{
  "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": 0.0453087,
      "auxiliary_loss_mlp": 0.03024399,
      "balance_loss_clip": 2.12519169,
      "balance_loss_mlp": 2.07797289,
      "epoch": 0.00012024289063909097,
      "flos": 24932483919360.0,
      "grad_norm": 40.05649503286671,
      "language_loss": 2.58637381,
      "learning_rate": 0.0,
      "loss": 1.90842855,
      "num_input_tokens_seen": 20375,
      "step": 1,
      "time_per_iteration": 13.402307987213135
    },
    {
      "auxiliary_loss_clip": 0.03016431,
      "auxiliary_loss_mlp": 0.02008305,
      "balance_loss_clip": 1.41552353,
      "balance_loss_mlp": 1.37354004,
      "epoch": 0.00024048578127818193,
      "flos": 30664624377600.0,
      "grad_norm": 55.35345245854129,
      "language_loss": 1.8912344,
      "learning_rate": 5.021476677069823e-07,
      "loss": 1.94148171,
      "num_input_tokens_seen": 39035,
      "step": 2,
      "time_per_iteration": 2.492805004119873
    },
    {
      "auxiliary_loss_clip": 0.03033623,
      "auxiliary_loss_mlp": 0.01978788,
      "balance_loss_clip": 1.41764009,
      "balance_loss_mlp": 1.35661113,
      "epoch": 0.0003607286719172729,
      "flos": 19026227969280.0,
      "grad_norm": 40.190956714339215,
      "language_loss": 1.61603725,
      "learning_rate": 7.958852231401551e-07,
      "loss": 1.6661613,
      "num_input_tokens_seen": 57600,
      "step": 3,
      "time_per_iteration": 2.388728618621826
    },
    {
      "auxiliary_loss_clip": 0.03012922,
      "auxiliary_loss_mlp": 0.02016422,
      "balance_loss_clip": 1.41320586,
      "balance_loss_mlp": 1.38737833,
      "epoch": 0.00048097156255636386,
      "flos": 19316314206720.0,
      "grad_norm": 37.177621431503354,
      "language_loss": 1.64672685,
      "learning_rate": 1.0042953354139647e-06,
      "loss": 1.69702029,
      "num_input_tokens_seen": 76465,
      "step": 4,
      "time_per_iteration": 2.452646255493164
    },
    {
      "auxiliary_loss_clip": 0.03018484,
      "auxiliary_loss_mlp": 0.01985602,
      "balance_loss_clip": 1.4120363,
      "balance_loss_mlp": 1.35465145,
      "epoch": 0.0006012144531954548,
      "flos": 13991264893440.0,
      "grad_norm": 59.586253523002846,
      "language_loss": 1.93763006,
      "learning_rate": 1.1659507774310057e-06,
      "loss": 1.9876709,
      "num_input_tokens_seen": 94350,
      "step": 5,
      "time_per_iteration": 2.6441478729248047
    },
    {
      "auxiliary_loss_clip": 0.03006309,
      "auxiliary_loss_mlp": 0.01974338,
      "balance_loss_clip": 1.41095996,
      "balance_loss_mlp": 1.35635757,
      "epoch": 0.0007214573438345458,
      "flos": 23148988225920.0,
      "grad_norm": 48.29982267042469,
      "language_loss": 1.61155796,
      "learning_rate": 1.2980328908471373e-06,
      "loss": 1.66136432,
      "num_input_tokens_seen": 114595,
      "step": 6,
      "time_per_iteration": 2.710550546646118
    },
    {
      "auxiliary_loss_clip": 0.02551452,
      "auxiliary_loss_mlp": 0.01942687,
      "balance_loss_clip": 1.43650091,
      "balance_loss_mlp": 1.34149063,
      "epoch": 0.0008417002344736367,
      "flos": 67663246170240.0,
      "grad_norm": 4.621009364545405,
      "language_loss": 0.81563127,
      "learning_rate": 1.4097067265369432e-06,
      "loss": 0.8605727,
      "num_input_tokens_seen": 179590,
      "step": 7,
      "time_per_iteration": 3.1298398971557617
    },
    {
      "auxiliary_loss_clip": 0.02981604,
      "auxiliary_loss_mlp": 0.02005351,
      "balance_loss_clip": 1.40704489,
      "balance_loss_mlp": 1.37172961,
      "epoch": 0.0009619431251127277,
      "flos": 21281381504640.0,
      "grad_norm": 41.27389399740074,
      "language_loss": 1.58512759,
      "learning_rate": 1.506443003120947e-06,
      "loss": 1.63499713,
      "num_input_tokens_seen": 195090,
      "step": 8,
      "time_per_iteration": 2.6470723152160645
    },
    {
      "auxiliary_loss_clip": 0.03027086,
      "auxiliary_loss_mlp": 0.02011793,
      "balance_loss_clip": 1.41574669,
      "balance_loss_mlp": 1.38961577,
      "epoch": 0.0010821860157518186,
      "flos": 23331342597120.0,
      "grad_norm": 17.883759919706183,
      "language_loss": 1.47798622,
      "learning_rate": 1.5917704462803102e-06,
      "loss": 1.52837491,
      "num_input_tokens_seen": 211635,
      "step": 9,
      "time_per_iteration": 2.6869075298309326
    },
    {
      "auxiliary_loss_clip": 0.03017421,
      "auxiliary_loss_mlp": 0.01984216,
      "balance_loss_clip": 1.41306543,
      "balance_loss_mlp": 1.37691617,
      "epoch": 0.0012024289063909096,
      "flos": 17010166337280.0,
      "grad_norm": 13.317398802474843,
      "language_loss": 1.52916026,
      "learning_rate": 1.6680984451379884e-06,
      "loss": 1.57917655,
      "num_input_tokens_seen": 224705,
      "step": 10,
      "time_per_iteration": 2.7115583419799805
    },
    {
      "auxiliary_loss_clip": 0.02984961,
      "auxiliary_loss_mlp": 0.02014909,
      "balance_loss_clip": 1.41232181,
      "balance_loss_mlp": 1.39769101,
      "epoch": 0.0013226717970300007,
      "flos": 21288133261440.0,
      "grad_norm": 13.708460645282123,
      "language_loss": 1.32536197,
      "learning_rate": 1.7371455188905097e-06,
      "loss": 1.37536073,
      "num_input_tokens_seen": 244635,
      "step": 11,
      "time_per_iteration": 2.696669816970825
    },
    {
      "auxiliary_loss_clip": 0.02999999,
      "auxiliary_loss_mlp": 0.02013158,
      "balance_loss_clip": 1.41574955,
      "balance_loss_mlp": 1.38182533,
      "epoch": 0.0014429146876690916,
      "flos": 27237884935680.0,
      "grad_norm": 10.932981067518668,
      "language_loss": 1.25328827,
      "learning_rate": 1.8001805585541196e-06,
      "loss": 1.30341983,
      "num_input_tokens_seen": 265765,
      "step": 12,
      "time_per_iteration": 2.7930660247802734
    },
    {
      "auxiliary_loss_clip": 0.0294344,
      "auxiliary_loss_mlp": 0.02013107,
      "balance_loss_clip": 1.4040041,
      "balance_loss_mlp": 1.3878777,
      "epoch": 0.0015631575783081825,
      "flos": 19062174504960.0,
      "grad_norm": 6.949076784902715,
      "language_loss": 1.29252386,
      "learning_rate": 1.8581671739548328e-06,
      "loss": 1.34208918,
      "num_input_tokens_seen": 283500,
      "step": 13,
      "time_per_iteration": 2.6577343940734863
    },
    {
      "auxiliary_loss_clip": 0.02959138,
      "auxiliary_loss_mlp": 0.01949059,
      "balance_loss_clip": 1.40717244,
      "balance_loss_mlp": 1.32154167,
      "epoch": 0.0016834004689472734,
      "flos": 48139473985920.0,
      "grad_norm": 6.260492747394391,
      "language_loss": 1.13400435,
      "learning_rate": 1.9118543942439254e-06,
      "loss": 1.18308628,
      "num_input_tokens_seen": 305685,
      "step": 14,
      "time_per_iteration": 4.275149345397949
    },
    {
      "auxiliary_loss_clip": 0.02930498,
      "auxiliary_loss_mlp": 0.01935818,
      "balance_loss_clip": 1.40850115,
      "balance_loss_mlp": 1.33462214,
      "epoch": 0.0018036433595863645,
      "flos": 34970026314240.0,
      "grad_norm": 6.161757028877586,
      "language_loss": 1.12679744,
      "learning_rate": 1.961836000571161e-06,
      "loss": 1.17546058,
      "num_input_tokens_seen": 327340,
      "step": 15,
      "time_per_iteration": 3.539961814880371
    },
    {
      "auxiliary_loss_clip": 0.0244068,
      "auxiliary_loss_mlp": 0.01857046,
      "balance_loss_clip": 1.42286229,
      "balance_loss_mlp": 1.2833159,
      "epoch": 0.0019238862502254555,
      "flos": 59768284440960.0,
      "grad_norm": 3.7845423747583786,
      "language_loss": 0.64651853,
      "learning_rate": 2.0085906708279293e-06,
      "loss": 0.6894958,
      "num_input_tokens_seen": 382710,
      "step": 16,
      "time_per_iteration": 3.1092963218688965
    },
    {
      "auxiliary_loss_clip": 0.02916534,
      "auxiliary_loss_mlp": 0.01903503,
      "balance_loss_clip": 1.40354609,
      "balance_loss_mlp": 1.31336939,
      "epoch": 0.0020441291408645466,
      "flos": 20814543417600.0,
      "grad_norm": 6.6980063015362665,
      "language_loss": 1.16166615,
      "learning_rate": 2.0525099325728135e-06,
      "loss": 1.20986652,
      "num_input_tokens_seen": 400890,
      "step": 17,
      "time_per_iteration": 2.672699213027954
    },
    {
      "auxiliary_loss_clip": 0.02407658,
      "auxiliary_loss_mlp": 0.01833555,
      "balance_loss_clip": 1.4206264,
      "balance_loss_mlp": 1.26592851,
      "epoch": 0.0021643720315036373,
      "flos": 63857001582720.0,
      "grad_norm": 3.531502406972119,
      "language_loss": 0.72226644,
      "learning_rate": 2.0939181139872922e-06,
      "loss": 0.7646786,
      "num_input_tokens_seen": 462605,
      "step": 18,
      "time_per_iteration": 3.1368837356567383
    },
    {
      "auxiliary_loss_clip": 0.02890265,
      "auxiliary_loss_mlp": 0.0192717,
      "balance_loss_clip": 1.39767098,
      "balance_loss_mlp": 1.31452942,
      "epoch": 0.0022846149221427284,
      "flos": 31284981192960.0,
      "grad_norm": 5.488954984064185,
      "language_loss": 1.01707244,
      "learning_rate": 2.1330868934640175e-06,
      "loss": 1.06524682,
      "num_input_tokens_seen": 483280,
      "step": 19,
      "time_per_iteration": 2.7848057746887207
    },
    {
      "auxiliary_loss_clip": 0.0236213,
      "auxiliary_loss_mlp": 0.01808363,
      "balance_loss_clip": 1.41383862,
      "balance_loss_mlp": 1.25294292,
      "epoch": 0.002404857812781819,
      "flos": 51083648161920.0,
      "grad_norm": 3.5808905885145133,
      "language_loss": 0.76495373,
      "learning_rate": 2.170246112844971e-06,
      "loss": 0.80665863,
      "num_input_tokens_seen": 537620,
      "step": 20,
      "time_per_iteration": 2.8943123817443848
    },
    {
      "auxiliary_loss_clip": 0.02858942,
      "auxiliary_loss_mlp": 0.01882676,
      "balance_loss_clip": 1.39366663,
      "balance_loss_mlp": 1.28758311,
      "epoch": 0.0025251007034209102,
      "flos": 15815347309440.0,
      "grad_norm": 4.108947108899217,
      "language_loss": 1.01803756,
      "learning_rate": 2.2055919496770983e-06,
      "loss": 1.06545377,
      "num_input_tokens_seen": 555760,
      "step": 21,
      "time_per_iteration": 2.625614881515503
    },
    {
      "auxiliary_loss_clip": 0.02818396,
      "auxiliary_loss_mlp": 0.01798778,
      "balance_loss_clip": 1.38654721,
      "balance_loss_mlp": 1.23382175,
      "epoch": 0.0026453435940600014,
      "flos": 37851857458560.0,
      "grad_norm": 3.7870140488020425,
      "language_loss": 0.89644194,
      "learning_rate": 2.2392931865974923e-06,
      "loss": 0.94261372,
      "num_input_tokens_seen": 578450,
      "step": 22,
      "time_per_iteration": 2.748885154724121
    },
    {
      "auxiliary_loss_clip": 0.02772658,
      "auxiliary_loss_mlp": 0.01815127,
      "balance_loss_clip": 1.38328421,
      "balance_loss_mlp": 1.24292254,
      "epoch": 0.002765586484699092,
      "flos": 21141976821120.0,
      "grad_norm": 4.152053156434109,
      "language_loss": 1.01760089,
      "learning_rate": 2.271496085962064e-06,
      "loss": 1.06347871,
      "num_input_tokens_seen": 596145,
      "step": 23,
      "time_per_iteration": 2.6569299697875977
    },
    {
      "auxiliary_loss_clip": 0.02769797,
      "auxiliary_loss_mlp": 0.018324,
      "balance_loss_clip": 1.38065851,
      "balance_loss_mlp": 1.25790691,
      "epoch": 0.002885829375338183,
      "flos": 20667381396480.0,
      "grad_norm": 3.6861529791062075,
      "language_loss": 1.02674317,
      "learning_rate": 2.3023282262611022e-06,
      "loss": 1.07276511,
      "num_input_tokens_seen": 614920,
      "step": 24,
      "time_per_iteration": 2.608624219894409
    },
    {
      "auxiliary_loss_clip": 0.0272516,
      "auxiliary_loss_mlp": 0.01761225,
      "balance_loss_clip": 1.37109566,
      "balance_loss_mlp": 1.19512403,
      "epoch": 0.003006072265977274,
      "flos": 34823869873920.0,
      "grad_norm": 3.7288752578216005,
      "language_loss": 0.92569017,
      "learning_rate": 2.3319015548620114e-06,
      "loss": 0.97055399,
      "num_input_tokens_seen": 636060,
      "step": 25,
      "time_per_iteration": 2.7860467433929443
    },
    {
      "auxiliary_loss_clip": 0.02729038,
      "auxiliary_loss_mlp": 0.01779431,
      "balance_loss_clip": 1.36872518,
      "balance_loss_mlp": 1.21867061,
      "epoch": 0.003126315156616365,
      "flos": 24422021118720.0,
      "grad_norm": 2.292046447616983,
      "language_loss": 0.92839521,
      "learning_rate": 2.3603148416618152e-06,
      "loss": 0.97347993,
      "num_input_tokens_seen": 655575,
      "step": 26,
      "time_per_iteration": 2.644622564315796
    },
    {
      "auxiliary_loss_clip": 0.02683952,
      "auxiliary_loss_mlp": 0.01766354,
      "balance_loss_clip": 1.36782992,
      "balance_loss_mlp": 1.21169722,
      "epoch": 0.003246558047255456,
      "flos": 23622326674560.0,
      "grad_norm": 2.979073986219296,
      "language_loss": 1.00929868,
      "learning_rate": 2.3876556694204647e-06,
      "loss": 1.05380177,
      "num_input_tokens_seen": 675730,
      "step": 27,
      "time_per_iteration": 2.6642537117004395
    },
    {
      "auxiliary_loss_clip": 0.0268306,
      "auxiliary_loss_mlp": 0.01733614,
      "balance_loss_clip": 1.36775827,
      "balance_loss_mlp": 1.18620479,
      "epoch": 0.003366800937894547,
      "flos": 17820275725440.0,
      "grad_norm": 3.0666357915817586,
      "language_loss": 0.90653968,
      "learning_rate": 2.414002061950908e-06,
      "loss": 0.95070642,
      "num_input_tokens_seen": 694605,
      "step": 28,
      "time_per_iteration": 2.59476900100708
    },
    {
      "auxiliary_loss_clip": 0.02680003,
      "auxiliary_loss_mlp": 0.01725822,
      "balance_loss_clip": 1.372316,
      "balance_loss_mlp": 1.18566072,
      "epoch": 0.003487043828533638,
      "flos": 24426115269120.0,
      "grad_norm": 2.2692897548172004,
      "language_loss": 0.9992311,
      "learning_rate": 2.4394238264681557e-06,
      "loss": 1.0432893,
      "num_input_tokens_seen": 714340,
      "step": 29,
      "time_per_iteration": 2.6500377655029297
    },
    {
      "auxiliary_loss_clip": 0.02640901,
      "auxiliary_loss_mlp": 0.01713735,
      "balance_loss_clip": 1.36137772,
      "balance_loss_mlp": 1.16708899,
      "epoch": 0.003607286719172729,
      "flos": 26140311002880.0,
      "grad_norm": 2.80938794610897,
      "language_loss": 0.99658728,
      "learning_rate": 2.4639836682781433e-06,
      "loss": 1.04013371,
      "num_input_tokens_seen": 734470,
      "step": 30,
      "time_per_iteration": 2.6594419479370117
    },
    {
      "auxiliary_loss_clip": 0.02589048,
      "auxiliary_loss_mlp": 0.01682643,
      "balance_loss_clip": 1.3586514,
      "balance_loss_mlp": 1.14934874,
      "epoch": 0.00372752960981182,
      "flos": 20593082113920.0,
      "grad_norm": 2.714006965676855,
      "language_loss": 1.00205088,
      "learning_rate": 2.487738122623307e-06,
      "loss": 1.04476786,
      "num_input_tokens_seen": 753380,
      "step": 31,
      "time_per_iteration": 2.6267364025115967
    },
    {
      "auxiliary_loss_clip": 0.02540853,
      "auxiliary_loss_mlp": 0.01660493,
      "balance_loss_clip": 1.34376931,
      "balance_loss_mlp": 1.13101327,
      "epoch": 0.003847772500450911,
      "flos": 22674608282880.0,
      "grad_norm": 2.7539894184855176,
      "language_loss": 0.99033946,
      "learning_rate": 2.510738338534912e-06,
      "loss": 1.03235292,
      "num_input_tokens_seen": 772105,
      "step": 32,
      "time_per_iteration": 2.6818654537200928
    },
    {
      "auxiliary_loss_clip": 0.02467029,
      "auxiliary_loss_mlp": 0.01653818,
      "balance_loss_clip": 1.32897449,
      "balance_loss_mlp": 1.12014186,
      "epoch": 0.003968015391090002,
      "flos": 17967796882560.0,
      "grad_norm": 4.69957988603279,
      "language_loss": 1.02554035,
      "learning_rate": 2.5330307420306648e-06,
      "loss": 1.06674874,
      "num_input_tokens_seen": 788955,
      "step": 33,
      "time_per_iteration": 2.5856425762176514
    },
    {
      "auxiliary_loss_clip": 0.02409353,
      "auxiliary_loss_mlp": 0.01641273,
      "balance_loss_clip": 1.32103622,
      "balance_loss_mlp": 1.12400031,
      "epoch": 0.004088258281729093,
      "flos": 27304103658240.0,
      "grad_norm": 2.3548962289021733,
      "language_loss": 0.88200492,
      "learning_rate": 2.554657600279796e-06,
      "loss": 0.92251116,
      "num_input_tokens_seen": 810230,
      "step": 34,
      "time_per_iteration": 2.6599254608154297
    },
    {
      "auxiliary_loss_clip": 0.02390725,
      "auxiliary_loss_mlp": 0.01612571,
      "balance_loss_clip": 1.32020497,
      "balance_loss_mlp": 1.10521579,
      "epoch": 0.004208501172368184,
      "flos": 23258587599360.0,
      "grad_norm": 2.3190751928295468,
      "language_loss": 1.0344131,
      "learning_rate": 2.5756575039679493e-06,
      "loss": 1.07444608,
      "num_input_tokens_seen": 829780,
      "step": 35,
      "time_per_iteration": 2.6185925006866455
    },
    {
      "auxiliary_loss_clip": 0.02352412,
      "auxiliary_loss_mlp": 0.0161133,
      "balance_loss_clip": 1.31636763,
      "balance_loss_mlp": 1.10817134,
      "epoch": 0.0043287440630072746,
      "flos": 17312104062720.0,
      "grad_norm": 1.8857284480016188,
      "language_loss": 0.9518497,
      "learning_rate": 2.5960657816942747e-06,
      "loss": 0.99148715,
      "num_input_tokens_seen": 848695,
      "step": 36,
      "time_per_iteration": 2.61562180519104
    },
    {
      "auxiliary_loss_clip": 0.01793365,
      "auxiliary_loss_mlp": 0.01704104,
      "balance_loss_clip": 1.28598237,
      "balance_loss_mlp": 1.27990949,
      "epoch": 0.004448986953646365,
      "flos": 53092491160320.0,
      "grad_norm": 1.3932761973085401,
      "language_loss": 0.60957325,
      "learning_rate": 2.6159148575788668e-06,
      "loss": 0.64454794,
      "num_input_tokens_seen": 906730,
      "step": 37,
      "time_per_iteration": 3.1273927688598633
    },
    {
      "auxiliary_loss_clip": 0.02277778,
      "auxiliary_loss_mlp": 0.01566146,
      "balance_loss_clip": 1.30672705,
      "balance_loss_mlp": 1.07786465,
      "epoch": 0.004569229844285457,
      "flos": 13444165866240.0,
      "grad_norm": 2.222725324332027,
      "language_loss": 0.98605675,
      "learning_rate": 2.635234561171e-06,
      "loss": 1.02449584,
      "num_input_tokens_seen": 925125,
      "step": 38,
      "time_per_iteration": 2.6355819702148438
    },
    {
      "auxiliary_loss_clip": 0.02258946,
      "auxiliary_loss_mlp": 0.0155264,
      "balance_loss_clip": 1.29661858,
      "balance_loss_mlp": 1.07160699,
      "epoch": 0.0046894727349245475,
      "flos": 16209609966720.0,
      "grad_norm": 2.2339572794325107,
      "language_loss": 0.94263422,
      "learning_rate": 2.6540523970949877e-06,
      "loss": 0.98075008,
      "num_input_tokens_seen": 939970,
      "step": 39,
      "time_per_iteration": 2.579984426498413
    },
    {
      "auxiliary_loss_clip": 0.02247426,
      "auxiliary_loss_mlp": 0.01539186,
      "balance_loss_clip": 1.30225897,
      "balance_loss_mlp": 1.06692672,
      "epoch": 0.004809715625563638,
      "flos": 23914244505600.0,
      "grad_norm": 2.827923899924266,
      "language_loss": 0.92589557,
      "learning_rate": 2.6723937805519533e-06,
      "loss": 0.96376169,
      "num_input_tokens_seen": 957470,
      "step": 40,
      "time_per_iteration": 3.4047935009002686
    },
    {
      "auxiliary_loss_clip": 0.02222023,
      "auxiliary_loss_mlp": 0.01528797,
      "balance_loss_clip": 1.29602635,
      "balance_loss_mlp": 1.07141471,
      "epoch": 0.00492995851620273,
      "flos": 20773030273920.0,
      "grad_norm": 2.4410972428213844,
      "language_loss": 0.93071747,
      "learning_rate": 2.690282243737839e-06,
      "loss": 0.96822572,
      "num_input_tokens_seen": 976405,
      "step": 41,
      "time_per_iteration": 4.178888559341431
    },
    {
      "auxiliary_loss_clip": 0.02198679,
      "auxiliary_loss_mlp": 0.0150265,
      "balance_loss_clip": 1.29209495,
      "balance_loss_mlp": 1.047557,
      "epoch": 0.0050502014068418205,
      "flos": 20338655103360.0,
      "grad_norm": 3.445176608857848,
      "language_loss": 0.9953258,
      "learning_rate": 2.7077396173840807e-06,
      "loss": 1.0323391,
      "num_input_tokens_seen": 994690,
      "step": 42,
      "time_per_iteration": 2.6067748069763184
    },
    {
      "auxiliary_loss_clip": 0.02180712,
      "auxiliary_loss_mlp": 0.01508579,
      "balance_loss_clip": 1.2938509,
      "balance_loss_mlp": 1.07332158,
      "epoch": 0.005170444297480911,
      "flos": 25994872834560.0,
      "grad_norm": 2.4747522085804543,
      "language_loss": 0.92700005,
      "learning_rate": 2.7247861909342594e-06,
      "loss": 0.96389294,
      "num_input_tokens_seen": 1015615,
      "step": 43,
      "time_per_iteration": 2.645594596862793
    },
    {
      "auxiliary_loss_clip": 0.02129997,
      "auxiliary_loss_mlp": 0.01489824,
      "balance_loss_clip": 1.27942896,
      "balance_loss_mlp": 1.046175,
      "epoch": 0.005290687188120003,
      "flos": 20954055841920.0,
      "grad_norm": 2.251205497913448,
      "language_loss": 0.8302021,
      "learning_rate": 2.7414408543044743e-06,
      "loss": 0.86640036,
      "num_input_tokens_seen": 1031255,
      "step": 44,
      "time_per_iteration": 2.5888729095458984
    },
    {
      "auxiliary_loss_clip": 0.0211171,
      "auxiliary_loss_mlp": 0.01499132,
      "balance_loss_clip": 1.27232671,
      "balance_loss_mlp": 1.06463802,
      "epoch": 0.005410930078759093,
      "flos": 15851401585920.0,
      "grad_norm": 4.378379538928598,
      "language_loss": 0.7915197,
      "learning_rate": 2.7577212237113157e-06,
      "loss": 0.82762814,
      "num_input_tokens_seen": 1048295,
      "step": 45,
      "time_per_iteration": 2.5761027336120605
    },
    {
      "auxiliary_loss_clip": 0.02086656,
      "auxiliary_loss_mlp": 0.01468456,
      "balance_loss_clip": 1.27612555,
      "balance_loss_mlp": 1.05227268,
      "epoch": 0.005531172969398184,
      "flos": 21104988791040.0,
      "grad_norm": 2.055607767293106,
      "language_loss": 1.0421474,
      "learning_rate": 2.7736437536690466e-06,
      "loss": 1.07769847,
      "num_input_tokens_seen": 1067925,
      "step": 46,
      "time_per_iteration": 2.6037964820861816
    },
    {
      "auxiliary_loss_clip": 0.02069318,
      "auxiliary_loss_mlp": 0.01463238,
      "balance_loss_clip": 1.26886511,
      "balance_loss_mlp": 1.05868936,
      "epoch": 0.005651415860037276,
      "flos": 20844887431680.0,
      "grad_norm": 1.9629075544797059,
      "language_loss": 1.07826447,
      "learning_rate": 2.789223836941131e-06,
      "loss": 1.11359,
      "num_input_tokens_seen": 1088060,
      "step": 47,
      "time_per_iteration": 2.6000306606292725
    },
    {
      "auxiliary_loss_clip": 0.02057098,
      "auxiliary_loss_mlp": 0.01470319,
      "balance_loss_clip": 1.26867902,
      "balance_loss_mlp": 1.05833173,
      "epoch": 0.005771658750676366,
      "flos": 13260195383040.0,
      "grad_norm": 2.401218333099771,
      "language_loss": 1.08662903,
      "learning_rate": 2.8044758939680847e-06,
      "loss": 1.12190318,
      "num_input_tokens_seen": 1104130,
      "step": 48,
      "time_per_iteration": 2.585463047027588
    },
    {
      "auxiliary_loss_clip": 0.02040563,
      "auxiliary_loss_mlp": 0.01459209,
      "balance_loss_clip": 1.26747572,
      "balance_loss_mlp": 1.06362462,
      "epoch": 0.005891901641315457,
      "flos": 24425396997120.0,
      "grad_norm": 2.7767191484068703,
      "language_loss": 1.02077508,
      "learning_rate": 2.8194134530738863e-06,
      "loss": 1.05577278,
      "num_input_tokens_seen": 1122900,
      "step": 49,
      "time_per_iteration": 2.638040542602539
    },
    {
      "auxiliary_loss_clip": 0.02003592,
      "auxiliary_loss_mlp": 0.01464111,
      "balance_loss_clip": 1.26080275,
      "balance_loss_mlp": 1.07653821,
      "epoch": 0.006012144531954548,
      "flos": 23076197314560.0,
      "grad_norm": 3.6882757101618946,
      "language_loss": 0.90302163,
      "learning_rate": 2.834049222568994e-06,
      "loss": 0.93769872,
      "num_input_tokens_seen": 1140250,
      "step": 50,
      "time_per_iteration": 2.6478142738342285
    },
    {
      "auxiliary_loss_clip": 0.0199236,
      "auxiliary_loss_mlp": 0.01419724,
      "balance_loss_clip": 1.25675392,
      "balance_loss_mlp": 1.04569304,
      "epoch": 0.006132387422593639,
      "flos": 22528775064960.0,
      "grad_norm": 2.0542826881899305,
      "language_loss": 0.92550802,
      "learning_rate": 2.848395155712969e-06,
      "loss": 0.95962894,
      "num_input_tokens_seen": 1160470,
      "step": 51,
      "time_per_iteration": 2.606874942779541
    },
    {
      "auxiliary_loss_clip": 0.0197401,
      "auxiliary_loss_mlp": 0.01414804,
      "balance_loss_clip": 1.25479364,
      "balance_loss_mlp": 1.05259895,
      "epoch": 0.00625263031323273,
      "flos": 27628340751360.0,
      "grad_norm": 2.3556716253285095,
      "language_loss": 0.97746921,
      "learning_rate": 2.8624625093687977e-06,
      "loss": 1.01135731,
      "num_input_tokens_seen": 1177605,
      "step": 52,
      "time_per_iteration": 2.656209707260132
    },
    {
      "auxiliary_loss_clip": 0.01969606,
      "auxiliary_loss_mlp": 0.01417453,
      "balance_loss_clip": 1.25473309,
      "balance_loss_mlp": 1.04552031,
      "epoch": 0.006372873203871821,
      "flos": 23110671392640.0,
      "grad_norm": 2.0736284955624487,
      "language_loss": 0.88969409,
      "learning_rate": 2.876261897070029e-06,
      "loss": 0.92356473,
      "num_input_tokens_seen": 1197735,
      "step": 53,
      "time_per_iteration": 2.6310503482818604
    },
    {
      "auxiliary_loss_clip": 0.01958248,
      "auxiliary_loss_mlp": 0.01398245,
      "balance_loss_clip": 1.24924254,
      "balance_loss_mlp": 1.04481316,
      "epoch": 0.006493116094510912,
      "flos": 22856028900480.0,
      "grad_norm": 2.464367229755811,
      "language_loss": 0.92479634,
      "learning_rate": 2.889803337127447e-06,
      "loss": 0.95836127,
      "num_input_tokens_seen": 1216335,
      "step": 54,
      "time_per_iteration": 2.6552133560180664
    },
    {
      "auxiliary_loss_clip": 0.01935928,
      "auxiliary_loss_mlp": 0.01421643,
      "balance_loss_clip": 1.2454046,
      "balance_loss_mlp": 1.06935537,
      "epoch": 0.006613358985150003,
      "flos": 23071708114560.0,
      "grad_norm": 2.629135537444029,
      "language_loss": 0.8473953,
      "learning_rate": 2.903096296321516e-06,
      "loss": 0.88097101,
      "num_input_tokens_seen": 1234480,
      "step": 55,
      "time_per_iteration": 2.631974220275879
    },
    {
      "auxiliary_loss_clip": 0.01929913,
      "auxiliary_loss_mlp": 0.01412091,
      "balance_loss_clip": 1.2450695,
      "balance_loss_mlp": 1.05999422,
      "epoch": 0.006733601875789094,
      "flos": 26537662229760.0,
      "grad_norm": 2.3184773434736736,
      "language_loss": 0.91684091,
      "learning_rate": 2.9161497296578907e-06,
      "loss": 0.950261,
      "num_input_tokens_seen": 1253870,
      "step": 56,
      "time_per_iteration": 2.6975319385528564
    },
    {
      "auxiliary_loss_clip": 0.01903625,
      "auxiliary_loss_mlp": 0.01412479,
      "balance_loss_clip": 1.24271703,
      "balance_loss_mlp": 1.06190872,
      "epoch": 0.006853844766428185,
      "flos": 15523178083200.0,
      "grad_norm": 2.265653791971172,
      "language_loss": 0.85676903,
      "learning_rate": 2.928972116604173e-06,
      "loss": 0.88993007,
      "num_input_tokens_seen": 1270145,
      "step": 57,
      "time_per_iteration": 2.6410205364227295
    },
    {
      "auxiliary_loss_clip": 0.01886049,
      "auxiliary_loss_mlp": 0.01372482,
      "balance_loss_clip": 1.23887622,
      "balance_loss_mlp": 1.04651618,
      "epoch": 0.006974087657067276,
      "flos": 24243760897920.0,
      "grad_norm": 2.764554548056646,
      "language_loss": 1.0186969,
      "learning_rate": 2.9415714941751377e-06,
      "loss": 1.05128217,
      "num_input_tokens_seen": 1291365,
      "step": 58,
      "time_per_iteration": 2.6696598529815674
    },
    {
      "auxiliary_loss_clip": 0.01882523,
      "auxiliary_loss_mlp": 0.01382038,
      "balance_loss_clip": 1.23821962,
      "balance_loss_mlp": 1.05435574,
      "epoch": 0.007094330547706367,
      "flos": 25772513690880.0,
      "grad_norm": 1.919562802469859,
      "language_loss": 0.9371078,
      "learning_rate": 2.9539554871897396e-06,
      "loss": 0.96975338,
      "num_input_tokens_seen": 1311535,
      "step": 59,
      "time_per_iteration": 2.676222085952759
    },
    {
      "auxiliary_loss_clip": 0.01842611,
      "auxiliary_loss_mlp": 0.01373359,
      "balance_loss_clip": 1.22992456,
      "balance_loss_mlp": 1.05578554,
      "epoch": 0.007214573438345458,
      "flos": 21319015979520.0,
      "grad_norm": 2.2086513936204226,
      "language_loss": 0.97362268,
      "learning_rate": 2.9661313359851253e-06,
      "loss": 1.00578237,
      "num_input_tokens_seen": 1329420,
      "step": 60,
      "time_per_iteration": 2.651477098464966
    },
    {
      "auxiliary_loss_clip": 0.01834743,
      "auxiliary_loss_mlp": 0.01371004,
      "balance_loss_clip": 1.23185146,
      "balance_loss_mlp": 1.05896175,
      "epoch": 0.007334816328984549,
      "flos": 24937088192640.0,
      "grad_norm": 2.087639075503606,
      "language_loss": 0.93683457,
      "learning_rate": 2.978105921839922e-06,
      "loss": 0.96889198,
      "num_input_tokens_seen": 1349965,
      "step": 61,
      "time_per_iteration": 2.666473150253296
    },
    {
      "auxiliary_loss_clip": 0.01832506,
      "auxiliary_loss_mlp": 0.01347752,
      "balance_loss_clip": 1.22967124,
      "balance_loss_mlp": 1.04505575,
      "epoch": 0.00745505921962364,
      "flos": 18510586277760.0,
      "grad_norm": 2.107774906143673,
      "language_loss": 0.72052026,
      "learning_rate": 2.9898857903302893e-06,
      "loss": 0.75232279,
      "num_input_tokens_seen": 1368915,
      "step": 62,
      "time_per_iteration": 2.613717794418335
    },
    {
      "auxiliary_loss_clip": 0.01815699,
      "auxiliary_loss_mlp": 0.01361521,
      "balance_loss_clip": 1.23170066,
      "balance_loss_mlp": 1.05920601,
      "epoch": 0.007575302110262731,
      "flos": 18477656484480.0,
      "grad_norm": 4.376929165172936,
      "language_loss": 0.87953103,
      "learning_rate": 3.001477172817253e-06,
      "loss": 0.91130316,
      "num_input_tokens_seen": 1386805,
      "step": 63,
      "time_per_iteration": 2.619391918182373
    },
    {
      "auxiliary_loss_clip": 0.01803686,
      "auxiliary_loss_mlp": 0.0132975,
      "balance_loss_clip": 1.22568798,
      "balance_loss_mlp": 1.04231215,
      "epoch": 0.007695545000901822,
      "flos": 24973178382720.0,
      "grad_norm": 2.5748419331048304,
      "language_loss": 0.96118534,
      "learning_rate": 3.012886006241894e-06,
      "loss": 0.99251968,
      "num_input_tokens_seen": 1406190,
      "step": 64,
      "time_per_iteration": 2.643075466156006
    },
    {
      "auxiliary_loss_clip": 0.01797596,
      "auxiliary_loss_mlp": 0.01328233,
      "balance_loss_clip": 1.22866285,
      "balance_loss_mlp": 1.04728007,
      "epoch": 0.007815787891540913,
      "flos": 21324223451520.0,
      "grad_norm": 2.1128966948706367,
      "language_loss": 0.88245243,
      "learning_rate": 3.0241179513858383e-06,
      "loss": 0.91371071,
      "num_input_tokens_seen": 1425500,
      "step": 65,
      "time_per_iteration": 2.6216814517974854
    },
    {
      "auxiliary_loss_clip": 0.01791598,
      "auxiliary_loss_mlp": 0.01348424,
      "balance_loss_clip": 1.22285855,
      "balance_loss_mlp": 1.06022406,
      "epoch": 0.007936030782180003,
      "flos": 21575777374080.0,
      "grad_norm": 2.079268453587738,
      "language_loss": 0.87726587,
      "learning_rate": 3.035178409737647e-06,
      "loss": 0.90866613,
      "num_input_tokens_seen": 1442950,
      "step": 66,
      "time_per_iteration": 3.395198106765747
    },
    {
      "auxiliary_loss_clip": 0.01757625,
      "auxiliary_loss_mlp": 0.01315927,
      "balance_loss_clip": 1.21579909,
      "balance_loss_mlp": 1.05080557,
      "epoch": 0.008056273672819095,
      "flos": 20120785159680.0,
      "grad_norm": 3.1096826206468533,
      "language_loss": 0.88654399,
      "learning_rate": 3.046072539090907e-06,
      "loss": 0.91727948,
      "num_input_tokens_seen": 1460915,
      "step": 67,
      "time_per_iteration": 4.073195695877075
    },
    {
      "auxiliary_loss_clip": 0.01741218,
      "auxiliary_loss_mlp": 0.01332232,
      "balance_loss_clip": 1.21339822,
      "balance_loss_mlp": 1.06329536,
      "epoch": 0.008176516563458186,
      "flos": 18333116156160.0,
      "grad_norm": 2.2239207494181223,
      "language_loss": 1.04725313,
      "learning_rate": 3.056805267986779e-06,
      "loss": 1.07798767,
      "num_input_tokens_seen": 1478385,
      "step": 68,
      "time_per_iteration": 2.586170196533203
    },
    {
      "auxiliary_loss_clip": 0.01729047,
      "auxiliary_loss_mlp": 0.01298497,
      "balance_loss_clip": 1.21359575,
      "balance_loss_mlp": 1.04520082,
      "epoch": 0.008296759454097276,
      "flos": 21872076664320.0,
      "grad_norm": 2.039047415497215,
      "language_loss": 0.9515959,
      "learning_rate": 3.0673813091022194e-06,
      "loss": 0.98187125,
      "num_input_tokens_seen": 1497605,
      "step": 69,
      "time_per_iteration": 2.598644733428955
    },
    {
      "auxiliary_loss_clip": 0.01532954,
      "auxiliary_loss_mlp": 0.01270153,
      "balance_loss_clip": 1.26069069,
      "balance_loss_mlp": 1.06721103,
      "epoch": 0.008417002344736368,
      "flos": 63408228036480.0,
      "grad_norm": 1.3037477442917957,
      "language_loss": 0.62122869,
      "learning_rate": 3.0778051716749317e-06,
      "loss": 0.64925969,
      "num_input_tokens_seen": 1561150,
      "step": 70,
      "time_per_iteration": 3.265970468521118
    },
    {
      "auxiliary_loss_clip": 0.01700297,
      "auxiliary_loss_mlp": 0.01310869,
      "balance_loss_clip": 1.2007885,
      "balance_loss_mlp": 1.05928981,
      "epoch": 0.008537245235375458,
      "flos": 22966454286720.0,
      "grad_norm": 2.258663281492024,
      "language_loss": 0.9026612,
      "learning_rate": 3.0880811730470094e-06,
      "loss": 0.93277276,
      "num_input_tokens_seen": 1580605,
      "step": 71,
      "time_per_iteration": 2.613346815109253
    },
    {
      "auxiliary_loss_clip": 0.01516256,
      "auxiliary_loss_mlp": 0.01226503,
      "balance_loss_clip": 1.25357187,
      "balance_loss_mlp": 1.03576803,
      "epoch": 0.008657488126014549,
      "flos": 61984046712960.0,
      "grad_norm": 1.147801278374983,
      "language_loss": 0.58584422,
      "learning_rate": 3.098213449401257e-06,
      "loss": 0.61327183,
      "num_input_tokens_seen": 1647535,
      "step": 72,
      "time_per_iteration": 3.1337921619415283
    },
    {
      "auxiliary_loss_clip": 0.01692048,
      "auxiliary_loss_mlp": 0.0127031,
      "balance_loss_clip": 1.20379794,
      "balance_loss_mlp": 1.03875804,
      "epoch": 0.00877773101665364,
      "flos": 30296791152000.0,
      "grad_norm": 2.45303044047699,
      "language_loss": 0.98655325,
      "learning_rate": 3.1082059657570015e-06,
      "loss": 1.01617682,
      "num_input_tokens_seen": 1666770,
      "step": 73,
      "time_per_iteration": 2.69681715965271
    },
    {
      "auxiliary_loss_clip": 0.01674206,
      "auxiliary_loss_mlp": 0.01274133,
      "balance_loss_clip": 1.1974709,
      "balance_loss_mlp": 1.04563236,
      "epoch": 0.00889797390729273,
      "flos": 23514056104320.0,
      "grad_norm": 3.019281159872949,
      "language_loss": 0.96413434,
      "learning_rate": 3.1180625252858496e-06,
      "loss": 0.99361771,
      "num_input_tokens_seen": 1685200,
      "step": 74,
      "time_per_iteration": 2.637930393218994
    },
    {
      "auxiliary_loss_clip": 0.01658775,
      "auxiliary_loss_mlp": 0.01280239,
      "balance_loss_clip": 1.19496667,
      "balance_loss_mlp": 1.04811454,
      "epoch": 0.009018216797931822,
      "flos": 23075838178560.0,
      "grad_norm": 3.573720063022437,
      "language_loss": 0.80203533,
      "learning_rate": 3.1277867780021663e-06,
      "loss": 0.83142543,
      "num_input_tokens_seen": 1701835,
      "step": 75,
      "time_per_iteration": 2.619157075881958
    },
    {
      "auxiliary_loss_clip": 0.01646941,
      "auxiliary_loss_mlp": 0.01257828,
      "balance_loss_clip": 1.19131613,
      "balance_loss_mlp": 1.04420531,
      "epoch": 0.009138459688570914,
      "flos": 15918877284480.0,
      "grad_norm": 1.9776084973692787,
      "language_loss": 0.95502162,
      "learning_rate": 3.1373822288779824e-06,
      "loss": 0.98406923,
      "num_input_tokens_seen": 1718415,
      "step": 76,
      "time_per_iteration": 2.5785481929779053
    },
    {
      "auxiliary_loss_clip": 0.01646782,
      "auxiliary_loss_mlp": 0.01275688,
      "balance_loss_clip": 1.19202542,
      "balance_loss_mlp": 1.05767775,
      "epoch": 0.009258702579210003,
      "flos": 27016531372800.0,
      "grad_norm": 2.255416496328232,
      "language_loss": 0.79577249,
      "learning_rate": 3.1468522454274533e-06,
      "loss": 0.82499719,
      "num_input_tokens_seen": 1738770,
      "step": 77,
      "time_per_iteration": 2.7057700157165527
    },
    {
      "auxiliary_loss_clip": 0.01647718,
      "auxiliary_loss_mlp": 0.0126299,
      "balance_loss_clip": 1.19213307,
      "balance_loss_mlp": 1.05337214,
      "epoch": 0.009378945469849095,
      "flos": 26903196984960.0,
      "grad_norm": 2.0410713519052996,
      "language_loss": 0.91864669,
      "learning_rate": 3.15620006480197e-06,
      "loss": 0.94775379,
      "num_input_tokens_seen": 1758040,
      "step": 78,
      "time_per_iteration": 2.6397178173065186
    },
    {
      "auxiliary_loss_clip": 0.01636227,
      "auxiliary_loss_mlp": 0.01261692,
      "balance_loss_clip": 1.19049156,
      "balance_loss_mlp": 1.04921365,
      "epoch": 0.009499188360488187,
      "flos": 35694236327040.0,
      "grad_norm": 3.021894862146212,
      "language_loss": 0.74774021,
      "learning_rate": 3.1654288004333087e-06,
      "loss": 0.77671939,
      "num_input_tokens_seen": 1776705,
      "step": 79,
      "time_per_iteration": 2.6920456886291504
    },
    {
      "auxiliary_loss_clip": 0.01620125,
      "auxiliary_loss_mlp": 0.01246211,
      "balance_loss_clip": 1.18872166,
      "balance_loss_mlp": 1.05108976,
      "epoch": 0.009619431251127276,
      "flos": 21503201944320.0,
      "grad_norm": 2.515636025129678,
      "language_loss": 0.75920856,
      "learning_rate": 3.1745414482589353e-06,
      "loss": 0.78787196,
      "num_input_tokens_seen": 1795915,
      "step": 80,
      "time_per_iteration": 2.599285125732422
    },
    {
      "auxiliary_loss_clip": 0.01627985,
      "auxiliary_loss_mlp": 0.01236801,
      "balance_loss_clip": 1.19026232,
      "balance_loss_mlp": 1.04425442,
      "epoch": 0.009739674141766368,
      "flos": 17421056991360.0,
      "grad_norm": 4.000001525298694,
      "language_loss": 0.86949003,
      "learning_rate": 3.1835408925606204e-06,
      "loss": 0.89813787,
      "num_input_tokens_seen": 1814055,
      "step": 81,
      "time_per_iteration": 2.568049669265747
    },
    {
      "auxiliary_loss_clip": 0.01609387,
      "auxiliary_loss_mlp": 0.01233467,
      "balance_loss_clip": 1.18539381,
      "balance_loss_mlp": 1.04626036,
      "epoch": 0.00985991703240546,
      "flos": 27527109246720.0,
      "grad_norm": 2.2944116960339445,
      "language_loss": 0.89241147,
      "learning_rate": 3.1924299114448214e-06,
      "loss": 0.92084002,
      "num_input_tokens_seen": 1834535,
      "step": 82,
      "time_per_iteration": 2.6494734287261963
    },
    {
      "auxiliary_loss_clip": 0.0161636,
      "auxiliary_loss_mlp": 0.01230356,
      "balance_loss_clip": 1.18673503,
      "balance_loss_mlp": 1.04810929,
      "epoch": 0.00998015992304455,
      "flos": 13808084509440.0,
      "grad_norm": 2.350110162364427,
      "language_loss": 0.83233261,
      "learning_rate": 3.2012111819909055e-06,
      "loss": 0.86079979,
      "num_input_tokens_seen": 1851865,
      "step": 83,
      "time_per_iteration": 2.584118127822876
    },
    {
      "auxiliary_loss_clip": 0.01595411,
      "auxiliary_loss_mlp": 0.01223468,
      "balance_loss_clip": 1.18258286,
      "balance_loss_mlp": 1.04827833,
      "epoch": 0.010100402813683641,
      "flos": 20191385341440.0,
      "grad_norm": 8.975001858219356,
      "language_loss": 0.9501487,
      "learning_rate": 3.2098872850910627e-06,
      "loss": 0.97833753,
      "num_input_tokens_seen": 1868540,
      "step": 84,
      "time_per_iteration": 2.571390151977539
    },
    {
      "auxiliary_loss_clip": 0.01594961,
      "auxiliary_loss_mlp": 0.01225788,
      "balance_loss_clip": 1.18429995,
      "balance_loss_mlp": 1.05164695,
      "epoch": 0.010220645704322733,
      "flos": 17201642762880.0,
      "grad_norm": 2.1247216432110276,
      "language_loss": 0.89264452,
      "learning_rate": 3.2184607100038194e-06,
      "loss": 0.92085207,
      "num_input_tokens_seen": 1887180,
      "step": 85,
      "time_per_iteration": 2.6075520515441895
    },
    {
      "auxiliary_loss_clip": 0.01595658,
      "auxiliary_loss_mlp": 0.01214251,
      "balance_loss_clip": 1.18251324,
      "balance_loss_mlp": 1.03934729,
      "epoch": 0.010340888594961822,
      "flos": 21470415805440.0,
      "grad_norm": 2.1750126615985597,
      "language_loss": 0.93202782,
      "learning_rate": 3.2269338586412414e-06,
      "loss": 0.96012694,
      "num_input_tokens_seen": 1904765,
      "step": 86,
      "time_per_iteration": 2.61441969871521
    },
    {
      "auxiliary_loss_clip": 0.01577205,
      "auxiliary_loss_mlp": 0.01208234,
      "balance_loss_clip": 1.17754412,
      "balance_loss_mlp": 1.04486942,
      "epoch": 0.010461131485600914,
      "flos": 23002831785600.0,
      "grad_norm": 2.690522602561099,
      "language_loss": 0.96386796,
      "learning_rate": 3.2353090496083106e-06,
      "loss": 0.99172235,
      "num_input_tokens_seen": 1922600,
      "step": 87,
      "time_per_iteration": 2.63801646232605
    },
    {
      "auxiliary_loss_clip": 0.0156599,
      "auxiliary_loss_mlp": 0.01208705,
      "balance_loss_clip": 1.17437685,
      "balance_loss_mlp": 1.04715228,
      "epoch": 0.010581374376240005,
      "flos": 33546850571520.0,
      "grad_norm": 1.947382074391797,
      "language_loss": 0.8126632,
      "learning_rate": 3.2435885220114572e-06,
      "loss": 0.84041017,
      "num_input_tokens_seen": 1943950,
      "step": 88,
      "time_per_iteration": 2.6933977603912354
    },
    {
      "auxiliary_loss_clip": 0.01566454,
      "auxiliary_loss_mlp": 0.01201987,
      "balance_loss_clip": 1.17799294,
      "balance_loss_mlp": 1.04653835,
      "epoch": 0.010701617266879095,
      "flos": 21763087822080.0,
      "grad_norm": 3.4061866615408363,
      "language_loss": 0.93906343,
      "learning_rate": 3.2517744390519113e-06,
      "loss": 0.96674782,
      "num_input_tokens_seen": 1962815,
      "step": 89,
      "time_per_iteration": 2.603818416595459
    },
    {
      "auxiliary_loss_clip": 0.01557285,
      "auxiliary_loss_mlp": 0.01202036,
      "balance_loss_clip": 1.17054963,
      "balance_loss_mlp": 1.04382157,
      "epoch": 0.010821860157518187,
      "flos": 19060199256960.0,
      "grad_norm": 2.211320541466291,
      "language_loss": 0.75099432,
      "learning_rate": 3.259868891418298e-06,
      "loss": 0.77858758,
      "num_input_tokens_seen": 1980580,
      "step": 90,
      "time_per_iteration": 2.579664707183838
    },
    {
      "auxiliary_loss_clip": 0.01556705,
      "auxiliary_loss_mlp": 0.01212194,
      "balance_loss_clip": 1.17595363,
      "balance_loss_mlp": 1.05302536,
      "epoch": 0.010942103048157278,
      "flos": 25447378757760.0,
      "grad_norm": 2.072695404672927,
      "language_loss": 0.85075581,
      "learning_rate": 3.2678739004917757e-06,
      "loss": 0.87844473,
      "num_input_tokens_seen": 2000315,
      "step": 91,
      "time_per_iteration": 2.6627094745635986
    },
    {
      "auxiliary_loss_clip": 0.01550033,
      "auxiliary_loss_mlp": 0.01200792,
      "balance_loss_clip": 1.17341506,
      "balance_loss_mlp": 1.05125582,
      "epoch": 0.011062345938796368,
      "flos": 27493928058240.0,
      "grad_norm": 1.6212437058878755,
      "language_loss": 0.92045468,
      "learning_rate": 3.275791421376029e-06,
      "loss": 0.94796288,
      "num_input_tokens_seen": 2023760,
      "step": 92,
      "time_per_iteration": 3.4186458587646484
    },
    {
      "auxiliary_loss_clip": 0.01537474,
      "auxiliary_loss_mlp": 0.01188643,
      "balance_loss_clip": 1.16777909,
      "balance_loss_mlp": 1.04292107,
      "epoch": 0.01118258882943546,
      "flos": 16071210864000.0,
      "grad_norm": 2.009111919120048,
      "language_loss": 0.96106005,
      "learning_rate": 3.2836233457634622e-06,
      "loss": 0.98832124,
      "num_input_tokens_seen": 2041895,
      "step": 93,
      "time_per_iteration": 4.793899059295654
    },
    {
      "auxiliary_loss_clip": 0.01539835,
      "auxiliary_loss_mlp": 0.01183022,
      "balance_loss_clip": 1.1647526,
      "balance_loss_mlp": 1.04073405,
      "epoch": 0.011302831720074551,
      "flos": 20668602458880.0,
      "grad_norm": 3.3354969305425657,
      "language_loss": 0.85480964,
      "learning_rate": 3.2913715046481135e-06,
      "loss": 0.88203818,
      "num_input_tokens_seen": 2061640,
      "step": 94,
      "time_per_iteration": 2.5935192108154297
    },
    {
      "auxiliary_loss_clip": 0.01531191,
      "auxiliary_loss_mlp": 0.01182634,
      "balance_loss_clip": 1.16179776,
      "balance_loss_mlp": 1.04520988,
      "epoch": 0.011423074610713641,
      "flos": 13072238490240.0,
      "grad_norm": 2.3469126746941145,
      "language_loss": 0.88728631,
      "learning_rate": 3.299037670895023e-06,
      "loss": 0.91442454,
      "num_input_tokens_seen": 2078255,
      "step": 95,
      "time_per_iteration": 2.617413282394409
    },
    {
      "auxiliary_loss_clip": 0.01528625,
      "auxiliary_loss_mlp": 0.0117617,
      "balance_loss_clip": 1.16492355,
      "balance_loss_mlp": 1.0383637,
      "epoch": 0.011543317501352733,
      "flos": 30335646689280.0,
      "grad_norm": 2.3505429369002186,
      "language_loss": 0.80201304,
      "learning_rate": 3.3066235616750667e-06,
      "loss": 0.82906097,
      "num_input_tokens_seen": 2099490,
      "step": 96,
      "time_per_iteration": 2.6649017333984375
    },
    {
      "auxiliary_loss_clip": 0.01520993,
      "auxiliary_loss_mlp": 0.01180699,
      "balance_loss_clip": 1.15958881,
      "balance_loss_mlp": 1.05195248,
      "epoch": 0.011663560391991824,
      "flos": 15522962601600.0,
      "grad_norm": 2.1705821646909,
      "language_loss": 0.92170382,
      "learning_rate": 3.3141308407736276e-06,
      "loss": 0.94872069,
      "num_input_tokens_seen": 2116125,
      "step": 97,
      "time_per_iteration": 2.5824432373046875
    },
    {
      "auxiliary_loss_clip": 0.01518404,
      "auxiliary_loss_mlp": 0.01174778,
      "balance_loss_clip": 1.15843105,
      "balance_loss_mlp": 1.04622281,
      "epoch": 0.011783803282630914,
      "flos": 19902125116800.0,
      "grad_norm": 2.0482764235159547,
      "language_loss": 0.86570263,
      "learning_rate": 3.321561120780869e-06,
      "loss": 0.89263445,
      "num_input_tokens_seen": 2134835,
      "step": 98,
      "time_per_iteration": 2.581505537033081
    },
    {
      "auxiliary_loss_clip": 0.01512233,
      "auxiliary_loss_mlp": 0.0117075,
      "balance_loss_clip": 1.15965736,
      "balance_loss_mlp": 1.04915643,
      "epoch": 0.011904046173270006,
      "flos": 22340674517760.0,
      "grad_norm": 2.0297250163415006,
      "language_loss": 1.01557922,
      "learning_rate": 3.3289159651708192e-06,
      "loss": 1.04240894,
      "num_input_tokens_seen": 2152410,
      "step": 99,
      "time_per_iteration": 2.5981733798980713
    },
    {
      "auxiliary_loss_clip": 0.01502514,
      "auxiliary_loss_mlp": 0.0116499,
      "balance_loss_clip": 1.15611434,
      "balance_loss_mlp": 1.04520845,
      "epoch": 0.012024289063909096,
      "flos": 19100060375040.0,
      "grad_norm": 1.956163364192819,
      "language_loss": 0.97505409,
      "learning_rate": 3.3361968902759768e-06,
      "loss": 1.00172913,
      "num_input_tokens_seen": 2172090,
      "step": 100,
      "time_per_iteration": 2.5510027408599854
    },
    {
      "auxiliary_loss_clip": 0.01497784,
      "auxiliary_loss_mlp": 0.01159129,
      "balance_loss_clip": 1.15469229,
      "balance_loss_mlp": 1.05060124,
      "epoch": 0.012144531954548187,
      "flos": 15012205159680.0,
      "grad_norm": 2.1150027141963763,
      "language_loss": 0.9384501,
      "learning_rate": 3.343405367163663e-06,
      "loss": 0.96501923,
      "num_input_tokens_seen": 2189020,
      "step": 101,
      "time_per_iteration": 2.546816349029541
    },
    {
      "auxiliary_loss_clip": 0.0150648,
      "auxiliary_loss_mlp": 0.01153391,
      "balance_loss_clip": 1.16041195,
      "balance_loss_mlp": 1.04543519,
      "epoch": 0.012264774845187279,
      "flos": 15122020014720.0,
      "grad_norm": 2.3804031806115633,
      "language_loss": 0.81260633,
      "learning_rate": 3.350542823419951e-06,
      "loss": 0.83920509,
      "num_input_tokens_seen": 2205620,
      "step": 102,
      "time_per_iteration": 2.526825189590454
    },
    {
      "auxiliary_loss_clip": 0.01499931,
      "auxiliary_loss_mlp": 0.01157401,
      "balance_loss_clip": 1.15521574,
      "balance_loss_mlp": 1.04219735,
      "epoch": 0.012385017735826368,
      "flos": 13949248959360.0,
      "grad_norm": 3.5861911336007735,
      "language_loss": 0.87721407,
      "learning_rate": 3.3576106448465615e-06,
      "loss": 0.90378737,
      "num_input_tokens_seen": 2219000,
      "step": 103,
      "time_per_iteration": 2.5419628620147705
    },
    {
      "auxiliary_loss_clip": 0.01492201,
      "auxiliary_loss_mlp": 0.01152962,
      "balance_loss_clip": 1.15092552,
      "balance_loss_mlp": 1.04457712,
      "epoch": 0.01250526062646546,
      "flos": 23623260428160.0,
      "grad_norm": 3.699616498350589,
      "language_loss": 0.88345087,
      "learning_rate": 3.3646101770757797e-06,
      "loss": 0.90990245,
      "num_input_tokens_seen": 2237790,
      "step": 104,
      "time_per_iteration": 2.5710675716400146
    },
    {
      "auxiliary_loss_clip": 0.01485939,
      "auxiliary_loss_mlp": 0.01163598,
      "balance_loss_clip": 1.15223074,
      "balance_loss_mlp": 1.05363917,
      "epoch": 0.012625503517104552,
      "flos": 34640078958720.0,
      "grad_norm": 1.8306105402457191,
      "language_loss": 0.85498941,
      "learning_rate": 3.371542727108104e-06,
      "loss": 0.88148481,
      "num_input_tokens_seen": 2259965,
      "step": 105,
      "time_per_iteration": 2.69905948638916
    },
    {
      "auxiliary_loss_clip": 0.01485643,
      "auxiliary_loss_mlp": 0.01165434,
      "balance_loss_clip": 1.15037036,
      "balance_loss_mlp": 1.05857432,
      "epoch": 0.012745746407743641,
      "flos": 17821891837440.0,
      "grad_norm": 2.258412255062607,
      "language_loss": 0.89881307,
      "learning_rate": 3.3784095647770114e-06,
      "loss": 0.92532384,
      "num_input_tokens_seen": 2278610,
      "step": 106,
      "time_per_iteration": 2.5383150577545166
    },
    {
      "auxiliary_loss_clip": 0.01478552,
      "auxiliary_loss_mlp": 0.01154144,
      "balance_loss_clip": 1.14627028,
      "balance_loss_mlp": 1.04504395,
      "epoch": 0.012865989298382733,
      "flos": 20595057361920.0,
      "grad_norm": 2.2166129175887717,
      "language_loss": 0.88347495,
      "learning_rate": 3.3852119241449547e-06,
      "loss": 0.90980196,
      "num_input_tokens_seen": 2297730,
      "step": 107,
      "time_per_iteration": 2.6038389205932617
    },
    {
      "auxiliary_loss_clip": 0.01472784,
      "auxiliary_loss_mlp": 0.01139727,
      "balance_loss_clip": 1.14579749,
      "balance_loss_mlp": 1.04393005,
      "epoch": 0.012986232189021825,
      "flos": 23948969978880.0,
      "grad_norm": 2.39349538738734,
      "language_loss": 0.96223915,
      "learning_rate": 3.3919510048344295e-06,
      "loss": 0.98836434,
      "num_input_tokens_seen": 2315740,
      "step": 108,
      "time_per_iteration": 2.5915186405181885
    },
    {
      "auxiliary_loss_clip": 0.01473683,
      "auxiliary_loss_mlp": 0.01132227,
      "balance_loss_clip": 1.14801407,
      "balance_loss_mlp": 1.03681219,
      "epoch": 0.013106475079660914,
      "flos": 23725425686400.0,
      "grad_norm": 2.1325339280838946,
      "language_loss": 0.86618012,
      "learning_rate": 3.3986279732976907e-06,
      "loss": 0.89223921,
      "num_input_tokens_seen": 2334215,
      "step": 109,
      "time_per_iteration": 2.5888819694519043
    },
    {
      "auxiliary_loss_clip": 0.0146352,
      "auxiliary_loss_mlp": 0.0114036,
      "balance_loss_clip": 1.14381897,
      "balance_loss_mlp": 1.04995215,
      "epoch": 0.013226717970300006,
      "flos": 21102438925440.0,
      "grad_norm": 2.1408373036610957,
      "language_loss": 0.95308113,
      "learning_rate": 3.4052439640284983e-06,
      "loss": 0.97911996,
      "num_input_tokens_seen": 2353130,
      "step": 110,
      "time_per_iteration": 2.5511116981506348
    },
    {
      "auxiliary_loss_clip": 0.01467681,
      "auxiliary_loss_mlp": 0.01138962,
      "balance_loss_clip": 1.14625907,
      "balance_loss_mlp": 1.04740918,
      "epoch": 0.013346960860939098,
      "flos": 24863902231680.0,
      "grad_norm": 1.7257173750439767,
      "language_loss": 0.81011415,
      "learning_rate": 3.4118000807190217e-06,
      "loss": 0.83618057,
      "num_input_tokens_seen": 2374010,
      "step": 111,
      "time_per_iteration": 2.6079864501953125
    },
    {
      "auxiliary_loss_clip": 0.0147179,
      "auxiliary_loss_mlp": 0.01131018,
      "balance_loss_clip": 1.14578366,
      "balance_loss_mlp": 1.03727186,
      "epoch": 0.013467203751578187,
      "flos": 28181940140160.0,
      "grad_norm": 2.0161350817544466,
      "language_loss": 0.75918114,
      "learning_rate": 3.4182973973648723e-06,
      "loss": 0.78520924,
      "num_input_tokens_seen": 2395220,
      "step": 112,
      "time_per_iteration": 2.6113860607147217
    },
    {
      "auxiliary_loss_clip": 0.01457903,
      "auxiliary_loss_mlp": 0.01134087,
      "balance_loss_clip": 1.14017749,
      "balance_loss_mlp": 1.04591942,
      "epoch": 0.013587446642217279,
      "flos": 18916233546240.0,
      "grad_norm": 2.6246451773435306,
      "language_loss": 0.94689393,
      "learning_rate": 3.424736959321014e-06,
      "loss": 0.9728139,
      "num_input_tokens_seen": 2413025,
      "step": 113,
      "time_per_iteration": 2.5599708557128906
    },
    {
      "auxiliary_loss_clip": 0.01457444,
      "auxiliary_loss_mlp": 0.01128761,
      "balance_loss_clip": 1.14257932,
      "balance_loss_mlp": 1.04641151,
      "epoch": 0.01370768953285637,
      "flos": 23988615615360.0,
      "grad_norm": 1.8935774974448414,
      "language_loss": 0.88701338,
      "learning_rate": 3.431119784311155e-06,
      "loss": 0.91287541,
      "num_input_tokens_seen": 2432700,
      "step": 114,
      "time_per_iteration": 2.6713778972625732
    },
    {
      "auxiliary_loss_clip": 0.0145785,
      "auxiliary_loss_mlp": 0.01123061,
      "balance_loss_clip": 1.14176226,
      "balance_loss_mlp": 1.04690981,
      "epoch": 0.01382793242349546,
      "flos": 39202565512320.0,
      "grad_norm": 1.7680281655727876,
      "language_loss": 0.77450758,
      "learning_rate": 3.43744686339307e-06,
      "loss": 0.80031675,
      "num_input_tokens_seen": 2455020,
      "step": 115,
      "time_per_iteration": 2.7368478775024414
    },
    {
      "auxiliary_loss_clip": 0.01453303,
      "auxiliary_loss_mlp": 0.0112249,
      "balance_loss_clip": 1.13550115,
      "balance_loss_mlp": 1.04219031,
      "epoch": 0.013948175314134552,
      "flos": 41353506714240.0,
      "grad_norm": 2.0598336504458525,
      "language_loss": 0.90858972,
      "learning_rate": 3.44371916188212e-06,
      "loss": 0.93434763,
      "num_input_tokens_seen": 2475775,
      "step": 116,
      "time_per_iteration": 2.7373931407928467
    },
    {
      "auxiliary_loss_clip": 0.01447865,
      "auxiliary_loss_mlp": 0.0111746,
      "balance_loss_clip": 1.13291061,
      "balance_loss_mlp": 1.03835285,
      "epoch": 0.014068418204773643,
      "flos": 22453542028800.0,
      "grad_norm": 1.9963322096700664,
      "language_loss": 0.8617624,
      "learning_rate": 3.449937620235143e-06,
      "loss": 0.88741571,
      "num_input_tokens_seen": 2496370,
      "step": 117,
      "time_per_iteration": 3.320584297180176
    },
    {
      "auxiliary_loss_clip": 0.01446677,
      "auxiliary_loss_mlp": 0.01118599,
      "balance_loss_clip": 1.13664126,
      "balance_loss_mlp": 1.04006386,
      "epoch": 0.014188661095412733,
      "flos": 23805147922560.0,
      "grad_norm": 1.615364663719047,
      "language_loss": 0.8905986,
      "learning_rate": 3.456103154896722e-06,
      "loss": 0.91625136,
      "num_input_tokens_seen": 2517645,
      "step": 118,
      "time_per_iteration": 3.356013536453247
    },
    {
      "auxiliary_loss_clip": 0.01438376,
      "auxiliary_loss_mlp": 0.01116657,
      "balance_loss_clip": 1.13284338,
      "balance_loss_mlp": 1.03874183,
      "epoch": 0.014308903986051825,
      "flos": 23660248458240.0,
      "grad_norm": 2.0119213078617357,
      "language_loss": 0.92168069,
      "learning_rate": 3.462216659109757e-06,
      "loss": 0.94723105,
      "num_input_tokens_seen": 2537825,
      "step": 119,
      "time_per_iteration": 3.3525164127349854
    },
    {
      "auxiliary_loss_clip": 0.01451287,
      "auxiliary_loss_mlp": 0.01118704,
      "balance_loss_clip": 1.14068723,
      "balance_loss_mlp": 1.04469943,
      "epoch": 0.014429146876690916,
      "flos": 20667991927680.0,
      "grad_norm": 2.4015067077815315,
      "language_loss": 0.85239851,
      "learning_rate": 3.4682790036921077e-06,
      "loss": 0.87809849,
      "num_input_tokens_seen": 2556485,
      "step": 120,
      "time_per_iteration": 3.3394882678985596
    },
    {
      "auxiliary_loss_clip": 0.01437363,
      "auxiliary_loss_mlp": 0.01101006,
      "balance_loss_clip": 1.13607359,
      "balance_loss_mlp": 1.03515482,
      "epoch": 0.014549389767330006,
      "flos": 20229199384320.0,
      "grad_norm": 2.2282694052089433,
      "language_loss": 0.83147061,
      "learning_rate": 3.4742910377810193e-06,
      "loss": 0.85685426,
      "num_input_tokens_seen": 2573945,
      "step": 121,
      "time_per_iteration": 2.556511402130127
    },
    {
      "auxiliary_loss_clip": 0.0143662,
      "auxiliary_loss_mlp": 0.01104604,
      "balance_loss_clip": 1.13140368,
      "balance_loss_mlp": 1.03679752,
      "epoch": 0.014669632657969098,
      "flos": 18004174381440.0,
      "grad_norm": 2.043247982997348,
      "language_loss": 0.88811433,
      "learning_rate": 3.4802535895469042e-06,
      "loss": 0.91352654,
      "num_input_tokens_seen": 2592695,
      "step": 122,
      "time_per_iteration": 2.5309178829193115
    },
    {
      "auxiliary_loss_clip": 0.01431493,
      "auxiliary_loss_mlp": 0.01101067,
      "balance_loss_clip": 1.13161528,
      "balance_loss_mlp": 1.03264093,
      "epoch": 0.01478987554860819,
      "flos": 22741796672640.0,
      "grad_norm": 3.254136886781692,
      "language_loss": 0.89421999,
      "learning_rate": 3.4861674668779934e-06,
      "loss": 0.91954565,
      "num_input_tokens_seen": 2610925,
      "step": 123,
      "time_per_iteration": 2.5972630977630615
    },
    {
      "auxiliary_loss_clip": 0.01433476,
      "auxiliary_loss_mlp": 0.01103961,
      "balance_loss_clip": 1.13010907,
      "balance_loss_mlp": 1.03892064,
      "epoch": 0.01491011843924728,
      "flos": 17198590106880.0,
      "grad_norm": 1.9208478689631097,
      "language_loss": 0.83843648,
      "learning_rate": 3.492033458037272e-06,
      "loss": 0.86381084,
      "num_input_tokens_seen": 2629495,
      "step": 124,
      "time_per_iteration": 2.598525047302246
    },
    {
      "auxiliary_loss_clip": 0.0143205,
      "auxiliary_loss_mlp": 0.01114682,
      "balance_loss_clip": 1.12865806,
      "balance_loss_mlp": 1.04792428,
      "epoch": 0.01503036132988637,
      "flos": 17673867889920.0,
      "grad_norm": 2.2970378727272442,
      "language_loss": 0.87255955,
      "learning_rate": 3.497852332293018e-06,
      "loss": 0.89802688,
      "num_input_tokens_seen": 2645070,
      "step": 125,
      "time_per_iteration": 2.53228759765625
    },
    {
      "auxiliary_loss_clip": 0.01431716,
      "auxiliary_loss_mlp": 0.01100438,
      "balance_loss_clip": 1.13099372,
      "balance_loss_mlp": 1.0388782,
      "epoch": 0.015150604220525462,
      "flos": 18878239935360.0,
      "grad_norm": 5.67102839289843,
      "language_loss": 0.96538627,
      "learning_rate": 3.5036248405242356e-06,
      "loss": 0.99070781,
      "num_input_tokens_seen": 2663825,
      "step": 126,
      "time_per_iteration": 2.539271831512451
    },
    {
      "auxiliary_loss_clip": 0.01425853,
      "auxiliary_loss_mlp": 0.01107097,
      "balance_loss_clip": 1.12774515,
      "balance_loss_mlp": 1.04162753,
      "epoch": 0.015270847111164552,
      "flos": 39420184060800.0,
      "grad_norm": 2.323708118500421,
      "language_loss": 0.82781053,
      "learning_rate": 3.509351715802146e-06,
      "loss": 0.85314012,
      "num_input_tokens_seen": 2684710,
      "step": 127,
      "time_per_iteration": 2.7133429050445557
    },
    {
      "auxiliary_loss_clip": 0.01428186,
      "auxiliary_loss_mlp": 0.01102574,
      "balance_loss_clip": 1.12756145,
      "balance_loss_mlp": 1.03925014,
      "epoch": 0.015391090001803644,
      "flos": 43762466286720.0,
      "grad_norm": 2.1120578981172513,
      "language_loss": 0.78384489,
      "learning_rate": 3.5150336739488763e-06,
      "loss": 0.80915254,
      "num_input_tokens_seen": 2706995,
      "step": 128,
      "time_per_iteration": 2.7446038722991943
    },
    {
      "auxiliary_loss_clip": 0.01417719,
      "auxiliary_loss_mlp": 0.01097335,
      "balance_loss_clip": 1.1275034,
      "balance_loss_mlp": 1.03923237,
      "epoch": 0.015511332892442733,
      "flos": 18916341287040.0,
      "grad_norm": 2.1187648466415268,
      "language_loss": 0.83900869,
      "learning_rate": 3.5206714140744143e-06,
      "loss": 0.86415923,
      "num_input_tokens_seen": 2727050,
      "step": 129,
      "time_per_iteration": 2.584784507751465
    },
    {
      "auxiliary_loss_clip": 0.01423313,
      "auxiliary_loss_mlp": 0.01102943,
      "balance_loss_clip": 1.1324482,
      "balance_loss_mlp": 1.04310036,
      "epoch": 0.015631575783081827,
      "flos": 24535283679360.0,
      "grad_norm": 2.5386299347939905,
      "language_loss": 0.87734556,
      "learning_rate": 3.5262656190928208e-06,
      "loss": 0.9026081,
      "num_input_tokens_seen": 2745350,
      "step": 130,
      "time_per_iteration": 2.7079966068267822
    },
    {
      "auxiliary_loss_clip": 0.01392286,
      "auxiliary_loss_mlp": 0.01064443,
      "balance_loss_clip": 1.18541312,
      "balance_loss_mlp": 1.03297186,
      "epoch": 0.015751818673720917,
      "flos": 62328536098560.0,
      "grad_norm": 1.0456292657828032,
      "language_loss": 0.71486366,
      "learning_rate": 3.5318169562186737e-06,
      "loss": 0.7394309,
      "num_input_tokens_seen": 2814195,
      "step": 131,
      "time_per_iteration": 3.241239547729492
    },
    {
      "auxiliary_loss_clip": 0.01419333,
      "auxiliary_loss_mlp": 0.01098658,
      "balance_loss_clip": 1.12860799,
      "balance_loss_mlp": 1.04067421,
      "epoch": 0.015872061564360006,
      "flos": 23878549365120.0,
      "grad_norm": 1.8465496460238873,
      "language_loss": 0.82010138,
      "learning_rate": 3.5373260774446292e-06,
      "loss": 0.84528124,
      "num_input_tokens_seen": 2834645,
      "step": 132,
      "time_per_iteration": 2.6498820781707764
    },
    {
      "auxiliary_loss_clip": 0.01406771,
      "auxiliary_loss_mlp": 0.0109452,
      "balance_loss_clip": 1.12216783,
      "balance_loss_mlp": 1.04192472,
      "epoch": 0.0159923044549991,
      "flos": 23367899664000.0,
      "grad_norm": 1.816610276639977,
      "language_loss": 0.90245569,
      "learning_rate": 3.542793620000961e-06,
      "loss": 0.92746854,
      "num_input_tokens_seen": 2854120,
      "step": 133,
      "time_per_iteration": 2.7068278789520264
    },
    {
      "auxiliary_loss_clip": 0.01404637,
      "auxiliary_loss_mlp": 0.01096995,
      "balance_loss_clip": 1.12097573,
      "balance_loss_mlp": 1.04685521,
      "epoch": 0.01611254734563819,
      "flos": 17858305249920.0,
      "grad_norm": 2.1487602638720626,
      "language_loss": 0.8693921,
      "learning_rate": 3.5482202067978894e-06,
      "loss": 0.89440846,
      "num_input_tokens_seen": 2871330,
      "step": 134,
      "time_per_iteration": 2.6706464290618896
    },
    {
      "auxiliary_loss_clip": 0.01409732,
      "auxiliary_loss_mlp": 0.01096974,
      "balance_loss_clip": 1.12347353,
      "balance_loss_mlp": 1.04404569,
      "epoch": 0.01623279023627728,
      "flos": 20954774113920.0,
      "grad_norm": 2.1016012392816954,
      "language_loss": 0.75919592,
      "learning_rate": 3.553606446851471e-06,
      "loss": 0.78426301,
      "num_input_tokens_seen": 2888070,
      "step": 135,
      "time_per_iteration": 2.669142484664917
    },
    {
      "auxiliary_loss_clip": 0.01402727,
      "auxiliary_loss_mlp": 0.01080486,
      "balance_loss_clip": 1.12173915,
      "balance_loss_mlp": 1.03327882,
      "epoch": 0.016353033126916373,
      "flos": 15742412743680.0,
      "grad_norm": 1.7929539720210832,
      "language_loss": 0.83258963,
      "learning_rate": 3.5589529356937613e-06,
      "loss": 0.85742176,
      "num_input_tokens_seen": 2906465,
      "step": 136,
      "time_per_iteration": 2.606795310974121
    },
    {
      "auxiliary_loss_clip": 0.01409082,
      "auxiliary_loss_mlp": 0.01084158,
      "balance_loss_clip": 1.12415743,
      "balance_loss_mlp": 1.03201604,
      "epoch": 0.016473276017555463,
      "flos": 18807280617600.0,
      "grad_norm": 1.9193274855913012,
      "language_loss": 0.7679103,
      "learning_rate": 3.5642602557679627e-06,
      "loss": 0.79284275,
      "num_input_tokens_seen": 2924915,
      "step": 137,
      "time_per_iteration": 2.5697855949401855
    },
    {
      "auxiliary_loss_clip": 0.0140553,
      "auxiliary_loss_mlp": 0.01094286,
      "balance_loss_clip": 1.1272825,
      "balance_loss_mlp": 1.04800904,
      "epoch": 0.016593518908194552,
      "flos": 24352641999360.0,
      "grad_norm": 2.3085990215255356,
      "language_loss": 0.8419444,
      "learning_rate": 3.569528976809202e-06,
      "loss": 0.86694252,
      "num_input_tokens_seen": 2942130,
      "step": 138,
      "time_per_iteration": 2.6073203086853027
    },
    {
      "auxiliary_loss_clip": 0.01404215,
      "auxiliary_loss_mlp": 0.01093265,
      "balance_loss_clip": 1.12228751,
      "balance_loss_mlp": 1.04553354,
      "epoch": 0.016713761798833646,
      "flos": 22346133384960.0,
      "grad_norm": 7.540496256687313,
      "language_loss": 0.90022206,
      "learning_rate": 3.5747596562115522e-06,
      "loss": 0.92519683,
      "num_input_tokens_seen": 2962745,
      "step": 139,
      "time_per_iteration": 2.5947086811065674
    },
    {
      "auxiliary_loss_clip": 0.01403611,
      "auxiliary_loss_mlp": 0.01091802,
      "balance_loss_clip": 1.12087345,
      "balance_loss_mlp": 1.04066145,
      "epoch": 0.016834004689472735,
      "flos": 17821820010240.0,
      "grad_norm": 2.8676458507051104,
      "language_loss": 0.90720916,
      "learning_rate": 3.5799528393819138e-06,
      "loss": 0.93216336,
      "num_input_tokens_seen": 2981825,
      "step": 140,
      "time_per_iteration": 2.55751633644104
    },
    {
      "auxiliary_loss_clip": 0.01395526,
      "auxiliary_loss_mlp": 0.010871,
      "balance_loss_clip": 1.11654389,
      "balance_loss_mlp": 1.04084635,
      "epoch": 0.016954247580111825,
      "flos": 20519501103360.0,
      "grad_norm": 2.0087098375941714,
      "language_loss": 0.8796646,
      "learning_rate": 3.585109060081286e-06,
      "loss": 0.90449089,
      "num_input_tokens_seen": 3001625,
      "step": 141,
      "time_per_iteration": 2.563750743865967
    },
    {
      "auxiliary_loss_clip": 0.01404108,
      "auxiliary_loss_mlp": 0.01092154,
      "balance_loss_clip": 1.11993766,
      "balance_loss_mlp": 1.0448997,
      "epoch": 0.017074490470750915,
      "flos": 22088869200000.0,
      "grad_norm": 1.7031538438543403,
      "language_loss": 0.78560388,
      "learning_rate": 3.590228840753992e-06,
      "loss": 0.81056648,
      "num_input_tokens_seen": 3022055,
      "step": 142,
      "time_per_iteration": 2.597093343734741
    },
    {
      "auxiliary_loss_clip": 0.01397211,
      "auxiliary_loss_mlp": 0.01080199,
      "balance_loss_clip": 1.12086499,
      "balance_loss_mlp": 1.0378561,
      "epoch": 0.01719473336139001,
      "flos": 15997270717440.0,
      "grad_norm": 1.9843233589568476,
      "language_loss": 0.87284386,
      "learning_rate": 3.5953126928453423e-06,
      "loss": 0.897618,
      "num_input_tokens_seen": 3039605,
      "step": 143,
      "time_per_iteration": 3.283538818359375
    },
    {
      "auxiliary_loss_clip": 0.01387841,
      "auxiliary_loss_mlp": 0.01085554,
      "balance_loss_clip": 1.11618924,
      "balance_loss_mlp": 1.0472157,
      "epoch": 0.017314976252029098,
      "flos": 22492038430080.0,
      "grad_norm": 1.8916080118088239,
      "language_loss": 0.80455351,
      "learning_rate": 3.600361117108239e-06,
      "loss": 0.82928741,
      "num_input_tokens_seen": 3059405,
      "step": 144,
      "time_per_iteration": 2.581068754196167
    },
    {
      "auxiliary_loss_clip": 0.01393918,
      "auxiliary_loss_mlp": 0.01080574,
      "balance_loss_clip": 1.11899602,
      "balance_loss_mlp": 1.0428561,
      "epoch": 0.017435219142668188,
      "flos": 22018053536640.0,
      "grad_norm": 23.520495357926972,
      "language_loss": 0.97172225,
      "learning_rate": 3.6053746038991616e-06,
      "loss": 0.99646723,
      "num_input_tokens_seen": 3078490,
      "step": 145,
      "time_per_iteration": 4.081583023071289
    },
    {
      "auxiliary_loss_clip": 0.01356617,
      "auxiliary_loss_mlp": 0.0101542,
      "balance_loss_clip": 1.16361785,
      "balance_loss_mlp": 0.99949324,
      "epoch": 0.01755546203330728,
      "flos": 72240526149120.0,
      "grad_norm": 1.0542570106530018,
      "language_loss": 0.58483028,
      "learning_rate": 3.6103536334639843e-06,
      "loss": 0.60855067,
      "num_input_tokens_seen": 3131755,
      "step": 146,
      "time_per_iteration": 3.8999128341674805
    },
    {
      "auxiliary_loss_clip": 0.01390304,
      "auxiliary_loss_mlp": 0.01072192,
      "balance_loss_clip": 1.11865115,
      "balance_loss_mlp": 1.03414071,
      "epoch": 0.01767570492394637,
      "flos": 25337061112320.0,
      "grad_norm": 2.3486331832668954,
      "language_loss": 0.85526109,
      "learning_rate": 3.615298676214041e-06,
      "loss": 0.87988597,
      "num_input_tokens_seen": 3152035,
      "step": 147,
      "time_per_iteration": 2.6055521965026855
    },
    {
      "auxiliary_loss_clip": 0.01392812,
      "auxiliary_loss_mlp": 0.01081255,
      "balance_loss_clip": 1.11835718,
      "balance_loss_mlp": 1.03950739,
      "epoch": 0.01779594781458546,
      "flos": 20449188230400.0,
      "grad_norm": 2.2871681618341677,
      "language_loss": 0.88878107,
      "learning_rate": 3.6202101929928317e-06,
      "loss": 0.91352165,
      "num_input_tokens_seen": 3170625,
      "step": 148,
      "time_per_iteration": 2.563321590423584
    },
    {
      "auxiliary_loss_clip": 0.01388086,
      "auxiliary_loss_mlp": 0.01082942,
      "balance_loss_clip": 1.11643493,
      "balance_loss_mlp": 1.04710722,
      "epoch": 0.017916190705224554,
      "flos": 16253601148800.0,
      "grad_norm": 2.4734217331695585,
      "language_loss": 0.88555336,
      "learning_rate": 3.6250886353337413e-06,
      "loss": 0.91026354,
      "num_input_tokens_seen": 3188155,
      "step": 149,
      "time_per_iteration": 2.5292279720306396
    },
    {
      "auxiliary_loss_clip": 0.01387655,
      "auxiliary_loss_mlp": 0.01075863,
      "balance_loss_clip": 1.11793447,
      "balance_loss_mlp": 1.03976703,
      "epoch": 0.018036433595863644,
      "flos": 23330588411520.0,
      "grad_norm": 5.779166525723225,
      "language_loss": 0.86364198,
      "learning_rate": 3.6299344457091488e-06,
      "loss": 0.88827711,
      "num_input_tokens_seen": 3209015,
      "step": 150,
      "time_per_iteration": 2.5694382190704346
    },
    {
      "auxiliary_loss_clip": 0.01383418,
      "auxiliary_loss_mlp": 0.01086631,
      "balance_loss_clip": 1.11586452,
      "balance_loss_mlp": 1.05186915,
      "epoch": 0.018156676486502734,
      "flos": 18588010043520.0,
      "grad_norm": 2.5639517608020586,
      "language_loss": 0.93838865,
      "learning_rate": 3.634748057771256e-06,
      "loss": 0.96308911,
      "num_input_tokens_seen": 3224955,
      "step": 151,
      "time_per_iteration": 2.5167553424835205
    },
    {
      "auxiliary_loss_clip": 0.01384914,
      "auxiliary_loss_mlp": 0.01063019,
      "balance_loss_clip": 1.11534309,
      "balance_loss_mlp": 1.029688,
      "epoch": 0.018276919377141827,
      "flos": 25448707560960.0,
      "grad_norm": 1.7868278427582986,
      "language_loss": 0.85562885,
      "learning_rate": 3.639529896584965e-06,
      "loss": 0.88010812,
      "num_input_tokens_seen": 3246330,
      "step": 152,
      "time_per_iteration": 2.7165353298187256
    },
    {
      "auxiliary_loss_clip": 0.01385604,
      "auxiliary_loss_mlp": 0.01062984,
      "balance_loss_clip": 1.11629844,
      "balance_loss_mlp": 1.02834129,
      "epoch": 0.018397162267780917,
      "flos": 20047311889920.0,
      "grad_norm": 3.483305583844024,
      "language_loss": 0.88695335,
      "learning_rate": 3.6442803788531233e-06,
      "loss": 0.9114393,
      "num_input_tokens_seen": 3264290,
      "step": 153,
      "time_per_iteration": 2.539876699447632
    },
    {
      "auxiliary_loss_clip": 0.01386922,
      "auxiliary_loss_mlp": 0.01074536,
      "balance_loss_clip": 1.1182282,
      "balance_loss_mlp": 1.03934586,
      "epoch": 0.018517405158420007,
      "flos": 27565282425600.0,
      "grad_norm": 1.9726615999409454,
      "language_loss": 0.958161,
      "learning_rate": 3.6489999131344357e-06,
      "loss": 0.98277557,
      "num_input_tokens_seen": 3287065,
      "step": 154,
      "time_per_iteration": 2.6451680660247803
    },
    {
      "auxiliary_loss_clip": 0.01373233,
      "auxiliary_loss_mlp": 0.01072319,
      "balance_loss_clip": 1.11346209,
      "balance_loss_mlp": 1.04417408,
      "epoch": 0.0186376480490591,
      "flos": 19354056422400.0,
      "grad_norm": 2.0122236204218,
      "language_loss": 0.90547979,
      "learning_rate": 3.653688900054313e-06,
      "loss": 0.92993534,
      "num_input_tokens_seen": 3305595,
      "step": 155,
      "time_per_iteration": 2.5648252964019775
    },
    {
      "auxiliary_loss_clip": 0.01380418,
      "auxiliary_loss_mlp": 0.01065952,
      "balance_loss_clip": 1.11159551,
      "balance_loss_mlp": 1.03312135,
      "epoch": 0.01875789093969819,
      "flos": 26687840993280.0,
      "grad_norm": 1.988869987656064,
      "language_loss": 0.75904888,
      "learning_rate": 3.6583477325089526e-06,
      "loss": 0.78351259,
      "num_input_tokens_seen": 3326135,
      "step": 156,
      "time_per_iteration": 2.5859429836273193
    },
    {
      "auxiliary_loss_clip": 0.01375544,
      "auxiliary_loss_mlp": 0.01063489,
      "balance_loss_clip": 1.11340833,
      "balance_loss_mlp": 1.03020597,
      "epoch": 0.01887813383033728,
      "flos": 24353001135360.0,
      "grad_norm": 5.3745801022829225,
      "language_loss": 1.04340291,
      "learning_rate": 3.6629767958628916e-06,
      "loss": 1.06779325,
      "num_input_tokens_seen": 3343510,
      "step": 157,
      "time_per_iteration": 2.5606372356414795
    },
    {
      "auxiliary_loss_clip": 0.01377812,
      "auxiliary_loss_mlp": 0.01075163,
      "balance_loss_clip": 1.11274338,
      "balance_loss_mlp": 1.04357278,
      "epoch": 0.018998376720976373,
      "flos": 14647532330880.0,
      "grad_norm": 2.2014024090164535,
      "language_loss": 0.85299426,
      "learning_rate": 3.667576468140291e-06,
      "loss": 0.87752402,
      "num_input_tokens_seen": 3361325,
      "step": 158,
      "time_per_iteration": 2.5289859771728516
    },
    {
      "auxiliary_loss_clip": 0.01370209,
      "auxiliary_loss_mlp": 0.01062541,
      "balance_loss_clip": 1.10920525,
      "balance_loss_mlp": 1.03149867,
      "epoch": 0.019118619611615463,
      "flos": 29305261146240.0,
      "grad_norm": 3.0921559314585005,
      "language_loss": 0.88829696,
      "learning_rate": 3.672147120210184e-06,
      "loss": 0.91262448,
      "num_input_tokens_seen": 3377925,
      "step": 159,
      "time_per_iteration": 2.5899529457092285
    },
    {
      "auxiliary_loss_clip": 0.01373643,
      "auxiliary_loss_mlp": 0.01069704,
      "balance_loss_clip": 1.11394787,
      "balance_loss_mlp": 1.03880513,
      "epoch": 0.019238862502254553,
      "flos": 20886723797760.0,
      "grad_norm": 2.5643006719893746,
      "language_loss": 0.86361909,
      "learning_rate": 3.6766891159659177e-06,
      "loss": 0.88805252,
      "num_input_tokens_seen": 3396335,
      "step": 160,
      "time_per_iteration": 2.5433030128479004
    },
    {
      "auxiliary_loss_clip": 0.01372395,
      "auxiliary_loss_mlp": 0.0106666,
      "balance_loss_clip": 1.11453927,
      "balance_loss_mlp": 1.03803825,
      "epoch": 0.019359105392893646,
      "flos": 21360672777600.0,
      "grad_norm": 3.9924227753519195,
      "language_loss": 0.87971127,
      "learning_rate": 3.6812028124990075e-06,
      "loss": 0.90410185,
      "num_input_tokens_seen": 3413605,
      "step": 161,
      "time_per_iteration": 2.527780294418335
    },
    {
      "auxiliary_loss_clip": 0.0136565,
      "auxiliary_loss_mlp": 0.01065416,
      "balance_loss_clip": 1.10913253,
      "balance_loss_mlp": 1.03638899,
      "epoch": 0.019479348283532736,
      "flos": 16283729681280.0,
      "grad_norm": 2.6639411118039056,
      "language_loss": 0.81401479,
      "learning_rate": 3.6856885602676016e-06,
      "loss": 0.8383255,
      "num_input_tokens_seen": 3429640,
      "step": 162,
      "time_per_iteration": 2.5389866828918457
    },
    {
      "auxiliary_loss_clip": 0.0137052,
      "auxiliary_loss_mlp": 0.01066572,
      "balance_loss_clip": 1.10977626,
      "balance_loss_mlp": 1.03791428,
      "epoch": 0.019599591174171826,
      "flos": 22091239497600.0,
      "grad_norm": 2.828854765587588,
      "language_loss": 0.94201756,
      "learning_rate": 3.6901467032597733e-06,
      "loss": 0.9663884,
      "num_input_tokens_seen": 3448125,
      "step": 163,
      "time_per_iteration": 2.5421605110168457
    },
    {
      "auxiliary_loss_clip": 0.01369463,
      "auxiliary_loss_mlp": 0.01073033,
      "balance_loss_clip": 1.10961306,
      "balance_loss_mlp": 1.04110885,
      "epoch": 0.01971983406481092,
      "flos": 19609668581760.0,
      "grad_norm": 2.5245878021827974,
      "language_loss": 0.87415975,
      "learning_rate": 3.694577579151804e-06,
      "loss": 0.89858472,
      "num_input_tokens_seen": 3466535,
      "step": 164,
      "time_per_iteration": 2.551051616668701
    },
    {
      "auxiliary_loss_clip": 0.01370307,
      "auxiliary_loss_mlp": 0.01058437,
      "balance_loss_clip": 1.11197925,
      "balance_loss_mlp": 1.03203201,
      "epoch": 0.01984007695545001,
      "flos": 19099342103040.0,
      "grad_norm": 2.4763386313198685,
      "language_loss": 0.7380904,
      "learning_rate": 3.6989815194616703e-06,
      "loss": 0.76237786,
      "num_input_tokens_seen": 3483730,
      "step": 165,
      "time_per_iteration": 2.515458345413208
    },
    {
      "auxiliary_loss_clip": 0.01373229,
      "auxiliary_loss_mlp": 0.01067448,
      "balance_loss_clip": 1.10993731,
      "balance_loss_mlp": 1.0358336,
      "epoch": 0.0199603198460891,
      "flos": 20848406964480.0,
      "grad_norm": 2.7159205300939555,
      "language_loss": 0.79654348,
      "learning_rate": 3.703358849697888e-06,
      "loss": 0.82095027,
      "num_input_tokens_seen": 3503640,
      "step": 166,
      "time_per_iteration": 2.5567636489868164
    },
    {
      "auxiliary_loss_clip": 0.01364723,
      "auxiliary_loss_mlp": 0.01063335,
      "balance_loss_clip": 1.1095885,
      "balance_loss_mlp": 1.03844368,
      "epoch": 0.020080562736728192,
      "flos": 21870747861120.0,
      "grad_norm": 2.5866212836040825,
      "language_loss": 0.82760274,
      "learning_rate": 3.7077098895038803e-06,
      "loss": 0.85188329,
      "num_input_tokens_seen": 3523010,
      "step": 167,
      "time_per_iteration": 2.5505545139312744
    },
    {
      "auxiliary_loss_clip": 0.01367798,
      "auxiliary_loss_mlp": 0.01073622,
      "balance_loss_clip": 1.11010289,
      "balance_loss_mlp": 1.04340208,
      "epoch": 0.020200805627367282,
      "flos": 21688788539520.0,
      "grad_norm": 2.263880812393261,
      "language_loss": 0.96934098,
      "learning_rate": 3.712034952798045e-06,
      "loss": 0.99375516,
      "num_input_tokens_seen": 3541125,
      "step": 168,
      "time_per_iteration": 2.571009635925293
    },
    {
      "auxiliary_loss_clip": 0.01371553,
      "auxiliary_loss_mlp": 0.01074678,
      "balance_loss_clip": 1.11230755,
      "balance_loss_mlp": 1.04101372,
      "epoch": 0.02032104851800637,
      "flos": 33543043729920.0,
      "grad_norm": 2.2596605948596005,
      "language_loss": 0.84637839,
      "learning_rate": 3.7163343479096656e-06,
      "loss": 0.87084067,
      "num_input_tokens_seen": 3562700,
      "step": 169,
      "time_per_iteration": 2.6729812622070312
    },
    {
      "auxiliary_loss_clip": 0.01364236,
      "auxiliary_loss_mlp": 0.01066157,
      "balance_loss_clip": 1.10935056,
      "balance_loss_mlp": 1.04126632,
      "epoch": 0.020441291408645465,
      "flos": 31686965274240.0,
      "grad_norm": 2.2950568624300445,
      "language_loss": 0.83116275,
      "learning_rate": 3.720608377710802e-06,
      "loss": 0.85546672,
      "num_input_tokens_seen": 3582790,
      "step": 170,
      "time_per_iteration": 3.4660086631774902
    },
    {
      "auxiliary_loss_clip": 0.01359864,
      "auxiliary_loss_mlp": 0.01067473,
      "balance_loss_clip": 1.106866,
      "balance_loss_mlp": 1.03849292,
      "epoch": 0.020561534299284555,
      "flos": 20886687884160.0,
      "grad_norm": 2.06733280657129,
      "language_loss": 0.86449701,
      "learning_rate": 3.7248573397443277e-06,
      "loss": 0.88877034,
      "num_input_tokens_seen": 3601715,
      "step": 171,
      "time_per_iteration": 3.3638720512390137
    },
    {
      "auxiliary_loss_clip": 0.01360444,
      "auxiliary_loss_mlp": 0.01076727,
      "balance_loss_clip": 1.1090709,
      "balance_loss_mlp": 1.05133545,
      "epoch": 0.020681777189923645,
      "flos": 20996610480000.0,
      "grad_norm": 2.7477778407156976,
      "language_loss": 0.97406673,
      "learning_rate": 3.729081526348224e-06,
      "loss": 0.99843848,
      "num_input_tokens_seen": 3620245,
      "step": 172,
      "time_per_iteration": 3.4886553287506104
    },
    {
      "auxiliary_loss_clip": 0.01368393,
      "auxiliary_loss_mlp": 0.01066634,
      "balance_loss_clip": 1.11037779,
      "balance_loss_mlp": 1.03840566,
      "epoch": 0.020802020080562738,
      "flos": 28257532312320.0,
      "grad_norm": 1.9485453569147566,
      "language_loss": 0.84984386,
      "learning_rate": 3.7332812247762777e-06,
      "loss": 0.87419415,
      "num_input_tokens_seen": 3641545,
      "step": 173,
      "time_per_iteration": 2.6462221145629883
    },
    {
      "auxiliary_loss_clip": 0.01362183,
      "auxiliary_loss_mlp": 0.01061125,
      "balance_loss_clip": 1.10998619,
      "balance_loss_mlp": 1.03551865,
      "epoch": 0.020922262971201828,
      "flos": 19681274344320.0,
      "grad_norm": 2.6087817782895577,
      "language_loss": 0.95416784,
      "learning_rate": 3.737456717315293e-06,
      "loss": 0.97840095,
      "num_input_tokens_seen": 3660510,
      "step": 174,
      "time_per_iteration": 2.5701661109924316
    },
    {
      "auxiliary_loss_clip": 0.01361903,
      "auxiliary_loss_mlp": 0.01067359,
      "balance_loss_clip": 1.1093781,
      "balance_loss_mlp": 1.04294479,
      "epoch": 0.021042505861840918,
      "flos": 15666353694720.0,
      "grad_norm": 2.85651182102527,
      "language_loss": 0.90663564,
      "learning_rate": 3.7416082813989552e-06,
      "loss": 0.93092823,
      "num_input_tokens_seen": 3677505,
      "step": 175,
      "time_per_iteration": 2.5261216163635254
    },
    {
      "auxiliary_loss_clip": 0.01365339,
      "auxiliary_loss_mlp": 0.01058131,
      "balance_loss_clip": 1.10889912,
      "balance_loss_mlp": 1.03128517,
      "epoch": 0.02116274875248001,
      "flos": 21142012734720.0,
      "grad_norm": 2.27495886535375,
      "language_loss": 0.89653349,
      "learning_rate": 3.745736189718439e-06,
      "loss": 0.92076814,
      "num_input_tokens_seen": 3696760,
      "step": 176,
      "time_per_iteration": 2.5576980113983154
    },
    {
      "auxiliary_loss_clip": 0.01353085,
      "auxiliary_loss_mlp": 0.01063778,
      "balance_loss_clip": 1.10475862,
      "balance_loss_mlp": 1.03761196,
      "epoch": 0.0212829916431191,
      "flos": 24715770543360.0,
      "grad_norm": 4.478556307336332,
      "language_loss": 0.72251844,
      "learning_rate": 3.749840710329894e-06,
      "loss": 0.74668705,
      "num_input_tokens_seen": 3717465,
      "step": 177,
      "time_per_iteration": 2.576570987701416
    },
    {
      "auxiliary_loss_clip": 0.0136697,
      "auxiliary_loss_mlp": 0.01063153,
      "balance_loss_clip": 1.11330462,
      "balance_loss_mlp": 1.03360069,
      "epoch": 0.02140323453375819,
      "flos": 16645493508480.0,
      "grad_norm": 2.9622824242025048,
      "language_loss": 0.97859812,
      "learning_rate": 3.7539221067588938e-06,
      "loss": 1.00289941,
      "num_input_tokens_seen": 3731440,
      "step": 178,
      "time_per_iteration": 2.5010204315185547
    },
    {
      "auxiliary_loss_clip": 0.01365724,
      "auxiliary_loss_mlp": 0.01064083,
      "balance_loss_clip": 1.10997736,
      "balance_loss_mlp": 1.03506708,
      "epoch": 0.021523477424397284,
      "flos": 20299332689280.0,
      "grad_norm": 5.93421171144816,
      "language_loss": 0.93807912,
      "learning_rate": 3.757980638101964e-06,
      "loss": 0.96237713,
      "num_input_tokens_seen": 3744935,
      "step": 179,
      "time_per_iteration": 2.4971325397491455
    },
    {
      "auxiliary_loss_clip": 0.01364781,
      "auxiliary_loss_mlp": 0.01065596,
      "balance_loss_clip": 1.11039925,
      "balance_loss_mlp": 1.03867805,
      "epoch": 0.021643720315036374,
      "flos": 26104005331200.0,
      "grad_norm": 2.9195362308667208,
      "language_loss": 0.89362854,
      "learning_rate": 3.7620165591252806e-06,
      "loss": 0.91793227,
      "num_input_tokens_seen": 3763035,
      "step": 180,
      "time_per_iteration": 2.5811259746551514
    },
    {
      "auxiliary_loss_clip": 0.01355009,
      "auxiliary_loss_mlp": 0.01058748,
      "balance_loss_clip": 1.10662699,
      "balance_loss_mlp": 1.03687263,
      "epoch": 0.021763963205675464,
      "flos": 24787663614720.0,
      "grad_norm": 2.029100978552359,
      "language_loss": 0.94516218,
      "learning_rate": 3.766030120360636e-06,
      "loss": 0.96929967,
      "num_input_tokens_seen": 3782665,
      "step": 181,
      "time_per_iteration": 2.5605764389038086
    },
    {
      "auxiliary_loss_clip": 0.01358633,
      "auxiliary_loss_mlp": 0.01058302,
      "balance_loss_clip": 1.10661983,
      "balance_loss_mlp": 1.03479373,
      "epoch": 0.021884206096314557,
      "flos": 25813559957760.0,
      "grad_norm": 2.0805572368950656,
      "language_loss": 0.90404499,
      "learning_rate": 3.7700215681987578e-06,
      "loss": 0.92821437,
      "num_input_tokens_seen": 3802435,
      "step": 182,
      "time_per_iteration": 2.576889991760254
    },
    {
      "auxiliary_loss_clip": 0.01357283,
      "auxiliary_loss_mlp": 0.01053508,
      "balance_loss_clip": 1.10716259,
      "balance_loss_mlp": 1.02820003,
      "epoch": 0.022004448986953647,
      "flos": 20082719721600.0,
      "grad_norm": 1.7172382503260541,
      "language_loss": 0.82364362,
      "learning_rate": 3.7739911449800767e-06,
      "loss": 0.84775156,
      "num_input_tokens_seen": 3822490,
      "step": 183,
      "time_per_iteration": 2.524829149246216
    },
    {
      "auxiliary_loss_clip": 0.01356056,
      "auxiliary_loss_mlp": 0.01061446,
      "balance_loss_clip": 1.10409713,
      "balance_loss_mlp": 1.03636479,
      "epoch": 0.022124691877592736,
      "flos": 20480609652480.0,
      "grad_norm": 2.008319145713184,
      "language_loss": 0.80596179,
      "learning_rate": 3.7779390890830114e-06,
      "loss": 0.83013678,
      "num_input_tokens_seen": 3841140,
      "step": 184,
      "time_per_iteration": 2.5443224906921387
    },
    {
      "auxiliary_loss_clip": 0.01356497,
      "auxiliary_loss_mlp": 0.01070826,
      "balance_loss_clip": 1.10481596,
      "balance_loss_mlp": 1.04462421,
      "epoch": 0.02224493476823183,
      "flos": 23586847015680.0,
      "grad_norm": 1.7740214872047086,
      "language_loss": 0.85788798,
      "learning_rate": 3.7818656350098723e-06,
      "loss": 0.88216126,
      "num_input_tokens_seen": 3862090,
      "step": 185,
      "time_per_iteration": 2.5533270835876465
    },
    {
      "auxiliary_loss_clip": 0.01351839,
      "auxiliary_loss_mlp": 0.01058987,
      "balance_loss_clip": 1.10524523,
      "balance_loss_mlp": 1.03328502,
      "epoch": 0.02236517765887092,
      "flos": 16909940413440.0,
      "grad_norm": 2.747550339854854,
      "language_loss": 0.77303183,
      "learning_rate": 3.7857710134704447e-06,
      "loss": 0.79714012,
      "num_input_tokens_seen": 3881025,
      "step": 186,
      "time_per_iteration": 2.522667646408081
    },
    {
      "auxiliary_loss_clip": 0.01352295,
      "auxiliary_loss_mlp": 0.01055836,
      "balance_loss_clip": 1.10568726,
      "balance_loss_mlp": 1.03375876,
      "epoch": 0.02248542054951001,
      "flos": 43508182930560.0,
      "grad_norm": 2.1613814753744527,
      "language_loss": 0.79474986,
      "learning_rate": 3.7896554514633234e-06,
      "loss": 0.81883121,
      "num_input_tokens_seen": 3905310,
      "step": 187,
      "time_per_iteration": 2.7318899631500244
    },
    {
      "auxiliary_loss_clip": 0.01348984,
      "auxiliary_loss_mlp": 0.01058854,
      "balance_loss_clip": 1.10222936,
      "balance_loss_mlp": 1.0367763,
      "epoch": 0.022605663440149103,
      "flos": 23367648268800.0,
      "grad_norm": 1.966986344068411,
      "language_loss": 0.84070307,
      "learning_rate": 3.7935191723550955e-06,
      "loss": 0.8647815,
      "num_input_tokens_seen": 3924265,
      "step": 188,
      "time_per_iteration": 2.5925166606903076
    },
    {
      "auxiliary_loss_clip": 0.01349223,
      "auxiliary_loss_mlp": 0.01057086,
      "balance_loss_clip": 1.10202718,
      "balance_loss_mlp": 1.03584242,
      "epoch": 0.022725906330788193,
      "flos": 29019915504000.0,
      "grad_norm": 2.0474183489670765,
      "language_loss": 0.88372689,
      "learning_rate": 3.797362395957408e-06,
      "loss": 0.90778995,
      "num_input_tokens_seen": 3944830,
      "step": 189,
      "time_per_iteration": 2.5977184772491455
    },
    {
      "auxiliary_loss_clip": 0.01354089,
      "auxiliary_loss_mlp": 0.01068095,
      "balance_loss_clip": 1.10812485,
      "balance_loss_mlp": 1.04343081,
      "epoch": 0.022846149221427282,
      "flos": 24496176746880.0,
      "grad_norm": 4.474439095221145,
      "language_loss": 0.7825408,
      "learning_rate": 3.8011853386020055e-06,
      "loss": 0.80676264,
      "num_input_tokens_seen": 3965735,
      "step": 190,
      "time_per_iteration": 2.5752341747283936
    },
    {
      "auxiliary_loss_clip": 0.01353701,
      "auxiliary_loss_mlp": 0.01064243,
      "balance_loss_clip": 1.10770166,
      "balance_loss_mlp": 1.04021025,
      "epoch": 0.022966392112066376,
      "flos": 15523537219200.0,
      "grad_norm": 2.7017871545029033,
      "language_loss": 0.89773953,
      "learning_rate": 3.804988213213804e-06,
      "loss": 0.92191899,
      "num_input_tokens_seen": 3983975,
      "step": 191,
      "time_per_iteration": 2.508645534515381
    },
    {
      "auxiliary_loss_clip": 0.01319066,
      "auxiliary_loss_mlp": 0.01009938,
      "balance_loss_clip": 1.14052725,
      "balance_loss_mlp": 1.00245142,
      "epoch": 0.023086635002705466,
      "flos": 55650408433920.0,
      "grad_norm": 1.025726961185328,
      "language_loss": 0.63183266,
      "learning_rate": 3.808771229382049e-06,
      "loss": 0.65512264,
      "num_input_tokens_seen": 4043440,
      "step": 192,
      "time_per_iteration": 3.0472300052642822
    },
    {
      "auxiliary_loss_clip": 0.01352973,
      "auxiliary_loss_mlp": 0.01059919,
      "balance_loss_clip": 1.10634768,
      "balance_loss_mlp": 1.03852129,
      "epoch": 0.023206877893344555,
      "flos": 19313441118720.0,
      "grad_norm": 2.024541451471149,
      "language_loss": 0.84481251,
      "learning_rate": 3.8125345934296324e-06,
      "loss": 0.86894149,
      "num_input_tokens_seen": 4061750,
      "step": 193,
      "time_per_iteration": 2.5483314990997314
    },
    {
      "auxiliary_loss_clip": 0.01352932,
      "auxiliary_loss_mlp": 0.01062588,
      "balance_loss_clip": 1.10376501,
      "balance_loss_mlp": 1.03786397,
      "epoch": 0.02332712078398365,
      "flos": 23072965090560.0,
      "grad_norm": 2.197625374148137,
      "language_loss": 0.87897283,
      "learning_rate": 3.81627850848061e-06,
      "loss": 0.90312815,
      "num_input_tokens_seen": 4082345,
      "step": 194,
      "time_per_iteration": 2.6013219356536865
    },
    {
      "auxiliary_loss_clip": 0.01346423,
      "auxiliary_loss_mlp": 0.01058147,
      "balance_loss_clip": 1.10204494,
      "balance_loss_mlp": 1.03558087,
      "epoch": 0.02344736367462274,
      "flos": 24425971614720.0,
      "grad_norm": 2.263599799751454,
      "language_loss": 0.86284614,
      "learning_rate": 3.820003174525994e-06,
      "loss": 0.88689184,
      "num_input_tokens_seen": 4101770,
      "step": 195,
      "time_per_iteration": 2.552396297454834
    },
    {
      "auxiliary_loss_clip": 0.01348494,
      "auxiliary_loss_mlp": 0.01057783,
      "balance_loss_clip": 1.1046145,
      "balance_loss_mlp": 1.03608704,
      "epoch": 0.02356760656526183,
      "flos": 21579799697280.0,
      "grad_norm": 2.2176254273404306,
      "language_loss": 0.82858652,
      "learning_rate": 3.823708788487851e-06,
      "loss": 0.85264927,
      "num_input_tokens_seen": 4118770,
      "step": 196,
      "time_per_iteration": 2.5425665378570557
    },
    {
      "auxiliary_loss_clip": 0.01348521,
      "auxiliary_loss_mlp": 0.01057181,
      "balance_loss_clip": 1.10566902,
      "balance_loss_mlp": 1.03592634,
      "epoch": 0.02368784945590092,
      "flos": 25193598192000.0,
      "grad_norm": 1.9828394880083167,
      "language_loss": 0.84870803,
      "learning_rate": 3.827395544281781e-06,
      "loss": 0.87276506,
      "num_input_tokens_seen": 4141110,
      "step": 197,
      "time_per_iteration": 4.130285978317261
    },
    {
      "auxiliary_loss_clip": 0.01349631,
      "auxiliary_loss_mlp": 0.01065273,
      "balance_loss_clip": 1.10562205,
      "balance_loss_mlp": 1.04292083,
      "epoch": 0.02380809234654001,
      "flos": 27562481164800.0,
      "grad_norm": 1.7996929358659917,
      "language_loss": 0.78978896,
      "learning_rate": 3.831063632877802e-06,
      "loss": 0.81393802,
      "num_input_tokens_seen": 4161430,
      "step": 198,
      "time_per_iteration": 3.312865972518921
    },
    {
      "auxiliary_loss_clip": 0.01347084,
      "auxiliary_loss_mlp": 0.01056915,
      "balance_loss_clip": 1.10723591,
      "balance_loss_mlp": 1.03753185,
      "epoch": 0.0239283352371791,
      "flos": 18259786540800.0,
      "grad_norm": 2.4344548622797615,
      "language_loss": 0.75627375,
      "learning_rate": 3.834713242359712e-06,
      "loss": 0.78031373,
      "num_input_tokens_seen": 4179260,
      "step": 199,
      "time_per_iteration": 2.5321884155273438
    },
    {
      "auxiliary_loss_clip": 0.01347247,
      "auxiliary_loss_mlp": 0.01057285,
      "balance_loss_clip": 1.1024226,
      "balance_loss_mlp": 1.03558946,
      "epoch": 0.02404857812781819,
      "flos": 21395110942080.0,
      "grad_norm": 2.494129525173912,
      "language_loss": 0.87119997,
      "learning_rate": 3.838344557982959e-06,
      "loss": 0.89524531,
      "num_input_tokens_seen": 4200640,
      "step": 200,
      "time_per_iteration": 3.3851282596588135
    },
    {
      "auxiliary_loss_clip": 0.01348536,
      "auxiliary_loss_mlp": 0.01049197,
      "balance_loss_clip": 1.10475302,
      "balance_loss_mlp": 1.02673745,
      "epoch": 0.024168821018457284,
      "flos": 16654256426880.0,
      "grad_norm": 2.470625376244771,
      "language_loss": 0.84843361,
      "learning_rate": 3.841957762231063e-06,
      "loss": 0.87241089,
      "num_input_tokens_seen": 4218170,
      "step": 201,
      "time_per_iteration": 2.5357394218444824
    },
    {
      "auxiliary_loss_clip": 0.01341853,
      "auxiliary_loss_mlp": 0.01057896,
      "balance_loss_clip": 1.10033822,
      "balance_loss_mlp": 1.03584242,
      "epoch": 0.024289063909096374,
      "flos": 22820872464000.0,
      "grad_norm": 2.476660987822322,
      "language_loss": 0.87674701,
      "learning_rate": 3.8455530348706454e-06,
      "loss": 0.90074456,
      "num_input_tokens_seen": 4237770,
      "step": 202,
      "time_per_iteration": 2.5619590282440186
    },
    {
      "auxiliary_loss_clip": 0.01345136,
      "auxiliary_loss_mlp": 0.010552,
      "balance_loss_clip": 1.10541463,
      "balance_loss_mlp": 1.03547096,
      "epoch": 0.024409306799735464,
      "flos": 17748598135680.0,
      "grad_norm": 2.362070777642752,
      "language_loss": 0.77437711,
      "learning_rate": 3.849130553005099e-06,
      "loss": 0.79838049,
      "num_input_tokens_seen": 4255985,
      "step": 203,
      "time_per_iteration": 2.5003836154937744
    },
    {
      "auxiliary_loss_clip": 0.01344699,
      "auxiliary_loss_mlp": 0.01049559,
      "balance_loss_clip": 1.10241246,
      "balance_loss_mlp": 1.02667093,
      "epoch": 0.024529549690374557,
      "flos": 21616213109760.0,
      "grad_norm": 1.8032894091984923,
      "language_loss": 0.83524847,
      "learning_rate": 3.852690491126933e-06,
      "loss": 0.85919106,
      "num_input_tokens_seen": 4276035,
      "step": 204,
      "time_per_iteration": 2.5506951808929443
    },
    {
      "auxiliary_loss_clip": 0.01343673,
      "auxiliary_loss_mlp": 0.01055467,
      "balance_loss_clip": 1.10174227,
      "balance_loss_mlp": 1.03177977,
      "epoch": 0.024649792581013647,
      "flos": 25551662918400.0,
      "grad_norm": 4.238194123992816,
      "language_loss": 0.91067505,
      "learning_rate": 3.856233021168845e-06,
      "loss": 0.9346664,
      "num_input_tokens_seen": 4295730,
      "step": 205,
      "time_per_iteration": 2.551254987716675
    },
    {
      "auxiliary_loss_clip": 0.01338023,
      "auxiliary_loss_mlp": 0.010472,
      "balance_loss_clip": 1.09752989,
      "balance_loss_mlp": 1.02825785,
      "epoch": 0.024770035471652737,
      "flos": 34495574544000.0,
      "grad_norm": 2.226045364864961,
      "language_loss": 0.91115505,
      "learning_rate": 3.859758312553544e-06,
      "loss": 0.93500733,
      "num_input_tokens_seen": 4317950,
      "step": 206,
      "time_per_iteration": 2.6521823406219482
    },
    {
      "auxiliary_loss_clip": 0.01342001,
      "auxiliary_loss_mlp": 0.01051684,
      "balance_loss_clip": 1.10351491,
      "balance_loss_mlp": 1.03128719,
      "epoch": 0.02489027836229183,
      "flos": 21505428587520.0,
      "grad_norm": 1.9814023794480469,
      "language_loss": 0.91689438,
      "learning_rate": 3.8632665322423735e-06,
      "loss": 0.94083124,
      "num_input_tokens_seen": 4337605,
      "step": 207,
      "time_per_iteration": 2.5182037353515625
    },
    {
      "auxiliary_loss_clip": 0.01338489,
      "auxiliary_loss_mlp": 0.01059293,
      "balance_loss_clip": 1.1004703,
      "balance_loss_mlp": 1.04073191,
      "epoch": 0.02501052125293092,
      "flos": 23219013790080.0,
      "grad_norm": 1.914145409308628,
      "language_loss": 0.85959029,
      "learning_rate": 3.866757844782762e-06,
      "loss": 0.88356805,
      "num_input_tokens_seen": 4358110,
      "step": 208,
      "time_per_iteration": 2.5664336681365967
    },
    {
      "auxiliary_loss_clip": 0.01342622,
      "auxiliary_loss_mlp": 0.01054137,
      "balance_loss_clip": 1.10157871,
      "balance_loss_mlp": 1.03409767,
      "epoch": 0.02513076414357001,
      "flos": 26388920010240.0,
      "grad_norm": 15.1708282356871,
      "language_loss": 0.91061246,
      "learning_rate": 3.870232412354527e-06,
      "loss": 0.93457997,
      "num_input_tokens_seen": 4374955,
      "step": 209,
      "time_per_iteration": 2.6321704387664795
    },
    {
      "auxiliary_loss_clip": 0.01335507,
      "auxiliary_loss_mlp": 0.0105523,
      "balance_loss_clip": 1.09646273,
      "balance_loss_mlp": 1.03583479,
      "epoch": 0.025251007034209103,
      "flos": 13590430047360.0,
      "grad_norm": 2.1241385342590227,
      "language_loss": 0.92736185,
      "learning_rate": 3.873690394815086e-06,
      "loss": 0.95126927,
      "num_input_tokens_seen": 4391535,
      "step": 210,
      "time_per_iteration": 2.5423169136047363
    },
    {
      "auxiliary_loss_clip": 0.01339219,
      "auxiliary_loss_mlp": 0.01054992,
      "balance_loss_clip": 1.09936965,
      "balance_loss_mlp": 1.034917,
      "epoch": 0.025371249924848193,
      "flos": 15049229103360.0,
      "grad_norm": 3.321852608143173,
      "language_loss": 0.91218531,
      "learning_rate": 3.877131949743587e-06,
      "loss": 0.93612742,
      "num_input_tokens_seen": 4408400,
      "step": 211,
      "time_per_iteration": 2.4876456260681152
    },
    {
      "auxiliary_loss_clip": 0.0133876,
      "auxiliary_loss_mlp": 0.01055501,
      "balance_loss_clip": 1.10025716,
      "balance_loss_mlp": 1.03516412,
      "epoch": 0.025491492815487283,
      "flos": 25553853648000.0,
      "grad_norm": 2.103648655188884,
      "language_loss": 0.78103673,
      "learning_rate": 3.880557232483993e-06,
      "loss": 0.80497932,
      "num_input_tokens_seen": 4427840,
      "step": 212,
      "time_per_iteration": 2.6018950939178467
    },
    {
      "auxiliary_loss_clip": 0.01335488,
      "auxiliary_loss_mlp": 0.01051906,
      "balance_loss_clip": 1.09696269,
      "balance_loss_mlp": 1.03212941,
      "epoch": 0.025611735706126376,
      "flos": 20630752502400.0,
      "grad_norm": 2.1900250434084696,
      "language_loss": 0.86721957,
      "learning_rate": 3.883966396187164e-06,
      "loss": 0.89109349,
      "num_input_tokens_seen": 4447110,
      "step": 213,
      "time_per_iteration": 2.566042423248291
    },
    {
      "auxiliary_loss_clip": 0.01336886,
      "auxiliary_loss_mlp": 0.01052611,
      "balance_loss_clip": 1.10119438,
      "balance_loss_mlp": 1.03348958,
      "epoch": 0.025731978596765466,
      "flos": 19062282245760.0,
      "grad_norm": 1.9916228233515108,
      "language_loss": 0.89934182,
      "learning_rate": 3.887359591851937e-06,
      "loss": 0.92323679,
      "num_input_tokens_seen": 4464715,
      "step": 214,
      "time_per_iteration": 2.525160789489746
    },
    {
      "auxiliary_loss_clip": 0.01333278,
      "auxiliary_loss_mlp": 0.01050859,
      "balance_loss_clip": 1.09851432,
      "balance_loss_mlp": 1.03328776,
      "epoch": 0.025852221487404556,
      "flos": 22163814927360.0,
      "grad_norm": 8.454173393897214,
      "language_loss": 0.92169595,
      "learning_rate": 3.890736968365265e-06,
      "loss": 0.94553733,
      "num_input_tokens_seen": 4485030,
      "step": 215,
      "time_per_iteration": 2.5270135402679443
    },
    {
      "auxiliary_loss_clip": 0.013382,
      "auxiliary_loss_mlp": 0.0105043,
      "balance_loss_clip": 1.09811544,
      "balance_loss_mlp": 1.03010511,
      "epoch": 0.02597246437804365,
      "flos": 26541971861760.0,
      "grad_norm": 1.8550744407279083,
      "language_loss": 0.8513149,
      "learning_rate": 3.894098672541412e-06,
      "loss": 0.87520117,
      "num_input_tokens_seen": 4505935,
      "step": 216,
      "time_per_iteration": 2.5637948513031006
    },
    {
      "auxiliary_loss_clip": 0.01336082,
      "auxiliary_loss_mlp": 0.01057457,
      "balance_loss_clip": 1.09959114,
      "balance_loss_mlp": 1.03671455,
      "epoch": 0.02609270726868274,
      "flos": 32671671696000.0,
      "grad_norm": 1.7073178841011303,
      "language_loss": 0.75569415,
      "learning_rate": 3.89744484916025e-06,
      "loss": 0.77962959,
      "num_input_tokens_seen": 4527045,
      "step": 217,
      "time_per_iteration": 2.588130235671997
    },
    {
      "auxiliary_loss_clip": 0.01340088,
      "auxiliary_loss_mlp": 0.01050608,
      "balance_loss_clip": 1.10134017,
      "balance_loss_mlp": 1.02823234,
      "epoch": 0.02621295015932183,
      "flos": 26243553669120.0,
      "grad_norm": 2.255343365486263,
      "language_loss": 0.87289047,
      "learning_rate": 3.900775641004673e-06,
      "loss": 0.89679754,
      "num_input_tokens_seen": 4546360,
      "step": 218,
      "time_per_iteration": 2.5551741123199463
    },
    {
      "auxiliary_loss_clip": 0.01343493,
      "auxiliary_loss_mlp": 0.01060884,
      "balance_loss_clip": 1.10227466,
      "balance_loss_mlp": 1.03877068,
      "epoch": 0.026333193049960922,
      "flos": 42921402353280.0,
      "grad_norm": 3.705127125970801,
      "language_loss": 0.74619102,
      "learning_rate": 3.904091188897156e-06,
      "loss": 0.77023482,
      "num_input_tokens_seen": 4565495,
      "step": 219,
      "time_per_iteration": 2.7027428150177
    },
    {
      "auxiliary_loss_clip": 0.01338782,
      "auxiliary_loss_mlp": 0.01053421,
      "balance_loss_clip": 1.10037887,
      "balance_loss_mlp": 1.03310752,
      "epoch": 0.026453435940600012,
      "flos": 17963846386560.0,
      "grad_norm": 2.333891137870851,
      "language_loss": 0.82201034,
      "learning_rate": 3.90739163173548e-06,
      "loss": 0.84593236,
      "num_input_tokens_seen": 4583330,
      "step": 220,
      "time_per_iteration": 2.491384983062744
    },
    {
      "auxiliary_loss_clip": 0.01335663,
      "auxiliary_loss_mlp": 0.01057733,
      "balance_loss_clip": 1.0970881,
      "balance_loss_mlp": 1.03733683,
      "epoch": 0.026573678831239102,
      "flos": 18984319776000.0,
      "grad_norm": 3.9093254662745123,
      "language_loss": 0.88454098,
      "learning_rate": 3.910677106527646e-06,
      "loss": 0.90847492,
      "num_input_tokens_seen": 4600520,
      "step": 221,
      "time_per_iteration": 2.5039801597595215
    },
    {
      "auxiliary_loss_clip": 0.01331758,
      "auxiliary_loss_mlp": 0.01062439,
      "balance_loss_clip": 1.09794939,
      "balance_loss_mlp": 1.0421021,
      "epoch": 0.026693921721878195,
      "flos": 29241448634880.0,
      "grad_norm": 2.1988648216028372,
      "language_loss": 0.84371102,
      "learning_rate": 3.913947748426004e-06,
      "loss": 0.86765301,
      "num_input_tokens_seen": 4617340,
      "step": 222,
      "time_per_iteration": 2.5915815830230713
    },
    {
      "auxiliary_loss_clip": 0.01335595,
      "auxiliary_loss_mlp": 0.0105167,
      "balance_loss_clip": 1.09959495,
      "balance_loss_mlp": 1.03071356,
      "epoch": 0.026814164612517285,
      "flos": 14128083797760.0,
      "grad_norm": 2.5807941621450246,
      "language_loss": 0.76323134,
      "learning_rate": 3.9172036907606136e-06,
      "loss": 0.78710401,
      "num_input_tokens_seen": 4630820,
      "step": 223,
      "time_per_iteration": 2.4758107662200928
    },
    {
      "auxiliary_loss_clip": 0.01338329,
      "auxiliary_loss_mlp": 0.01054051,
      "balance_loss_clip": 1.09975648,
      "balance_loss_mlp": 1.03399992,
      "epoch": 0.026934407503156375,
      "flos": 23511973115520.0,
      "grad_norm": 1.8244897677455512,
      "language_loss": 0.95209765,
      "learning_rate": 3.920445065071855e-06,
      "loss": 0.97602141,
      "num_input_tokens_seen": 4651985,
      "step": 224,
      "time_per_iteration": 4.094212293624878
    },
    {
      "auxiliary_loss_clip": 0.01333694,
      "auxiliary_loss_mlp": 0.01061993,
      "balance_loss_clip": 1.09745455,
      "balance_loss_mlp": 1.04264522,
      "epoch": 0.027054650393795468,
      "flos": 28950356816640.0,
      "grad_norm": 2.1903199206348716,
      "language_loss": 0.79833156,
      "learning_rate": 3.923672001142322e-06,
      "loss": 0.82228839,
      "num_input_tokens_seen": 4672295,
      "step": 225,
      "time_per_iteration": 2.5812952518463135
    },
    {
      "auxiliary_loss_clip": 0.01334386,
      "auxiliary_loss_mlp": 0.01051926,
      "balance_loss_clip": 1.09725332,
      "balance_loss_mlp": 1.03280449,
      "epoch": 0.027174893284434558,
      "flos": 31431568596480.0,
      "grad_norm": 1.9130635602140988,
      "language_loss": 0.84407049,
      "learning_rate": 3.926884627027996e-06,
      "loss": 0.86793363,
      "num_input_tokens_seen": 4696065,
      "step": 226,
      "time_per_iteration": 3.30666184425354
    },
    {
      "auxiliary_loss_clip": 0.0133,
      "auxiliary_loss_mlp": 0.01045361,
      "balance_loss_clip": 1.09446359,
      "balance_loss_mlp": 1.02639532,
      "epoch": 0.027295136175073648,
      "flos": 22054466949120.0,
      "grad_norm": 2.266675272555932,
      "language_loss": 0.77258754,
      "learning_rate": 3.930083069088744e-06,
      "loss": 0.79634106,
      "num_input_tokens_seen": 4716065,
      "step": 227,
      "time_per_iteration": 2.57615065574646
    },
    {
      "auxiliary_loss_clip": 0.01317147,
      "auxiliary_loss_mlp": 0.01027762,
      "balance_loss_clip": 1.12835515,
      "balance_loss_mlp": 1.02084768,
      "epoch": 0.02741537906571274,
      "flos": 60800752972800.0,
      "grad_norm": 0.9793810352439645,
      "language_loss": 0.59292054,
      "learning_rate": 3.933267452018137e-06,
      "loss": 0.61636961,
      "num_input_tokens_seen": 4775860,
      "step": 228,
      "time_per_iteration": 3.1271283626556396
    },
    {
      "auxiliary_loss_clip": 0.01330305,
      "auxiliary_loss_mlp": 0.01056578,
      "balance_loss_clip": 1.09739828,
      "balance_loss_mlp": 1.03802955,
      "epoch": 0.02753562195635183,
      "flos": 24606278910720.0,
      "grad_norm": 2.0805264017294913,
      "language_loss": 0.84471035,
      "learning_rate": 3.936437898872622e-06,
      "loss": 0.86857915,
      "num_input_tokens_seen": 4795835,
      "step": 229,
      "time_per_iteration": 2.5416674613952637
    },
    {
      "auxiliary_loss_clip": 0.01332696,
      "auxiliary_loss_mlp": 0.01048781,
      "balance_loss_clip": 1.09696043,
      "balance_loss_mlp": 1.02836049,
      "epoch": 0.02765586484699092,
      "flos": 34094236907520.0,
      "grad_norm": 4.196638163116638,
      "language_loss": 0.79951966,
      "learning_rate": 3.9395945311000525e-06,
      "loss": 0.82333446,
      "num_input_tokens_seen": 4817460,
      "step": 230,
      "time_per_iteration": 2.6501245498657227
    },
    {
      "auxiliary_loss_clip": 0.01328296,
      "auxiliary_loss_mlp": 0.01062981,
      "balance_loss_clip": 1.09433126,
      "balance_loss_mlp": 1.04295397,
      "epoch": 0.027776107737630014,
      "flos": 14829922615680.0,
      "grad_norm": 4.799849890255863,
      "language_loss": 0.91169065,
      "learning_rate": 3.942737468567608e-06,
      "loss": 0.93560344,
      "num_input_tokens_seen": 4835475,
      "step": 231,
      "time_per_iteration": 2.4865331649780273
    },
    {
      "auxiliary_loss_clip": 0.01332173,
      "auxiliary_loss_mlp": 0.01047512,
      "balance_loss_clip": 1.09514129,
      "balance_loss_mlp": 1.02831948,
      "epoch": 0.027896350628269104,
      "flos": 47920347066240.0,
      "grad_norm": 2.407686638977946,
      "language_loss": 0.86245292,
      "learning_rate": 3.9458668295891026e-06,
      "loss": 0.88624966,
      "num_input_tokens_seen": 4857760,
      "step": 232,
      "time_per_iteration": 2.766127586364746
    },
    {
      "auxiliary_loss_clip": 0.01330782,
      "auxiliary_loss_mlp": 0.01051355,
      "balance_loss_clip": 1.09353232,
      "balance_loss_mlp": 1.03194809,
      "epoch": 0.028016593518908194,
      "flos": 21684550734720.0,
      "grad_norm": 3.3940873960802813,
      "language_loss": 0.87038279,
      "learning_rate": 3.948982730951712e-06,
      "loss": 0.89420426,
      "num_input_tokens_seen": 4875855,
      "step": 233,
      "time_per_iteration": 2.536513328552246
    },
    {
      "auxiliary_loss_clip": 0.01335091,
      "auxiliary_loss_mlp": 0.01056696,
      "balance_loss_clip": 1.09713316,
      "balance_loss_mlp": 1.03318834,
      "epoch": 0.028136836409547287,
      "flos": 18439483305600.0,
      "grad_norm": 4.662616251948372,
      "language_loss": 0.82083666,
      "learning_rate": 3.9520852879421254e-06,
      "loss": 0.84475452,
      "num_input_tokens_seen": 4893200,
      "step": 234,
      "time_per_iteration": 2.5240654945373535
    },
    {
      "auxiliary_loss_clip": 0.01332404,
      "auxiliary_loss_mlp": 0.01051934,
      "balance_loss_clip": 1.09618068,
      "balance_loss_mlp": 1.03401673,
      "epoch": 0.028257079300186377,
      "flos": 31576934937600.0,
      "grad_norm": 2.294840581378164,
      "language_loss": 0.81954813,
      "learning_rate": 3.955174614372137e-06,
      "loss": 0.84339154,
      "num_input_tokens_seen": 4912965,
      "step": 235,
      "time_per_iteration": 2.612030267715454
    },
    {
      "auxiliary_loss_clip": 0.0132922,
      "auxiliary_loss_mlp": 0.01049563,
      "balance_loss_clip": 1.09645557,
      "balance_loss_mlp": 1.03040648,
      "epoch": 0.028377322190825467,
      "flos": 23513337832320.0,
      "grad_norm": 2.729156368168894,
      "language_loss": 0.84348512,
      "learning_rate": 3.9582508226037045e-06,
      "loss": 0.86727297,
      "num_input_tokens_seen": 4933105,
      "step": 236,
      "time_per_iteration": 2.550361156463623
    },
    {
      "auxiliary_loss_clip": 0.01332618,
      "auxiliary_loss_mlp": 0.01061397,
      "balance_loss_clip": 1.09373152,
      "balance_loss_mlp": 1.03892636,
      "epoch": 0.02849756508146456,
      "flos": 20479604071680.0,
      "grad_norm": 2.857816227850878,
      "language_loss": 0.94016969,
      "learning_rate": 3.9613140235734636e-06,
      "loss": 0.9641099,
      "num_input_tokens_seen": 4950085,
      "step": 237,
      "time_per_iteration": 2.472715377807617
    },
    {
      "auxiliary_loss_clip": 0.01332312,
      "auxiliary_loss_mlp": 0.01045453,
      "balance_loss_clip": 1.09715402,
      "balance_loss_mlp": 1.02667785,
      "epoch": 0.02861780797210365,
      "flos": 14283362292480.0,
      "grad_norm": 4.194121235526327,
      "language_loss": 0.80957949,
      "learning_rate": 3.96436432681674e-06,
      "loss": 0.8333571,
      "num_input_tokens_seen": 4968075,
      "step": 238,
      "time_per_iteration": 2.5335562229156494
    },
    {
      "auxiliary_loss_clip": 0.01334338,
      "auxiliary_loss_mlp": 0.01054725,
      "balance_loss_clip": 1.09744024,
      "balance_loss_mlp": 1.03437567,
      "epoch": 0.02873805086274274,
      "flos": 25808532053760.0,
      "grad_norm": 2.060409859808705,
      "language_loss": 0.88946575,
      "learning_rate": 3.967401840491044e-06,
      "loss": 0.91335636,
      "num_input_tokens_seen": 4987355,
      "step": 239,
      "time_per_iteration": 2.5432772636413574
    },
    {
      "auxiliary_loss_clip": 0.01328087,
      "auxiliary_loss_mlp": 0.01057179,
      "balance_loss_clip": 1.09561753,
      "balance_loss_mlp": 1.03855896,
      "epoch": 0.028858293753381833,
      "flos": 17304238984320.0,
      "grad_norm": 3.0439542351028046,
      "language_loss": 0.87860805,
      "learning_rate": 3.97042667139909e-06,
      "loss": 0.90246069,
      "num_input_tokens_seen": 5004680,
      "step": 240,
      "time_per_iteration": 2.512861728668213
    },
    {
      "auxiliary_loss_clip": 0.0132862,
      "auxiliary_loss_mlp": 0.01052316,
      "balance_loss_clip": 1.09397149,
      "balance_loss_mlp": 1.03241992,
      "epoch": 0.028978536644020923,
      "flos": 23038347358080.0,
      "grad_norm": 2.03767138979313,
      "language_loss": 0.87639117,
      "learning_rate": 3.973438925011327e-06,
      "loss": 0.90020055,
      "num_input_tokens_seen": 5022965,
      "step": 241,
      "time_per_iteration": 2.496835947036743
    },
    {
      "auxiliary_loss_clip": 0.0132654,
      "auxiliary_loss_mlp": 0.01047732,
      "balance_loss_clip": 1.09358788,
      "balance_loss_mlp": 1.02875447,
      "epoch": 0.029098779534660012,
      "flos": 28329712692480.0,
      "grad_norm": 2.6352511280463022,
      "language_loss": 0.91494656,
      "learning_rate": 3.976438705488002e-06,
      "loss": 0.93868935,
      "num_input_tokens_seen": 5042625,
      "step": 242,
      "time_per_iteration": 2.5821566581726074
    },
    {
      "auxiliary_loss_clip": 0.0133004,
      "auxiliary_loss_mlp": 0.0106253,
      "balance_loss_clip": 1.09737158,
      "balance_loss_mlp": 1.04287243,
      "epoch": 0.029219022425299106,
      "flos": 13881665520000.0,
      "grad_norm": 2.8638357480503385,
      "language_loss": 0.93090218,
      "learning_rate": 3.9794261157007744e-06,
      "loss": 0.9548279,
      "num_input_tokens_seen": 5060380,
      "step": 243,
      "time_per_iteration": 2.48112416267395
    },
    {
      "auxiliary_loss_clip": 0.01333415,
      "auxiliary_loss_mlp": 0.010599,
      "balance_loss_clip": 1.09950519,
      "balance_loss_mlp": 1.03788197,
      "epoch": 0.029339265315938196,
      "flos": 19422501788160.0,
      "grad_norm": 2.5439542615663093,
      "language_loss": 0.85124165,
      "learning_rate": 3.982401257253887e-06,
      "loss": 0.87517476,
      "num_input_tokens_seen": 5078720,
      "step": 244,
      "time_per_iteration": 2.5113370418548584
    },
    {
      "auxiliary_loss_clip": 0.01329587,
      "auxiliary_loss_mlp": 0.01052992,
      "balance_loss_clip": 1.09556663,
      "balance_loss_mlp": 1.034693,
      "epoch": 0.029459508206577285,
      "flos": 15669550005120.0,
      "grad_norm": 3.6735227563624417,
      "language_loss": 0.8975901,
      "learning_rate": 3.985364230504893e-06,
      "loss": 0.92141593,
      "num_input_tokens_seen": 5096605,
      "step": 245,
      "time_per_iteration": 2.4872395992279053
    },
    {
      "auxiliary_loss_clip": 0.01330785,
      "auxiliary_loss_mlp": 0.01054508,
      "balance_loss_clip": 1.09677505,
      "balance_loss_mlp": 1.03448081,
      "epoch": 0.02957975109721638,
      "flos": 28220975245440.0,
      "grad_norm": 2.133019447498821,
      "language_loss": 0.84576523,
      "learning_rate": 3.988315134584976e-06,
      "loss": 0.86961818,
      "num_input_tokens_seen": 5116285,
      "step": 246,
      "time_per_iteration": 2.577279806137085
    },
    {
      "auxiliary_loss_clip": 0.01337697,
      "auxiliary_loss_mlp": 0.01056189,
      "balance_loss_clip": 1.0988425,
      "balance_loss_mlp": 1.03548253,
      "epoch": 0.02969999398785547,
      "flos": 24315869450880.0,
      "grad_norm": 1.8027259019294628,
      "language_loss": 0.80639583,
      "learning_rate": 3.991254067418851e-06,
      "loss": 0.83033466,
      "num_input_tokens_seen": 5136825,
      "step": 247,
      "time_per_iteration": 2.5394349098205566
    },
    {
      "auxiliary_loss_clip": 0.01325694,
      "auxiliary_loss_mlp": 0.01063459,
      "balance_loss_clip": 1.09413528,
      "balance_loss_mlp": 1.04563725,
      "epoch": 0.02982023687849456,
      "flos": 35078584193280.0,
      "grad_norm": 2.826097546735251,
      "language_loss": 0.83170283,
      "learning_rate": 3.994181125744254e-06,
      "loss": 0.85559434,
      "num_input_tokens_seen": 5158630,
      "step": 248,
      "time_per_iteration": 2.648813009262085
    },
    {
      "auxiliary_loss_clip": 0.0132894,
      "auxiliary_loss_mlp": 0.0106246,
      "balance_loss_clip": 1.09665143,
      "balance_loss_mlp": 1.04321957,
      "epoch": 0.02994047976913365,
      "flos": 26177155378560.0,
      "grad_norm": 1.9274659594757733,
      "language_loss": 0.74127686,
      "learning_rate": 3.99709640513106e-06,
      "loss": 0.76519084,
      "num_input_tokens_seen": 5179510,
      "step": 249,
      "time_per_iteration": 2.541853904724121
    },
    {
      "auxiliary_loss_clip": 0.01335702,
      "auxiliary_loss_mlp": 0.01056672,
      "balance_loss_clip": 1.09877229,
      "balance_loss_mlp": 1.03472519,
      "epoch": 0.03006072265977274,
      "flos": 25625028447360.0,
      "grad_norm": 2.639220075085666,
      "language_loss": 0.85327381,
      "learning_rate": 4e-06,
      "loss": 0.87719756,
      "num_input_tokens_seen": 5199345,
      "step": 250,
      "time_per_iteration": 3.3674063682556152
    },
    {
      "auxiliary_loss_clip": 0.0133235,
      "auxiliary_loss_mlp": 0.0105963,
      "balance_loss_clip": 1.10075378,
      "balance_loss_mlp": 1.04021049,
      "epoch": 0.03018096555041183,
      "flos": 22127078292480.0,
      "grad_norm": 2.797832132696717,
      "language_loss": 0.88679457,
      "learning_rate": 3.999999848300794e-06,
      "loss": 0.91071439,
      "num_input_tokens_seen": 5218330,
      "step": 251,
      "time_per_iteration": 2.5344204902648926
    },
    {
      "auxiliary_loss_clip": 0.01324506,
      "auxiliary_loss_mlp": 0.01047208,
      "balance_loss_clip": 1.09216452,
      "balance_loss_mlp": 1.02925515,
      "epoch": 0.030301208441050925,
      "flos": 30188197359360.0,
      "grad_norm": 1.7729267942120406,
      "language_loss": 0.89119875,
      "learning_rate": 3.999999393203203e-06,
      "loss": 0.91491592,
      "num_input_tokens_seen": 5240740,
      "step": 252,
      "time_per_iteration": 3.3662946224212646
    },
    {
      "auxiliary_loss_clip": 0.01323646,
      "auxiliary_loss_mlp": 0.01058355,
      "balance_loss_clip": 1.09078753,
      "balance_loss_mlp": 1.03929329,
      "epoch": 0.030421451331690014,
      "flos": 23621392920960.0,
      "grad_norm": 1.9072666815437946,
      "language_loss": 0.85072172,
      "learning_rate": 3.999998634707293e-06,
      "loss": 0.87454176,
      "num_input_tokens_seen": 5260290,
      "step": 253,
      "time_per_iteration": 2.5464928150177
    },
    {
      "auxiliary_loss_clip": 0.01331429,
      "auxiliary_loss_mlp": 0.01061993,
      "balance_loss_clip": 1.09886837,
      "balance_loss_mlp": 1.04160786,
      "epoch": 0.030541694222329104,
      "flos": 27928446883200.0,
      "grad_norm": 3.0094924849644085,
      "language_loss": 0.96523541,
      "learning_rate": 3.999997572813182e-06,
      "loss": 0.98916966,
      "num_input_tokens_seen": 5278100,
      "step": 254,
      "time_per_iteration": 2.538090467453003
    },
    {
      "auxiliary_loss_clip": 0.01328456,
      "auxiliary_loss_mlp": 0.01055714,
      "balance_loss_clip": 1.09506273,
      "balance_loss_mlp": 1.03644967,
      "epoch": 0.030661937112968194,
      "flos": 18588441006720.0,
      "grad_norm": 1.907144006268887,
      "language_loss": 0.87671506,
      "learning_rate": 3.999996207521028e-06,
      "loss": 0.90055686,
      "num_input_tokens_seen": 5296810,
      "step": 255,
      "time_per_iteration": 2.4777615070343018
    },
    {
      "auxiliary_loss_clip": 0.01328575,
      "auxiliary_loss_mlp": 0.01057344,
      "balance_loss_clip": 1.09551883,
      "balance_loss_mlp": 1.03579092,
      "epoch": 0.030782180003607287,
      "flos": 12969139478400.0,
      "grad_norm": 2.630087479242214,
      "language_loss": 0.81917191,
      "learning_rate": 3.999994538831039e-06,
      "loss": 0.84303105,
      "num_input_tokens_seen": 5313395,
      "step": 256,
      "time_per_iteration": 2.4693548679351807
    },
    {
      "auxiliary_loss_clip": 0.01328031,
      "auxiliary_loss_mlp": 0.01053787,
      "balance_loss_clip": 1.09430885,
      "balance_loss_mlp": 1.03367591,
      "epoch": 0.030902422894246377,
      "flos": 23335364920320.0,
      "grad_norm": 8.519923681958868,
      "language_loss": 0.85939026,
      "learning_rate": 3.99999256674347e-06,
      "loss": 0.88320839,
      "num_input_tokens_seen": 5333545,
      "step": 257,
      "time_per_iteration": 2.531607151031494
    },
    {
      "auxiliary_loss_clip": 0.01296267,
      "auxiliary_loss_mlp": 0.01008478,
      "balance_loss_clip": 1.11796868,
      "balance_loss_mlp": 1.0019927,
      "epoch": 0.031022665784885467,
      "flos": 55094151438720.0,
      "grad_norm": 1.0132202412455447,
      "language_loss": 0.53468871,
      "learning_rate": 3.999990291258618e-06,
      "loss": 0.55773616,
      "num_input_tokens_seen": 5392235,
      "step": 258,
      "time_per_iteration": 3.0783286094665527
    },
    {
      "auxiliary_loss_clip": 0.01326762,
      "auxiliary_loss_mlp": 0.01057195,
      "balance_loss_clip": 1.09442806,
      "balance_loss_mlp": 1.03748977,
      "epoch": 0.03114290867552456,
      "flos": 19317786664320.0,
      "grad_norm": 2.3677448873833984,
      "language_loss": 0.86438769,
      "learning_rate": 3.999987712376829e-06,
      "loss": 0.88822722,
      "num_input_tokens_seen": 5410555,
      "step": 259,
      "time_per_iteration": 2.4961798191070557
    },
    {
      "auxiliary_loss_clip": 0.01328668,
      "auxiliary_loss_mlp": 0.01059615,
      "balance_loss_clip": 1.09715486,
      "balance_loss_mlp": 1.04081631,
      "epoch": 0.031263151566163654,
      "flos": 20959442881920.0,
      "grad_norm": 2.309682837408446,
      "language_loss": 0.82062805,
      "learning_rate": 3.999984830098494e-06,
      "loss": 0.84451097,
      "num_input_tokens_seen": 5430135,
      "step": 260,
      "time_per_iteration": 2.4797780513763428
    },
    {
      "auxiliary_loss_clip": 0.01331891,
      "auxiliary_loss_mlp": 0.01055408,
      "balance_loss_clip": 1.0967623,
      "balance_loss_mlp": 1.03485608,
      "epoch": 0.03138339445680274,
      "flos": 14793006412800.0,
      "grad_norm": 2.7322718632988288,
      "language_loss": 0.9802134,
      "learning_rate": 3.999981644424051e-06,
      "loss": 1.00408649,
      "num_input_tokens_seen": 5444935,
      "step": 261,
      "time_per_iteration": 2.465020179748535
    },
    {
      "auxiliary_loss_clip": 0.01330338,
      "auxiliary_loss_mlp": 0.01051452,
      "balance_loss_clip": 1.09725332,
      "balance_loss_mlp": 1.0316391,
      "epoch": 0.03150363734744183,
      "flos": 11655599022720.0,
      "grad_norm": 2.2541732946491306,
      "language_loss": 0.86155766,
      "learning_rate": 3.999978155353982e-06,
      "loss": 0.88537562,
      "num_input_tokens_seen": 5462080,
      "step": 262,
      "time_per_iteration": 2.4766831398010254
    },
    {
      "auxiliary_loss_clip": 0.01323473,
      "auxiliary_loss_mlp": 0.01065818,
      "balance_loss_clip": 1.0926218,
      "balance_loss_mlp": 1.04511166,
      "epoch": 0.03162388023808092,
      "flos": 33727732485120.0,
      "grad_norm": 4.161232950573966,
      "language_loss": 0.80336976,
      "learning_rate": 3.9999743628888186e-06,
      "loss": 0.82726264,
      "num_input_tokens_seen": 5483870,
      "step": 263,
      "time_per_iteration": 2.617018699645996
    },
    {
      "auxiliary_loss_clip": 0.01325075,
      "auxiliary_loss_mlp": 0.01042232,
      "balance_loss_clip": 1.09219074,
      "balance_loss_mlp": 1.0230031,
      "epoch": 0.03174412312872001,
      "flos": 20810952057600.0,
      "grad_norm": 9.978602729849717,
      "language_loss": 0.89641875,
      "learning_rate": 3.999970267029133e-06,
      "loss": 0.92009181,
      "num_input_tokens_seen": 5502830,
      "step": 264,
      "time_per_iteration": 2.5015156269073486
    },
    {
      "auxiliary_loss_clip": 0.01321173,
      "auxiliary_loss_mlp": 0.01060116,
      "balance_loss_clip": 1.09389937,
      "balance_loss_mlp": 1.04069638,
      "epoch": 0.0318643660193591,
      "flos": 23727939638400.0,
      "grad_norm": 1.8503969254326815,
      "language_loss": 0.80222452,
      "learning_rate": 3.999965867775548e-06,
      "loss": 0.82603741,
      "num_input_tokens_seen": 5523225,
      "step": 265,
      "time_per_iteration": 2.556765079498291
    },
    {
      "auxiliary_loss_clip": 0.01323105,
      "auxiliary_loss_mlp": 0.01061523,
      "balance_loss_clip": 1.09384465,
      "balance_loss_mlp": 1.04317629,
      "epoch": 0.0319846089099982,
      "flos": 13917863450880.0,
      "grad_norm": 2.736637202703191,
      "language_loss": 0.86848921,
      "learning_rate": 3.9999611651287315e-06,
      "loss": 0.89233541,
      "num_input_tokens_seen": 5541380,
      "step": 266,
      "time_per_iteration": 2.471550941467285
    },
    {
      "auxiliary_loss_clip": 0.01328158,
      "auxiliary_loss_mlp": 0.01054323,
      "balance_loss_clip": 1.09495115,
      "balance_loss_mlp": 1.03356862,
      "epoch": 0.03210485180063729,
      "flos": 14753253035520.0,
      "grad_norm": 2.5628512471578584,
      "language_loss": 0.7873143,
      "learning_rate": 3.999956159089396e-06,
      "loss": 0.81113911,
      "num_input_tokens_seen": 5558830,
      "step": 267,
      "time_per_iteration": 2.476736545562744
    },
    {
      "auxiliary_loss_clip": 0.01326713,
      "auxiliary_loss_mlp": 0.01057854,
      "balance_loss_clip": 1.09469223,
      "balance_loss_mlp": 1.03972244,
      "epoch": 0.03222509469127638,
      "flos": 28913153304960.0,
      "grad_norm": 2.250910903993193,
      "language_loss": 0.793401,
      "learning_rate": 3.999950849658302e-06,
      "loss": 0.81724668,
      "num_input_tokens_seen": 5577750,
      "step": 268,
      "time_per_iteration": 2.5841994285583496
    },
    {
      "auxiliary_loss_clip": 0.01325965,
      "auxiliary_loss_mlp": 0.01058177,
      "balance_loss_clip": 1.09414136,
      "balance_loss_mlp": 1.03786385,
      "epoch": 0.03234533758191547,
      "flos": 16946389739520.0,
      "grad_norm": 2.2257782945992743,
      "language_loss": 0.84361476,
      "learning_rate": 3.999945236836254e-06,
      "loss": 0.8674562,
      "num_input_tokens_seen": 5596715,
      "step": 269,
      "time_per_iteration": 2.490841865539551
    },
    {
      "auxiliary_loss_clip": 0.01329006,
      "auxiliary_loss_mlp": 0.01057391,
      "balance_loss_clip": 1.09690475,
      "balance_loss_mlp": 1.03626752,
      "epoch": 0.03246558047255456,
      "flos": 18989096284800.0,
      "grad_norm": 2.569715786990049,
      "language_loss": 0.94627094,
      "learning_rate": 3.999939320624103e-06,
      "loss": 0.97013497,
      "num_input_tokens_seen": 5611865,
      "step": 270,
      "time_per_iteration": 2.4710710048675537
    },
    {
      "auxiliary_loss_clip": 0.01326989,
      "auxiliary_loss_mlp": 0.01051009,
      "balance_loss_clip": 1.09583044,
      "balance_loss_mlp": 1.03099358,
      "epoch": 0.03258582336319365,
      "flos": 23728334688000.0,
      "grad_norm": 2.495996761069405,
      "language_loss": 0.89760584,
      "learning_rate": 3.999933101022749e-06,
      "loss": 0.92138577,
      "num_input_tokens_seen": 5632270,
      "step": 271,
      "time_per_iteration": 2.546382427215576
    },
    {
      "auxiliary_loss_clip": 0.01327292,
      "auxiliary_loss_mlp": 0.01051441,
      "balance_loss_clip": 1.09495282,
      "balance_loss_mlp": 1.03284478,
      "epoch": 0.032706066253832745,
      "flos": 27670823562240.0,
      "grad_norm": 1.8727922379610746,
      "language_loss": 0.87050784,
      "learning_rate": 3.999926578033132e-06,
      "loss": 0.89429522,
      "num_input_tokens_seen": 5652085,
      "step": 272,
      "time_per_iteration": 2.530982732772827
    },
    {
      "auxiliary_loss_clip": 0.0132161,
      "auxiliary_loss_mlp": 0.01057184,
      "balance_loss_clip": 1.09209037,
      "balance_loss_mlp": 1.03547549,
      "epoch": 0.032826309144471835,
      "flos": 45624685968000.0,
      "grad_norm": 3.0050853689461428,
      "language_loss": 0.62911057,
      "learning_rate": 3.999919751656244e-06,
      "loss": 0.65289855,
      "num_input_tokens_seen": 5678985,
      "step": 273,
      "time_per_iteration": 2.800109624862671
    },
    {
      "auxiliary_loss_clip": 0.01328418,
      "auxiliary_loss_mlp": 0.01060482,
      "balance_loss_clip": 1.09550667,
      "balance_loss_mlp": 1.03948891,
      "epoch": 0.032946552035110925,
      "flos": 25812374808960.0,
      "grad_norm": 2.322352006589266,
      "language_loss": 0.7599318,
      "learning_rate": 3.9999126218931195e-06,
      "loss": 0.78382081,
      "num_input_tokens_seen": 5697020,
      "step": 274,
      "time_per_iteration": 2.6181647777557373
    },
    {
      "auxiliary_loss_clip": 0.01329075,
      "auxiliary_loss_mlp": 0.01043304,
      "balance_loss_clip": 1.10030007,
      "balance_loss_mlp": 1.02433741,
      "epoch": 0.033066794925750015,
      "flos": 15121984101120.0,
      "grad_norm": 2.811901821651355,
      "language_loss": 0.89784193,
      "learning_rate": 3.99990518874484e-06,
      "loss": 0.92156571,
      "num_input_tokens_seen": 5713460,
      "step": 275,
      "time_per_iteration": 2.5285236835479736
    },
    {
      "auxiliary_loss_clip": 0.01326152,
      "auxiliary_loss_mlp": 0.01056915,
      "balance_loss_clip": 1.09786391,
      "balance_loss_mlp": 1.03812742,
      "epoch": 0.033187037816389105,
      "flos": 22776593973120.0,
      "grad_norm": 2.2833585325987493,
      "language_loss": 0.92448968,
      "learning_rate": 3.999897452212534e-06,
      "loss": 0.94832033,
      "num_input_tokens_seen": 5730790,
      "step": 276,
      "time_per_iteration": 3.212458848953247
    },
    {
      "auxiliary_loss_clip": 0.01324309,
      "auxiliary_loss_mlp": 0.01057462,
      "balance_loss_clip": 1.09496808,
      "balance_loss_mlp": 1.03720808,
      "epoch": 0.033307280707028195,
      "flos": 23331414424320.0,
      "grad_norm": 2.246199531471302,
      "language_loss": 1.00024891,
      "learning_rate": 3.999889412297374e-06,
      "loss": 1.02406669,
      "num_input_tokens_seen": 5750215,
      "step": 277,
      "time_per_iteration": 4.014597415924072
    },
    {
      "auxiliary_loss_clip": 0.01324652,
      "auxiliary_loss_mlp": 0.01056125,
      "balance_loss_clip": 1.09345031,
      "balance_loss_mlp": 1.03658628,
      "epoch": 0.03342752359766729,
      "flos": 28840290566400.0,
      "grad_norm": 2.171332193358441,
      "language_loss": 0.79345751,
      "learning_rate": 3.999881069000581e-06,
      "loss": 0.81726527,
      "num_input_tokens_seen": 5769945,
      "step": 278,
      "time_per_iteration": 2.5589611530303955
    },
    {
      "auxiliary_loss_clip": 0.01326878,
      "auxiliary_loss_mlp": 0.01049083,
      "balance_loss_clip": 1.09371853,
      "balance_loss_mlp": 1.02837682,
      "epoch": 0.03354776648830638,
      "flos": 19384544090880.0,
      "grad_norm": 2.6396558253376146,
      "language_loss": 0.86682546,
      "learning_rate": 3.99987242232342e-06,
      "loss": 0.89058506,
      "num_input_tokens_seen": 5784950,
      "step": 279,
      "time_per_iteration": 3.2472403049468994
    },
    {
      "auxiliary_loss_clip": 0.01323198,
      "auxiliary_loss_mlp": 0.01053247,
      "balance_loss_clip": 1.09604263,
      "balance_loss_mlp": 1.03470969,
      "epoch": 0.03366800937894547,
      "flos": 17858628472320.0,
      "grad_norm": 2.212349221809365,
      "language_loss": 0.79644454,
      "learning_rate": 3.9998634722672026e-06,
      "loss": 0.82020903,
      "num_input_tokens_seen": 5805005,
      "step": 280,
      "time_per_iteration": 2.491605281829834
    },
    {
      "auxiliary_loss_clip": 0.01326337,
      "auxiliary_loss_mlp": 0.01066022,
      "balance_loss_clip": 1.09720159,
      "balance_loss_mlp": 1.04605484,
      "epoch": 0.03378825226958456,
      "flos": 35951033635200.0,
      "grad_norm": 2.4358114478581654,
      "language_loss": 0.78494656,
      "learning_rate": 3.999854218833286e-06,
      "loss": 0.8088702,
      "num_input_tokens_seen": 5825825,
      "step": 281,
      "time_per_iteration": 2.6168274879455566
    },
    {
      "auxiliary_loss_clip": 0.01323577,
      "auxiliary_loss_mlp": 0.01059152,
      "balance_loss_clip": 1.09538186,
      "balance_loss_mlp": 1.04091346,
      "epoch": 0.03390849516022365,
      "flos": 25702488126720.0,
      "grad_norm": 2.022818390782637,
      "language_loss": 0.82071722,
      "learning_rate": 3.999844662023075e-06,
      "loss": 0.84454453,
      "num_input_tokens_seen": 5845700,
      "step": 282,
      "time_per_iteration": 2.511321544647217
    },
    {
      "auxiliary_loss_clip": 0.01317425,
      "auxiliary_loss_mlp": 0.01053907,
      "balance_loss_clip": 1.09283721,
      "balance_loss_mlp": 1.03666949,
      "epoch": 0.03402873805086274,
      "flos": 21284505987840.0,
      "grad_norm": 1.9396013610186866,
      "language_loss": 0.92154366,
      "learning_rate": 3.999834801838018e-06,
      "loss": 0.94525707,
      "num_input_tokens_seen": 5864680,
      "step": 283,
      "time_per_iteration": 2.5031917095184326
    },
    {
      "auxiliary_loss_clip": 0.01321393,
      "auxiliary_loss_mlp": 0.01053047,
      "balance_loss_clip": 1.09528017,
      "balance_loss_mlp": 1.03517771,
      "epoch": 0.03414898094150183,
      "flos": 22710913954560.0,
      "grad_norm": 2.03666911592958,
      "language_loss": 0.74111319,
      "learning_rate": 3.9998246382796115e-06,
      "loss": 0.76485753,
      "num_input_tokens_seen": 5884260,
      "step": 284,
      "time_per_iteration": 2.497770071029663
    },
    {
      "auxiliary_loss_clip": 0.01324433,
      "auxiliary_loss_mlp": 0.01058694,
      "balance_loss_clip": 1.09386194,
      "balance_loss_mlp": 1.0381546,
      "epoch": 0.03426922383214093,
      "flos": 18879927874560.0,
      "grad_norm": 2.276783457812551,
      "language_loss": 0.90867716,
      "learning_rate": 3.999814171349399e-06,
      "loss": 0.93250841,
      "num_input_tokens_seen": 5902120,
      "step": 285,
      "time_per_iteration": 2.496124505996704
    },
    {
      "auxiliary_loss_clip": 0.01326015,
      "auxiliary_loss_mlp": 0.01053323,
      "balance_loss_clip": 1.0973444,
      "balance_loss_mlp": 1.03540587,
      "epoch": 0.03438946672278002,
      "flos": 34752012716160.0,
      "grad_norm": 2.0360046683578106,
      "language_loss": 0.73605114,
      "learning_rate": 3.9998034010489655e-06,
      "loss": 0.75984448,
      "num_input_tokens_seen": 5925810,
      "step": 286,
      "time_per_iteration": 2.6471352577209473
    },
    {
      "auxiliary_loss_clip": 0.01324548,
      "auxiliary_loss_mlp": 0.01054929,
      "balance_loss_clip": 1.0960505,
      "balance_loss_mlp": 1.03757179,
      "epoch": 0.03450970961341911,
      "flos": 22164102236160.0,
      "grad_norm": 3.0498976756429377,
      "language_loss": 0.76116049,
      "learning_rate": 3.999792327379946e-06,
      "loss": 0.78495526,
      "num_input_tokens_seen": 5945185,
      "step": 287,
      "time_per_iteration": 2.5090694427490234
    },
    {
      "auxiliary_loss_clip": 0.01325303,
      "auxiliary_loss_mlp": 0.01050245,
      "balance_loss_clip": 1.09675694,
      "balance_loss_mlp": 1.03281713,
      "epoch": 0.034629952504058197,
      "flos": 21725740656000.0,
      "grad_norm": 2.792977563468203,
      "language_loss": 0.96175218,
      "learning_rate": 3.999780950344021e-06,
      "loss": 0.98550761,
      "num_input_tokens_seen": 5963375,
      "step": 288,
      "time_per_iteration": 2.4771006107330322
    },
    {
      "auxiliary_loss_clip": 0.01325275,
      "auxiliary_loss_mlp": 0.01049369,
      "balance_loss_clip": 1.09518385,
      "balance_loss_mlp": 1.027601,
      "epoch": 0.034750195394697286,
      "flos": 20047994248320.0,
      "grad_norm": 2.027509424432824,
      "language_loss": 0.8257435,
      "learning_rate": 3.999769269942916e-06,
      "loss": 0.84948993,
      "num_input_tokens_seen": 5983415,
      "step": 289,
      "time_per_iteration": 2.493520736694336
    },
    {
      "auxiliary_loss_clip": 0.01320128,
      "auxiliary_loss_mlp": 0.01055031,
      "balance_loss_clip": 1.09360933,
      "balance_loss_mlp": 1.03643429,
      "epoch": 0.034870438285336376,
      "flos": 27965865876480.0,
      "grad_norm": 2.121481484742185,
      "language_loss": 0.81162673,
      "learning_rate": 3.999757286178402e-06,
      "loss": 0.83537829,
      "num_input_tokens_seen": 6005850,
      "step": 290,
      "time_per_iteration": 2.543306350708008
    },
    {
      "auxiliary_loss_clip": 0.01323303,
      "auxiliary_loss_mlp": 0.010636,
      "balance_loss_clip": 1.09563518,
      "balance_loss_mlp": 1.04424024,
      "epoch": 0.03499068117597547,
      "flos": 22017514832640.0,
      "grad_norm": 2.086523879822895,
      "language_loss": 0.90592933,
      "learning_rate": 3.999744999052299e-06,
      "loss": 0.92979836,
      "num_input_tokens_seen": 6027240,
      "step": 291,
      "time_per_iteration": 2.562394857406616
    },
    {
      "auxiliary_loss_clip": 0.0128671,
      "auxiliary_loss_mlp": 0.01012902,
      "balance_loss_clip": 1.11327028,
      "balance_loss_mlp": 1.00703681,
      "epoch": 0.03511092406661456,
      "flos": 57242147725440.0,
      "grad_norm": 0.9591230893959098,
      "language_loss": 0.61219919,
      "learning_rate": 3.9997324085664675e-06,
      "loss": 0.63519526,
      "num_input_tokens_seen": 6087470,
      "step": 292,
      "time_per_iteration": 3.0745739936828613
    },
    {
      "auxiliary_loss_clip": 0.01319743,
      "auxiliary_loss_mlp": 0.01053928,
      "balance_loss_clip": 1.09142983,
      "balance_loss_mlp": 1.03561711,
      "epoch": 0.03523116695725365,
      "flos": 22928065626240.0,
      "grad_norm": 2.460062076307442,
      "language_loss": 0.92140335,
      "learning_rate": 3.999719514722821e-06,
      "loss": 0.94514006,
      "num_input_tokens_seen": 6107600,
      "step": 293,
      "time_per_iteration": 2.54526424407959
    },
    {
      "auxiliary_loss_clip": 0.0132064,
      "auxiliary_loss_mlp": 0.01050096,
      "balance_loss_clip": 1.09219575,
      "balance_loss_mlp": 1.03265548,
      "epoch": 0.03535140984789274,
      "flos": 36903241226880.0,
      "grad_norm": 14.850532076979507,
      "language_loss": 0.74638635,
      "learning_rate": 3.999706317523314e-06,
      "loss": 0.77009374,
      "num_input_tokens_seen": 6126160,
      "step": 294,
      "time_per_iteration": 2.626112222671509
    },
    {
      "auxiliary_loss_clip": 0.0131673,
      "auxiliary_loss_mlp": 0.01053142,
      "balance_loss_clip": 1.09249496,
      "balance_loss_mlp": 1.03505826,
      "epoch": 0.03547165273853183,
      "flos": 20449152316800.0,
      "grad_norm": 2.060968967095047,
      "language_loss": 0.8603425,
      "learning_rate": 3.999692816969948e-06,
      "loss": 0.88404125,
      "num_input_tokens_seen": 6145695,
      "step": 295,
      "time_per_iteration": 2.481374740600586
    },
    {
      "auxiliary_loss_clip": 0.01283503,
      "auxiliary_loss_mlp": 0.01007954,
      "balance_loss_clip": 1.11384797,
      "balance_loss_mlp": 1.00223207,
      "epoch": 0.03559189562917092,
      "flos": 69850564871040.0,
      "grad_norm": 1.0055991068282755,
      "language_loss": 0.69475597,
      "learning_rate": 3.999679013064772e-06,
      "loss": 0.71767056,
      "num_input_tokens_seen": 6212440,
      "step": 296,
      "time_per_iteration": 3.1304142475128174
    },
    {
      "auxiliary_loss_clip": 0.01315316,
      "auxiliary_loss_mlp": 0.01058982,
      "balance_loss_clip": 1.09249008,
      "balance_loss_mlp": 1.04193497,
      "epoch": 0.03571213851981002,
      "flos": 21651944163840.0,
      "grad_norm": 3.011515107846017,
      "language_loss": 0.85254478,
      "learning_rate": 3.99966490580988e-06,
      "loss": 0.87628782,
      "num_input_tokens_seen": 6229800,
      "step": 297,
      "time_per_iteration": 2.505828857421875
    },
    {
      "auxiliary_loss_clip": 0.013233,
      "auxiliary_loss_mlp": 0.01051723,
      "balance_loss_clip": 1.09614527,
      "balance_loss_mlp": 1.03270948,
      "epoch": 0.03583238141044911,
      "flos": 43945610757120.0,
      "grad_norm": 2.5292134574557132,
      "language_loss": 0.65658277,
      "learning_rate": 3.999650495207411e-06,
      "loss": 0.68033296,
      "num_input_tokens_seen": 6255825,
      "step": 298,
      "time_per_iteration": 2.6806800365448
    },
    {
      "auxiliary_loss_clip": 0.01315715,
      "auxiliary_loss_mlp": 0.01049853,
      "balance_loss_clip": 1.0920856,
      "balance_loss_mlp": 1.0328418,
      "epoch": 0.0359526243010882,
      "flos": 18910810592640.0,
      "grad_norm": 4.2646416163823515,
      "language_loss": 0.9031477,
      "learning_rate": 3.999635781259553e-06,
      "loss": 0.92680335,
      "num_input_tokens_seen": 6271090,
      "step": 299,
      "time_per_iteration": 2.471494197845459
    },
    {
      "auxiliary_loss_clip": 0.01272655,
      "auxiliary_loss_mlp": 0.01007885,
      "balance_loss_clip": 1.10798907,
      "balance_loss_mlp": 1.00211513,
      "epoch": 0.03607286719172729,
      "flos": 61668892782720.0,
      "grad_norm": 0.9178392524474561,
      "language_loss": 0.52277815,
      "learning_rate": 3.999620763968535e-06,
      "loss": 0.54558349,
      "num_input_tokens_seen": 6329965,
      "step": 300,
      "time_per_iteration": 2.9586501121520996
    },
    {
      "auxiliary_loss_clip": 0.01323079,
      "auxiliary_loss_mlp": 0.01049887,
      "balance_loss_clip": 1.09530163,
      "balance_loss_mlp": 1.03311396,
      "epoch": 0.03619311008236638,
      "flos": 27819062991360.0,
      "grad_norm": 2.1908678634763694,
      "language_loss": 0.86540449,
      "learning_rate": 3.999605443336638e-06,
      "loss": 0.88913417,
      "num_input_tokens_seen": 6352095,
      "step": 301,
      "time_per_iteration": 2.5821969509124756
    },
    {
      "auxiliary_loss_clip": 0.01320179,
      "auxiliary_loss_mlp": 0.01053489,
      "balance_loss_clip": 1.09262156,
      "balance_loss_mlp": 1.0340817,
      "epoch": 0.03631335297300547,
      "flos": 13621133197440.0,
      "grad_norm": 2.6700834292101097,
      "language_loss": 0.89139813,
      "learning_rate": 3.999589819366185e-06,
      "loss": 0.91513479,
      "num_input_tokens_seen": 6365885,
      "step": 302,
      "time_per_iteration": 2.472255229949951
    },
    {
      "auxiliary_loss_clip": 0.01316509,
      "auxiliary_loss_mlp": 0.01051107,
      "balance_loss_clip": 1.09076905,
      "balance_loss_mlp": 1.03137779,
      "epoch": 0.036433595863644565,
      "flos": 27631788456960.0,
      "grad_norm": 3.281292136026117,
      "language_loss": 0.84874022,
      "learning_rate": 3.999573892059547e-06,
      "loss": 0.87241638,
      "num_input_tokens_seen": 6385015,
      "step": 303,
      "time_per_iteration": 3.3288962841033936
    },
    {
      "auxiliary_loss_clip": 0.01326562,
      "auxiliary_loss_mlp": 0.0105965,
      "balance_loss_clip": 1.09554911,
      "balance_loss_mlp": 1.0395391,
      "epoch": 0.036553838754283655,
      "flos": 24572020314240.0,
      "grad_norm": 2.718027888192216,
      "language_loss": 0.81306165,
      "learning_rate": 3.999557661419138e-06,
      "loss": 0.83692384,
      "num_input_tokens_seen": 6405165,
      "step": 304,
      "time_per_iteration": 3.988342046737671
    },
    {
      "auxiliary_loss_clip": 0.01321581,
      "auxiliary_loss_mlp": 0.01042176,
      "balance_loss_clip": 1.09456217,
      "balance_loss_mlp": 1.02465189,
      "epoch": 0.036674081644922744,
      "flos": 23404313076480.0,
      "grad_norm": 3.1688567243544563,
      "language_loss": 0.81432939,
      "learning_rate": 3.9995411274474225e-06,
      "loss": 0.83796698,
      "num_input_tokens_seen": 6424445,
      "step": 305,
      "time_per_iteration": 2.489184856414795
    },
    {
      "auxiliary_loss_clip": 0.01323551,
      "auxiliary_loss_mlp": 0.01053611,
      "balance_loss_clip": 1.09553683,
      "balance_loss_mlp": 1.0358727,
      "epoch": 0.036794324535561834,
      "flos": 27489690253440.0,
      "grad_norm": 4.647853277057265,
      "language_loss": 0.81564802,
      "learning_rate": 3.999524290146908e-06,
      "loss": 0.83941966,
      "num_input_tokens_seen": 6444650,
      "step": 306,
      "time_per_iteration": 3.3386616706848145
    },
    {
      "auxiliary_loss_clip": 0.01324667,
      "auxiliary_loss_mlp": 0.01059676,
      "balance_loss_clip": 1.09589934,
      "balance_loss_mlp": 1.04049516,
      "epoch": 0.036914567426200924,
      "flos": 19463476227840.0,
      "grad_norm": 2.42341769940735,
      "language_loss": 0.92598432,
      "learning_rate": 3.9995071495201485e-06,
      "loss": 0.94982779,
      "num_input_tokens_seen": 6461755,
      "step": 307,
      "time_per_iteration": 2.5385689735412598
    },
    {
      "auxiliary_loss_clip": 0.01322306,
      "auxiliary_loss_mlp": 0.01056324,
      "balance_loss_clip": 1.09576869,
      "balance_loss_mlp": 1.03832304,
      "epoch": 0.037034810316840014,
      "flos": 22309324922880.0,
      "grad_norm": 2.6026151172489076,
      "language_loss": 0.98076129,
      "learning_rate": 3.999489705569744e-06,
      "loss": 1.0045476,
      "num_input_tokens_seen": 6479455,
      "step": 308,
      "time_per_iteration": 2.5232832431793213
    },
    {
      "auxiliary_loss_clip": 0.01320153,
      "auxiliary_loss_mlp": 0.01044042,
      "balance_loss_clip": 1.09404218,
      "balance_loss_mlp": 1.02530241,
      "epoch": 0.03715505320747911,
      "flos": 18588333265920.0,
      "grad_norm": 2.1492955699959118,
      "language_loss": 0.86386603,
      "learning_rate": 3.999471958298341e-06,
      "loss": 0.88750798,
      "num_input_tokens_seen": 6498365,
      "step": 309,
      "time_per_iteration": 2.507586717605591
    },
    {
      "auxiliary_loss_clip": 0.01326119,
      "auxiliary_loss_mlp": 0.01052026,
      "balance_loss_clip": 1.09911644,
      "balance_loss_mlp": 1.03298867,
      "epoch": 0.0372752960981182,
      "flos": 35955343267200.0,
      "grad_norm": 1.9799868767549416,
      "language_loss": 0.76216698,
      "learning_rate": 3.999453907708631e-06,
      "loss": 0.7859484,
      "num_input_tokens_seen": 6520770,
      "step": 310,
      "time_per_iteration": 2.6458089351654053
    },
    {
      "auxiliary_loss_clip": 0.01316498,
      "auxiliary_loss_mlp": 0.01044755,
      "balance_loss_clip": 1.09334171,
      "balance_loss_mlp": 1.02748203,
      "epoch": 0.03739553898875729,
      "flos": 20814040627200.0,
      "grad_norm": 1.8829938853206694,
      "language_loss": 0.81268615,
      "learning_rate": 3.999435553803353e-06,
      "loss": 0.8362987,
      "num_input_tokens_seen": 6540170,
      "step": 311,
      "time_per_iteration": 2.5235283374786377
    },
    {
      "auxiliary_loss_clip": 0.01320133,
      "auxiliary_loss_mlp": 0.01054541,
      "balance_loss_clip": 1.0928483,
      "balance_loss_mlp": 1.03701758,
      "epoch": 0.03751578187939638,
      "flos": 20264140339200.0,
      "grad_norm": 2.379492147221191,
      "language_loss": 0.83417982,
      "learning_rate": 3.999416896585292e-06,
      "loss": 0.85792661,
      "num_input_tokens_seen": 6557200,
      "step": 312,
      "time_per_iteration": 2.4931626319885254
    },
    {
      "auxiliary_loss_clip": 0.01319991,
      "auxiliary_loss_mlp": 0.01054994,
      "balance_loss_clip": 1.0940001,
      "balance_loss_mlp": 1.03625393,
      "epoch": 0.03763602477003547,
      "flos": 20668063754880.0,
      "grad_norm": 3.6707894514985147,
      "language_loss": 0.85528255,
      "learning_rate": 3.9993979360572775e-06,
      "loss": 0.87903249,
      "num_input_tokens_seen": 6577340,
      "step": 313,
      "time_per_iteration": 2.5432841777801514
    },
    {
      "auxiliary_loss_clip": 0.01322799,
      "auxiliary_loss_mlp": 0.01064693,
      "balance_loss_clip": 1.0964855,
      "balance_loss_mlp": 1.04527426,
      "epoch": 0.03775626766067456,
      "flos": 16691352197760.0,
      "grad_norm": 2.7196920311134063,
      "language_loss": 0.83156359,
      "learning_rate": 3.999378672222185e-06,
      "loss": 0.85543853,
      "num_input_tokens_seen": 6595125,
      "step": 314,
      "time_per_iteration": 2.512808084487915
    },
    {
      "auxiliary_loss_clip": 0.01316471,
      "auxiliary_loss_mlp": 0.0104871,
      "balance_loss_clip": 1.09336519,
      "balance_loss_mlp": 1.03013682,
      "epoch": 0.03787651055131366,
      "flos": 21141797253120.0,
      "grad_norm": 2.1787931936594367,
      "language_loss": 0.83227986,
      "learning_rate": 3.9993591050829385e-06,
      "loss": 0.85593164,
      "num_input_tokens_seen": 6612990,
      "step": 315,
      "time_per_iteration": 2.509321689605713
    },
    {
      "auxiliary_loss_clip": 0.01323122,
      "auxiliary_loss_mlp": 0.01050633,
      "balance_loss_clip": 1.09716415,
      "balance_loss_mlp": 1.03300178,
      "epoch": 0.037996753441952746,
      "flos": 22018089450240.0,
      "grad_norm": 1.8344304305829875,
      "language_loss": 0.79111093,
      "learning_rate": 3.999339234642506e-06,
      "loss": 0.81484848,
      "num_input_tokens_seen": 6632740,
      "step": 316,
      "time_per_iteration": 2.661634683609009
    },
    {
      "auxiliary_loss_clip": 0.0131665,
      "auxiliary_loss_mlp": 0.01055788,
      "balance_loss_clip": 1.09345555,
      "balance_loss_mlp": 1.03688157,
      "epoch": 0.038116996332591836,
      "flos": 27709391790720.0,
      "grad_norm": 2.1321205797872653,
      "language_loss": 0.83624756,
      "learning_rate": 3.9993190609038994e-06,
      "loss": 0.859972,
      "num_input_tokens_seen": 6651505,
      "step": 317,
      "time_per_iteration": 2.544438123703003
    },
    {
      "auxiliary_loss_clip": 0.01322737,
      "auxiliary_loss_mlp": 0.01056027,
      "balance_loss_clip": 1.09545267,
      "balance_loss_mlp": 1.03833592,
      "epoch": 0.038237239223230926,
      "flos": 21178067011200.0,
      "grad_norm": 1.971339206957877,
      "language_loss": 0.82996392,
      "learning_rate": 3.999298583870182e-06,
      "loss": 0.85375166,
      "num_input_tokens_seen": 6671090,
      "step": 318,
      "time_per_iteration": 2.4800949096679688
    },
    {
      "auxiliary_loss_clip": 0.01319622,
      "auxiliary_loss_mlp": 0.01056433,
      "balance_loss_clip": 1.09480166,
      "balance_loss_mlp": 1.03696573,
      "epoch": 0.038357482113870016,
      "flos": 25556618995200.0,
      "grad_norm": 2.0636028903430597,
      "language_loss": 0.77382755,
      "learning_rate": 3.999277803544458e-06,
      "loss": 0.79758811,
      "num_input_tokens_seen": 6691245,
      "step": 319,
      "time_per_iteration": 2.537961006164551
    },
    {
      "auxiliary_loss_clip": 0.01250718,
      "auxiliary_loss_mlp": 0.01009237,
      "balance_loss_clip": 1.09175301,
      "balance_loss_mlp": 1.00356317,
      "epoch": 0.038477725004509106,
      "flos": 59227578034560.0,
      "grad_norm": 0.9611590820729975,
      "language_loss": 0.62415749,
      "learning_rate": 3.999256719929882e-06,
      "loss": 0.64675707,
      "num_input_tokens_seen": 6752520,
      "step": 320,
      "time_per_iteration": 3.0490732192993164
    },
    {
      "auxiliary_loss_clip": 0.0124932,
      "auxiliary_loss_mlp": 0.01007614,
      "balance_loss_clip": 1.09122705,
      "balance_loss_mlp": 1.00205886,
      "epoch": 0.0385979678951482,
      "flos": 67317676398720.0,
      "grad_norm": 1.2252363518170464,
      "language_loss": 0.67075634,
      "learning_rate": 3.999235333029651e-06,
      "loss": 0.69332576,
      "num_input_tokens_seen": 6806460,
      "step": 321,
      "time_per_iteration": 3.010756015777588
    },
    {
      "auxiliary_loss_clip": 0.0132267,
      "auxiliary_loss_mlp": 0.01064827,
      "balance_loss_clip": 1.0976516,
      "balance_loss_mlp": 1.0470053,
      "epoch": 0.03871821078578729,
      "flos": 22746752749440.0,
      "grad_norm": 3.9328536169429773,
      "language_loss": 0.81840611,
      "learning_rate": 3.999213642847009e-06,
      "loss": 0.84228116,
      "num_input_tokens_seen": 6827045,
      "step": 322,
      "time_per_iteration": 2.5040204524993896
    },
    {
      "auxiliary_loss_clip": 0.01323531,
      "auxiliary_loss_mlp": 0.01056065,
      "balance_loss_clip": 1.09799361,
      "balance_loss_mlp": 1.039042,
      "epoch": 0.03883845367642638,
      "flos": 26280613526400.0,
      "grad_norm": 2.593136662808657,
      "language_loss": 0.91238958,
      "learning_rate": 3.999191649385247e-06,
      "loss": 0.93618548,
      "num_input_tokens_seen": 6848220,
      "step": 323,
      "time_per_iteration": 2.5659687519073486
    },
    {
      "auxiliary_loss_clip": 0.01243863,
      "auxiliary_loss_mlp": 0.01005526,
      "balance_loss_clip": 1.08705044,
      "balance_loss_mlp": 0.99978012,
      "epoch": 0.03895869656706547,
      "flos": 56962835568000.0,
      "grad_norm": 0.9106112968757789,
      "language_loss": 0.59752786,
      "learning_rate": 3.999169352647702e-06,
      "loss": 0.6200217,
      "num_input_tokens_seen": 6909400,
      "step": 324,
      "time_per_iteration": 3.027989625930786
    },
    {
      "auxiliary_loss_clip": 0.01324365,
      "auxiliary_loss_mlp": 0.01058075,
      "balance_loss_clip": 1.09721637,
      "balance_loss_mlp": 1.03673673,
      "epoch": 0.03907893945770456,
      "flos": 24863363527680.0,
      "grad_norm": 2.215747671991251,
      "language_loss": 0.82897913,
      "learning_rate": 3.999146752637755e-06,
      "loss": 0.85280353,
      "num_input_tokens_seen": 6930445,
      "step": 325,
      "time_per_iteration": 2.54158353805542
    },
    {
      "auxiliary_loss_clip": 0.01320857,
      "auxiliary_loss_mlp": 0.01056523,
      "balance_loss_clip": 1.09448624,
      "balance_loss_mlp": 1.03749704,
      "epoch": 0.03919918234834365,
      "flos": 18368595815040.0,
      "grad_norm": 9.535906088470048,
      "language_loss": 0.89558804,
      "learning_rate": 3.999123849358836e-06,
      "loss": 0.91936189,
      "num_input_tokens_seen": 6948110,
      "step": 326,
      "time_per_iteration": 2.5016963481903076
    },
    {
      "auxiliary_loss_clip": 0.01321184,
      "auxiliary_loss_mlp": 0.0106315,
      "balance_loss_clip": 1.09424639,
      "balance_loss_mlp": 1.04456556,
      "epoch": 0.03931942523898275,
      "flos": 25225414663680.0,
      "grad_norm": 2.2276459680669984,
      "language_loss": 0.74881506,
      "learning_rate": 3.999100642814418e-06,
      "loss": 0.77265841,
      "num_input_tokens_seen": 6968550,
      "step": 327,
      "time_per_iteration": 2.5203912258148193
    },
    {
      "auxiliary_loss_clip": 0.01317992,
      "auxiliary_loss_mlp": 0.01053591,
      "balance_loss_clip": 1.09574878,
      "balance_loss_mlp": 1.0359596,
      "epoch": 0.03943966812962184,
      "flos": 23257905240960.0,
      "grad_norm": 2.798006152409529,
      "language_loss": 0.8818835,
      "learning_rate": 3.999077133008022e-06,
      "loss": 0.90559924,
      "num_input_tokens_seen": 6987135,
      "step": 328,
      "time_per_iteration": 2.5406737327575684
    },
    {
      "auxiliary_loss_clip": 0.01320292,
      "auxiliary_loss_mlp": 0.01045046,
      "balance_loss_clip": 1.09754384,
      "balance_loss_mlp": 1.02598429,
      "epoch": 0.03955991102026093,
      "flos": 29168837291520.0,
      "grad_norm": 2.0287829755604294,
      "language_loss": 0.90715504,
      "learning_rate": 3.9990533199432145e-06,
      "loss": 0.93080842,
      "num_input_tokens_seen": 7008630,
      "step": 329,
      "time_per_iteration": 3.334651470184326
    },
    {
      "auxiliary_loss_clip": 0.01320126,
      "auxiliary_loss_mlp": 0.01046323,
      "balance_loss_clip": 1.09809017,
      "balance_loss_mlp": 1.02662969,
      "epoch": 0.03968015391090002,
      "flos": 17602441695360.0,
      "grad_norm": 2.4147811515091084,
      "language_loss": 0.75663054,
      "learning_rate": 3.999029203623608e-06,
      "loss": 0.78029501,
      "num_input_tokens_seen": 7026350,
      "step": 330,
      "time_per_iteration": 2.502889633178711
    },
    {
      "auxiliary_loss_clip": 0.0131987,
      "auxiliary_loss_mlp": 0.01051036,
      "balance_loss_clip": 1.09565091,
      "balance_loss_mlp": 1.03246319,
      "epoch": 0.03980039680153911,
      "flos": 21799285752960.0,
      "grad_norm": 2.125489175277354,
      "language_loss": 0.87222475,
      "learning_rate": 3.99900478405286e-06,
      "loss": 0.89593381,
      "num_input_tokens_seen": 7045660,
      "step": 331,
      "time_per_iteration": 3.3017563819885254
    },
    {
      "auxiliary_loss_clip": 0.01320358,
      "auxiliary_loss_mlp": 0.01050434,
      "balance_loss_clip": 1.09869528,
      "balance_loss_mlp": 1.03342307,
      "epoch": 0.0399206396921782,
      "flos": 15195134148480.0,
      "grad_norm": 2.224013142276502,
      "language_loss": 0.82468772,
      "learning_rate": 3.998980061234676e-06,
      "loss": 0.84839571,
      "num_input_tokens_seen": 7063575,
      "step": 332,
      "time_per_iteration": 2.49629282951355
    },
    {
      "auxiliary_loss_clip": 0.01321977,
      "auxiliary_loss_mlp": 0.01052284,
      "balance_loss_clip": 1.09872448,
      "balance_loss_mlp": 1.0334847,
      "epoch": 0.040040882582817294,
      "flos": 14422910630400.0,
      "grad_norm": 3.012408306174759,
      "language_loss": 0.75591087,
      "learning_rate": 3.9989550351728055e-06,
      "loss": 0.77965343,
      "num_input_tokens_seen": 7080505,
      "step": 333,
      "time_per_iteration": 2.483832359313965
    },
    {
      "auxiliary_loss_clip": 0.01318211,
      "auxiliary_loss_mlp": 0.01052984,
      "balance_loss_clip": 1.09577763,
      "balance_loss_mlp": 1.03512633,
      "epoch": 0.040161125473456384,
      "flos": 19280906375040.0,
      "grad_norm": 2.509483066879563,
      "language_loss": 0.84788758,
      "learning_rate": 3.998929705871046e-06,
      "loss": 0.87159956,
      "num_input_tokens_seen": 7097860,
      "step": 334,
      "time_per_iteration": 3.23042368888855
    },
    {
      "auxiliary_loss_clip": 0.01318625,
      "auxiliary_loss_mlp": 0.01057031,
      "balance_loss_clip": 1.09727407,
      "balance_loss_mlp": 1.03918481,
      "epoch": 0.040281368364095474,
      "flos": 17821101738240.0,
      "grad_norm": 2.7334646361500803,
      "language_loss": 0.88855875,
      "learning_rate": 3.99890407333324e-06,
      "loss": 0.91231525,
      "num_input_tokens_seen": 7116390,
      "step": 335,
      "time_per_iteration": 2.466595411300659
    },
    {
      "auxiliary_loss_clip": 0.01315015,
      "auxiliary_loss_mlp": 0.01056899,
      "balance_loss_clip": 1.09242332,
      "balance_loss_mlp": 1.03949499,
      "epoch": 0.040401611254734564,
      "flos": 19573757959680.0,
      "grad_norm": 1.79871489538905,
      "language_loss": 0.87017751,
      "learning_rate": 3.998878137563275e-06,
      "loss": 0.8938967,
      "num_input_tokens_seen": 7135940,
      "step": 336,
      "time_per_iteration": 2.473306179046631
    },
    {
      "auxiliary_loss_clip": 0.01314724,
      "auxiliary_loss_mlp": 0.01054928,
      "balance_loss_clip": 1.09319174,
      "balance_loss_mlp": 1.03637934,
      "epoch": 0.040521854145373654,
      "flos": 22054466949120.0,
      "grad_norm": 2.705836103907685,
      "language_loss": 0.85163975,
      "learning_rate": 3.998851898565085e-06,
      "loss": 0.87533635,
      "num_input_tokens_seen": 7155745,
      "step": 337,
      "time_per_iteration": 2.510936975479126
    },
    {
      "auxiliary_loss_clip": 0.01314148,
      "auxiliary_loss_mlp": 0.01060585,
      "balance_loss_clip": 1.09421158,
      "balance_loss_mlp": 1.04337144,
      "epoch": 0.04064209703601274,
      "flos": 22674644196480.0,
      "grad_norm": 2.717042403168639,
      "language_loss": 0.83184516,
      "learning_rate": 3.998825356342653e-06,
      "loss": 0.85559249,
      "num_input_tokens_seen": 7175920,
      "step": 338,
      "time_per_iteration": 2.4971539974212646
    },
    {
      "auxiliary_loss_clip": 0.01318205,
      "auxiliary_loss_mlp": 0.01048038,
      "balance_loss_clip": 1.09433258,
      "balance_loss_mlp": 1.0307169,
      "epoch": 0.04076233992665183,
      "flos": 38582172783360.0,
      "grad_norm": 2.297543267957964,
      "language_loss": 0.72833246,
      "learning_rate": 3.998798510900003e-06,
      "loss": 0.75199491,
      "num_input_tokens_seen": 7198720,
      "step": 339,
      "time_per_iteration": 2.6401944160461426
    },
    {
      "auxiliary_loss_clip": 0.01311722,
      "auxiliary_loss_mlp": 0.01058406,
      "balance_loss_clip": 1.09139347,
      "balance_loss_mlp": 1.03954685,
      "epoch": 0.04088258281729093,
      "flos": 25885309374720.0,
      "grad_norm": 2.6383081505592227,
      "language_loss": 0.83955556,
      "learning_rate": 3.998771362241207e-06,
      "loss": 0.86325681,
      "num_input_tokens_seen": 7219125,
      "step": 340,
      "time_per_iteration": 2.516371488571167
    },
    {
      "auxiliary_loss_clip": 0.0131011,
      "auxiliary_loss_mlp": 0.01049691,
      "balance_loss_clip": 1.09263408,
      "balance_loss_mlp": 1.03263211,
      "epoch": 0.04100282570793002,
      "flos": 19789832223360.0,
      "grad_norm": 2.4557662761924473,
      "language_loss": 0.87873179,
      "learning_rate": 3.998743910370385e-06,
      "loss": 0.9023298,
      "num_input_tokens_seen": 7237985,
      "step": 341,
      "time_per_iteration": 2.499544620513916
    },
    {
      "auxiliary_loss_clip": 0.01318989,
      "auxiliary_loss_mlp": 0.01058707,
      "balance_loss_clip": 1.10081315,
      "balance_loss_mlp": 1.04051626,
      "epoch": 0.04112306859856911,
      "flos": 22565152563840.0,
      "grad_norm": 2.960118318886243,
      "language_loss": 0.73116267,
      "learning_rate": 3.998716155291702e-06,
      "loss": 0.75493968,
      "num_input_tokens_seen": 7255825,
      "step": 342,
      "time_per_iteration": 2.4998021125793457
    },
    {
      "auxiliary_loss_clip": 0.01316767,
      "auxiliary_loss_mlp": 0.01054697,
      "balance_loss_clip": 1.09773159,
      "balance_loss_mlp": 1.03707814,
      "epoch": 0.0412433114892082,
      "flos": 25040654081280.0,
      "grad_norm": 1.6793288395944508,
      "language_loss": 0.90485191,
      "learning_rate": 3.998688097009366e-06,
      "loss": 0.92856658,
      "num_input_tokens_seen": 7276590,
      "step": 343,
      "time_per_iteration": 2.5319066047668457
    },
    {
      "auxiliary_loss_clip": 0.01312179,
      "auxiliary_loss_mlp": 0.01049075,
      "balance_loss_clip": 1.09442902,
      "balance_loss_mlp": 1.03244543,
      "epoch": 0.04136355437984729,
      "flos": 25191371548800.0,
      "grad_norm": 2.1352361229342747,
      "language_loss": 0.80126566,
      "learning_rate": 3.998659735527636e-06,
      "loss": 0.82487822,
      "num_input_tokens_seen": 7295680,
      "step": 344,
      "time_per_iteration": 2.5061516761779785
    },
    {
      "auxiliary_loss_clip": 0.01312049,
      "auxiliary_loss_mlp": 0.01046783,
      "balance_loss_clip": 1.09247053,
      "balance_loss_mlp": 1.02916408,
      "epoch": 0.04148379727048638,
      "flos": 22966777509120.0,
      "grad_norm": 2.264932707183971,
      "language_loss": 0.78225362,
      "learning_rate": 3.998631070850813e-06,
      "loss": 0.80584192,
      "num_input_tokens_seen": 7316300,
      "step": 345,
      "time_per_iteration": 2.5206079483032227
    },
    {
      "auxiliary_loss_clip": 0.01313467,
      "auxiliary_loss_mlp": 0.01049097,
      "balance_loss_clip": 1.09682095,
      "balance_loss_mlp": 1.03225279,
      "epoch": 0.041604040161125476,
      "flos": 14063481187200.0,
      "grad_norm": 2.3119258276885204,
      "language_loss": 0.8348726,
      "learning_rate": 3.9986021029832455e-06,
      "loss": 0.85849822,
      "num_input_tokens_seen": 7333615,
      "step": 346,
      "time_per_iteration": 2.453634262084961
    },
    {
      "auxiliary_loss_clip": 0.01314256,
      "auxiliary_loss_mlp": 0.01062258,
      "balance_loss_clip": 1.0939163,
      "balance_loss_mlp": 1.04245758,
      "epoch": 0.041724283051764566,
      "flos": 12091877614080.0,
      "grad_norm": 3.54593894740408,
      "language_loss": 0.9167006,
      "learning_rate": 3.9985728319293285e-06,
      "loss": 0.94046575,
      "num_input_tokens_seen": 7347590,
      "step": 347,
      "time_per_iteration": 2.4561381340026855
    },
    {
      "auxiliary_loss_clip": 0.01316415,
      "auxiliary_loss_mlp": 0.0105213,
      "balance_loss_clip": 1.09203684,
      "balance_loss_mlp": 1.03279448,
      "epoch": 0.041844525942403656,
      "flos": 12385303816320.0,
      "grad_norm": 2.4580124357932682,
      "language_loss": 0.85185802,
      "learning_rate": 3.998543257693501e-06,
      "loss": 0.87554353,
      "num_input_tokens_seen": 7364345,
      "step": 348,
      "time_per_iteration": 2.4632930755615234
    },
    {
      "auxiliary_loss_clip": 0.01314244,
      "auxiliary_loss_mlp": 0.01053648,
      "balance_loss_clip": 1.09558165,
      "balance_loss_mlp": 1.03663635,
      "epoch": 0.041964768833042745,
      "flos": 23769345041280.0,
      "grad_norm": 1.715792883834112,
      "language_loss": 0.87710202,
      "learning_rate": 3.998513380280251e-06,
      "loss": 0.90078098,
      "num_input_tokens_seen": 7384625,
      "step": 349,
      "time_per_iteration": 2.563983917236328
    },
    {
      "auxiliary_loss_clip": 0.01315056,
      "auxiliary_loss_mlp": 0.01050266,
      "balance_loss_clip": 1.09474993,
      "balance_loss_mlp": 1.03163385,
      "epoch": 0.042085011723681835,
      "flos": 11875336473600.0,
      "grad_norm": 3.101076320724572,
      "language_loss": 0.95042682,
      "learning_rate": 3.99848319969411e-06,
      "loss": 0.97408009,
      "num_input_tokens_seen": 7402225,
      "step": 350,
      "time_per_iteration": 2.4653100967407227
    },
    {
      "auxiliary_loss_clip": 0.01317591,
      "auxiliary_loss_mlp": 0.01067174,
      "balance_loss_clip": 1.09739757,
      "balance_loss_mlp": 1.04886329,
      "epoch": 0.042205254614320925,
      "flos": 16873957964160.0,
      "grad_norm": 2.426386542389784,
      "language_loss": 0.79001719,
      "learning_rate": 3.9984527159396564e-06,
      "loss": 0.81386489,
      "num_input_tokens_seen": 7420865,
      "step": 351,
      "time_per_iteration": 2.481995105743408
    },
    {
      "auxiliary_loss_clip": 0.01311307,
      "auxiliary_loss_mlp": 0.01056802,
      "balance_loss_clip": 1.09166586,
      "balance_loss_mlp": 1.03807449,
      "epoch": 0.04232549750496002,
      "flos": 25118508810240.0,
      "grad_norm": 2.4100889979099023,
      "language_loss": 0.84686339,
      "learning_rate": 3.9984219290215154e-06,
      "loss": 0.87054449,
      "num_input_tokens_seen": 7441040,
      "step": 352,
      "time_per_iteration": 2.5281734466552734
    },
    {
      "auxiliary_loss_clip": 0.01310338,
      "auxiliary_loss_mlp": 0.01054344,
      "balance_loss_clip": 1.09387255,
      "balance_loss_mlp": 1.03806043,
      "epoch": 0.04244574039559911,
      "flos": 26724541714560.0,
      "grad_norm": 1.6101034169642747,
      "language_loss": 0.89167005,
      "learning_rate": 3.998390838944356e-06,
      "loss": 0.91531688,
      "num_input_tokens_seen": 7462545,
      "step": 353,
      "time_per_iteration": 2.5560574531555176
    },
    {
      "auxiliary_loss_clip": 0.01312276,
      "auxiliary_loss_mlp": 0.01053551,
      "balance_loss_clip": 1.09377122,
      "balance_loss_mlp": 1.03611088,
      "epoch": 0.0425659832862382,
      "flos": 20923244951040.0,
      "grad_norm": 2.6106992183460367,
      "language_loss": 0.90379965,
      "learning_rate": 3.998359445712895e-06,
      "loss": 0.92745793,
      "num_input_tokens_seen": 7481650,
      "step": 354,
      "time_per_iteration": 2.508089065551758
    },
    {
      "auxiliary_loss_clip": 0.01310295,
      "auxiliary_loss_mlp": 0.01053494,
      "balance_loss_clip": 1.09203768,
      "balance_loss_mlp": 1.03524303,
      "epoch": 0.04268622617687729,
      "flos": 23331127115520.0,
      "grad_norm": 3.0945863033429366,
      "language_loss": 0.81077528,
      "learning_rate": 3.9983277493318955e-06,
      "loss": 0.83441323,
      "num_input_tokens_seen": 7500945,
      "step": 355,
      "time_per_iteration": 2.51237416267395
    },
    {
      "auxiliary_loss_clip": 0.01313877,
      "auxiliary_loss_mlp": 0.01054756,
      "balance_loss_clip": 1.09328079,
      "balance_loss_mlp": 1.03714931,
      "epoch": 0.04280646906751638,
      "flos": 25994010908160.0,
      "grad_norm": 1.5905860508273464,
      "language_loss": 0.81051123,
      "learning_rate": 3.998295749806165e-06,
      "loss": 0.83419752,
      "num_input_tokens_seen": 7522170,
      "step": 356,
      "time_per_iteration": 2.5298011302948
    },
    {
      "auxiliary_loss_clip": 0.01312758,
      "auxiliary_loss_mlp": 0.01057674,
      "balance_loss_clip": 1.09669018,
      "balance_loss_mlp": 1.04112744,
      "epoch": 0.04292671195815547,
      "flos": 26906824258560.0,
      "grad_norm": 1.9805784724207547,
      "language_loss": 0.8328504,
      "learning_rate": 3.998263447140558e-06,
      "loss": 0.85655475,
      "num_input_tokens_seen": 7542370,
      "step": 357,
      "time_per_iteration": 4.045665979385376
    },
    {
      "auxiliary_loss_clip": 0.01306777,
      "auxiliary_loss_mlp": 0.01048548,
      "balance_loss_clip": 1.08943486,
      "balance_loss_mlp": 1.03115559,
      "epoch": 0.04304695484879457,
      "flos": 39457315745280.0,
      "grad_norm": 1.9914183702951826,
      "language_loss": 0.81793469,
      "learning_rate": 3.998230841339976e-06,
      "loss": 0.84148794,
      "num_input_tokens_seen": 7564380,
      "step": 358,
      "time_per_iteration": 3.3478591442108154
    },
    {
      "auxiliary_loss_clip": 0.01310945,
      "auxiliary_loss_mlp": 0.01051068,
      "balance_loss_clip": 1.09409976,
      "balance_loss_mlp": 1.03459287,
      "epoch": 0.04316719773943366,
      "flos": 19646297475840.0,
      "grad_norm": 2.228356847729235,
      "language_loss": 0.85161239,
      "learning_rate": 3.998197932409363e-06,
      "loss": 0.87523252,
      "num_input_tokens_seen": 7582390,
      "step": 359,
      "time_per_iteration": 2.4723424911499023
    },
    {
      "auxiliary_loss_clip": 0.01308333,
      "auxiliary_loss_mlp": 0.01054212,
      "balance_loss_clip": 1.09359908,
      "balance_loss_mlp": 1.03768992,
      "epoch": 0.04328744063007275,
      "flos": 22452320966400.0,
      "grad_norm": 2.224291678477611,
      "language_loss": 0.86304688,
      "learning_rate": 3.9981647203537125e-06,
      "loss": 0.88667238,
      "num_input_tokens_seen": 7599890,
      "step": 360,
      "time_per_iteration": 3.2452163696289062
    },
    {
      "auxiliary_loss_clip": 0.01313542,
      "auxiliary_loss_mlp": 0.01048433,
      "balance_loss_clip": 1.09399867,
      "balance_loss_mlp": 1.03180289,
      "epoch": 0.04340768352071184,
      "flos": 21283033530240.0,
      "grad_norm": 2.474404106925693,
      "language_loss": 0.95925772,
      "learning_rate": 3.998131205178063e-06,
      "loss": 0.98287755,
      "num_input_tokens_seen": 7618360,
      "step": 361,
      "time_per_iteration": 2.4729456901550293
    },
    {
      "auxiliary_loss_clip": 0.01312222,
      "auxiliary_loss_mlp": 0.01053626,
      "balance_loss_clip": 1.0933435,
      "balance_loss_mlp": 1.0356257,
      "epoch": 0.04352792641135093,
      "flos": 11583705951360.0,
      "grad_norm": 3.322659308441934,
      "language_loss": 0.7678467,
      "learning_rate": 3.998097386887498e-06,
      "loss": 0.79150516,
      "num_input_tokens_seen": 7635435,
      "step": 362,
      "time_per_iteration": 2.4794559478759766
    },
    {
      "auxiliary_loss_clip": 0.0130634,
      "auxiliary_loss_mlp": 0.01050182,
      "balance_loss_clip": 1.09140229,
      "balance_loss_mlp": 1.03448248,
      "epoch": 0.04364816930199002,
      "flos": 23623547736960.0,
      "grad_norm": 1.828294003642291,
      "language_loss": 0.84879386,
      "learning_rate": 3.998063265487148e-06,
      "loss": 0.8723591,
      "num_input_tokens_seen": 7656485,
      "step": 363,
      "time_per_iteration": 2.5058865547180176
    },
    {
      "auxiliary_loss_clip": 0.01311384,
      "auxiliary_loss_mlp": 0.01049902,
      "balance_loss_clip": 1.09295678,
      "balance_loss_mlp": 1.03328466,
      "epoch": 0.043768412192629114,
      "flos": 14429734214400.0,
      "grad_norm": 2.8162056417414236,
      "language_loss": 0.81129456,
      "learning_rate": 3.99802884098219e-06,
      "loss": 0.83490741,
      "num_input_tokens_seen": 7674595,
      "step": 364,
      "time_per_iteration": 2.483816146850586
    },
    {
      "auxiliary_loss_clip": 0.01308282,
      "auxiliary_loss_mlp": 0.01053006,
      "balance_loss_clip": 1.09013879,
      "balance_loss_mlp": 1.03558969,
      "epoch": 0.043888655083268203,
      "flos": 26468893641600.0,
      "grad_norm": 2.7616814730510035,
      "language_loss": 0.82206571,
      "learning_rate": 3.997994113377845e-06,
      "loss": 0.84567863,
      "num_input_tokens_seen": 7693495,
      "step": 365,
      "time_per_iteration": 2.50985050201416
    },
    {
      "auxiliary_loss_clip": 0.01311899,
      "auxiliary_loss_mlp": 0.01058431,
      "balance_loss_clip": 1.09298277,
      "balance_loss_mlp": 1.04095531,
      "epoch": 0.04400889797390729,
      "flos": 27235263242880.0,
      "grad_norm": 2.0304573871313276,
      "language_loss": 0.83474493,
      "learning_rate": 3.9979590826793815e-06,
      "loss": 0.85844827,
      "num_input_tokens_seen": 7714685,
      "step": 366,
      "time_per_iteration": 2.5782947540283203
    },
    {
      "auxiliary_loss_clip": 0.01310411,
      "auxiliary_loss_mlp": 0.01047939,
      "balance_loss_clip": 1.09197485,
      "balance_loss_mlp": 1.03030753,
      "epoch": 0.04412914086454638,
      "flos": 20119528183680.0,
      "grad_norm": 3.395786684810804,
      "language_loss": 0.80999064,
      "learning_rate": 3.997923748892113e-06,
      "loss": 0.83357418,
      "num_input_tokens_seen": 7734005,
      "step": 367,
      "time_per_iteration": 2.477569341659546
    },
    {
      "auxiliary_loss_clip": 0.01309588,
      "auxiliary_loss_mlp": 0.01054376,
      "balance_loss_clip": 1.0953505,
      "balance_loss_mlp": 1.03867602,
      "epoch": 0.04424938375518547,
      "flos": 22604618632320.0,
      "grad_norm": 1.975378697192492,
      "language_loss": 0.88524801,
      "learning_rate": 3.9978881120214015e-06,
      "loss": 0.90888762,
      "num_input_tokens_seen": 7755525,
      "step": 368,
      "time_per_iteration": 2.5497565269470215
    },
    {
      "auxiliary_loss_clip": 0.0130841,
      "auxiliary_loss_mlp": 0.01050714,
      "balance_loss_clip": 1.09018385,
      "balance_loss_mlp": 1.03214109,
      "epoch": 0.04436962664582456,
      "flos": 24132365844480.0,
      "grad_norm": 2.3851077610681517,
      "language_loss": 0.7948674,
      "learning_rate": 3.997852172072652e-06,
      "loss": 0.81845862,
      "num_input_tokens_seen": 7776740,
      "step": 369,
      "time_per_iteration": 2.508631467819214
    },
    {
      "auxiliary_loss_clip": 0.01312293,
      "auxiliary_loss_mlp": 0.01051524,
      "balance_loss_clip": 1.09337115,
      "balance_loss_mlp": 1.03177094,
      "epoch": 0.04448986953646366,
      "flos": 18222906251520.0,
      "grad_norm": 3.756288889602186,
      "language_loss": 0.89277613,
      "learning_rate": 3.9978159290513155e-06,
      "loss": 0.91641426,
      "num_input_tokens_seen": 7794820,
      "step": 370,
      "time_per_iteration": 2.4808430671691895
    },
    {
      "auxiliary_loss_clip": 0.01314554,
      "auxiliary_loss_mlp": 0.0104964,
      "balance_loss_clip": 1.09474969,
      "balance_loss_mlp": 1.03130603,
      "epoch": 0.04461011242710275,
      "flos": 30117920400000.0,
      "grad_norm": 2.3562408256778666,
      "language_loss": 0.80191511,
      "learning_rate": 3.997779382962892e-06,
      "loss": 0.82555705,
      "num_input_tokens_seen": 7817705,
      "step": 371,
      "time_per_iteration": 2.5633630752563477
    },
    {
      "auxiliary_loss_clip": 0.01308151,
      "auxiliary_loss_mlp": 0.01049616,
      "balance_loss_clip": 1.09265888,
      "balance_loss_mlp": 1.03347468,
      "epoch": 0.04473035531774184,
      "flos": 29752529299200.0,
      "grad_norm": 1.977110167086918,
      "language_loss": 0.73805761,
      "learning_rate": 3.997742533812924e-06,
      "loss": 0.7616353,
      "num_input_tokens_seen": 7840970,
      "step": 372,
      "time_per_iteration": 2.5842678546905518
    },
    {
      "auxiliary_loss_clip": 0.01308302,
      "auxiliary_loss_mlp": 0.01057566,
      "balance_loss_clip": 1.09297359,
      "balance_loss_mlp": 1.04141307,
      "epoch": 0.04485059820838093,
      "flos": 13151565676800.0,
      "grad_norm": 2.569260290899036,
      "language_loss": 0.92188001,
      "learning_rate": 3.997705381607001e-06,
      "loss": 0.9455387,
      "num_input_tokens_seen": 7857785,
      "step": 373,
      "time_per_iteration": 2.4519026279449463
    },
    {
      "auxiliary_loss_clip": 0.01238268,
      "auxiliary_loss_mlp": 0.01006258,
      "balance_loss_clip": 1.08658481,
      "balance_loss_mlp": 1.00120342,
      "epoch": 0.04497084109902002,
      "flos": 68094209548800.0,
      "grad_norm": 0.9785359393505195,
      "language_loss": 0.60243452,
      "learning_rate": 3.997667926350761e-06,
      "loss": 0.62487978,
      "num_input_tokens_seen": 7916115,
      "step": 374,
      "time_per_iteration": 3.023674964904785
    },
    {
      "auxiliary_loss_clip": 0.012366,
      "auxiliary_loss_mlp": 0.0100547,
      "balance_loss_clip": 1.08558261,
      "balance_loss_mlp": 1.00055814,
      "epoch": 0.04509108398965911,
      "flos": 64342263346560.0,
      "grad_norm": 0.9045002647575208,
      "language_loss": 0.5779137,
      "learning_rate": 3.997630168049886e-06,
      "loss": 0.60033441,
      "num_input_tokens_seen": 7974480,
      "step": 375,
      "time_per_iteration": 3.0997185707092285
    },
    {
      "auxiliary_loss_clip": 0.01308322,
      "auxiliary_loss_mlp": 0.01051021,
      "balance_loss_clip": 1.09149265,
      "balance_loss_mlp": 1.03275836,
      "epoch": 0.045211326880298205,
      "flos": 22271115830400.0,
      "grad_norm": 2.2951793618715373,
      "language_loss": 0.77201092,
      "learning_rate": 3.997592106710101e-06,
      "loss": 0.79560435,
      "num_input_tokens_seen": 7993940,
      "step": 376,
      "time_per_iteration": 2.503448247909546
    },
    {
      "auxiliary_loss_clip": 0.01304344,
      "auxiliary_loss_mlp": 0.01050865,
      "balance_loss_clip": 1.08921754,
      "balance_loss_mlp": 1.03443825,
      "epoch": 0.045331569770937295,
      "flos": 32159441796480.0,
      "grad_norm": 2.330498549661473,
      "language_loss": 0.65711224,
      "learning_rate": 3.997553742337182e-06,
      "loss": 0.6806643,
      "num_input_tokens_seen": 8013365,
      "step": 377,
      "time_per_iteration": 2.5934371948242188
    },
    {
      "auxiliary_loss_clip": 0.01307314,
      "auxiliary_loss_mlp": 0.01050952,
      "balance_loss_clip": 1.09164011,
      "balance_loss_mlp": 1.03456104,
      "epoch": 0.045451812661576385,
      "flos": 22163455791360.0,
      "grad_norm": 1.7205129515694528,
      "language_loss": 0.91449189,
      "learning_rate": 3.997515074936949e-06,
      "loss": 0.93807459,
      "num_input_tokens_seen": 8034240,
      "step": 378,
      "time_per_iteration": 2.5195086002349854
    },
    {
      "auxiliary_loss_clip": 0.01306059,
      "auxiliary_loss_mlp": 0.01041449,
      "balance_loss_clip": 1.09126401,
      "balance_loss_mlp": 1.02602303,
      "epoch": 0.045572055552215475,
      "flos": 16581968305920.0,
      "grad_norm": 3.204897787339293,
      "language_loss": 0.87067598,
      "learning_rate": 3.997476104515268e-06,
      "loss": 0.89415109,
      "num_input_tokens_seen": 8052430,
      "step": 379,
      "time_per_iteration": 2.477285623550415
    },
    {
      "auxiliary_loss_clip": 0.01303911,
      "auxiliary_loss_mlp": 0.01051316,
      "balance_loss_clip": 1.09351289,
      "balance_loss_mlp": 1.03591394,
      "epoch": 0.045692298442854565,
      "flos": 17603375448960.0,
      "grad_norm": 2.649375368921909,
      "language_loss": 0.77656972,
      "learning_rate": 3.9974368310780485e-06,
      "loss": 0.80012202,
      "num_input_tokens_seen": 8069605,
      "step": 380,
      "time_per_iteration": 2.497349739074707
    },
    {
      "auxiliary_loss_clip": 0.0131565,
      "auxiliary_loss_mlp": 0.0105539,
      "balance_loss_clip": 1.09548306,
      "balance_loss_mlp": 1.03625655,
      "epoch": 0.045812541333493655,
      "flos": 26761098781440.0,
      "grad_norm": 2.6309445366576174,
      "language_loss": 0.74487996,
      "learning_rate": 3.997397254631251e-06,
      "loss": 0.76859033,
      "num_input_tokens_seen": 8090225,
      "step": 381,
      "time_per_iteration": 2.5604755878448486
    },
    {
      "auxiliary_loss_clip": 0.01225282,
      "auxiliary_loss_mlp": 0.01006215,
      "balance_loss_clip": 1.07678473,
      "balance_loss_mlp": 1.0011363,
      "epoch": 0.04593278422413275,
      "flos": 60250349894400.0,
      "grad_norm": 0.8221900557880444,
      "language_loss": 0.60065883,
      "learning_rate": 3.997357375180878e-06,
      "loss": 0.6229738,
      "num_input_tokens_seen": 8154505,
      "step": 382,
      "time_per_iteration": 3.1888010501861572
    },
    {
      "auxiliary_loss_clip": 0.01306088,
      "auxiliary_loss_mlp": 0.01042372,
      "balance_loss_clip": 1.09097886,
      "balance_loss_mlp": 1.02581382,
      "epoch": 0.04605302711477184,
      "flos": 21799249839360.0,
      "grad_norm": 2.326936222237418,
      "language_loss": 0.75258505,
      "learning_rate": 3.997317192732979e-06,
      "loss": 0.77606964,
      "num_input_tokens_seen": 8173285,
      "step": 383,
      "time_per_iteration": 3.244495391845703
    },
    {
      "auxiliary_loss_clip": 0.01308136,
      "auxiliary_loss_mlp": 0.01051861,
      "balance_loss_clip": 1.0921762,
      "balance_loss_mlp": 1.03468287,
      "epoch": 0.04617327000541093,
      "flos": 19459705299840.0,
      "grad_norm": 2.020913573673132,
      "language_loss": 0.82658958,
      "learning_rate": 3.99727670729365e-06,
      "loss": 0.85018957,
      "num_input_tokens_seen": 8191845,
      "step": 384,
      "time_per_iteration": 3.2668025493621826
    },
    {
      "auxiliary_loss_clip": 0.01308043,
      "auxiliary_loss_mlp": 0.0105406,
      "balance_loss_clip": 1.0950706,
      "balance_loss_mlp": 1.03927827,
      "epoch": 0.04629351289605002,
      "flos": 25411468135680.0,
      "grad_norm": 1.904129561887046,
      "language_loss": 0.77818513,
      "learning_rate": 3.997235918869033e-06,
      "loss": 0.80180615,
      "num_input_tokens_seen": 8212880,
      "step": 385,
      "time_per_iteration": 3.2936720848083496
    },
    {
      "auxiliary_loss_clip": 0.01301935,
      "auxiliary_loss_mlp": 0.01055079,
      "balance_loss_clip": 1.0915072,
      "balance_loss_mlp": 1.0395937,
      "epoch": 0.04641375578668911,
      "flos": 20558284813440.0,
      "grad_norm": 1.9188090409482603,
      "language_loss": 0.8271237,
      "learning_rate": 3.997194827465315e-06,
      "loss": 0.85069382,
      "num_input_tokens_seen": 8231475,
      "step": 386,
      "time_per_iteration": 2.526918649673462
    },
    {
      "auxiliary_loss_clip": 0.01304356,
      "auxiliary_loss_mlp": 0.01055337,
      "balance_loss_clip": 1.09081197,
      "balance_loss_mlp": 1.03825474,
      "epoch": 0.0465339986773282,
      "flos": 13188661447680.0,
      "grad_norm": 2.806370337426351,
      "language_loss": 0.91858411,
      "learning_rate": 3.997153433088728e-06,
      "loss": 0.94218099,
      "num_input_tokens_seen": 8248600,
      "step": 387,
      "time_per_iteration": 3.193587303161621
    },
    {
      "auxiliary_loss_clip": 0.01305685,
      "auxiliary_loss_mlp": 0.01064565,
      "balance_loss_clip": 1.09153104,
      "balance_loss_mlp": 1.04655266,
      "epoch": 0.0466542415679673,
      "flos": 25556547168000.0,
      "grad_norm": 2.3150392597946174,
      "language_loss": 0.81256604,
      "learning_rate": 3.997111735745554e-06,
      "loss": 0.83626854,
      "num_input_tokens_seen": 8271570,
      "step": 388,
      "time_per_iteration": 2.5755271911621094
    },
    {
      "auxiliary_loss_clip": 0.01306455,
      "auxiliary_loss_mlp": 0.0105409,
      "balance_loss_clip": 1.09364772,
      "balance_loss_mlp": 1.03642333,
      "epoch": 0.04677448445860639,
      "flos": 22236749493120.0,
      "grad_norm": 2.174623217207601,
      "language_loss": 0.82467991,
      "learning_rate": 3.997069735442118e-06,
      "loss": 0.84828538,
      "num_input_tokens_seen": 8291265,
      "step": 389,
      "time_per_iteration": 2.5046701431274414
    },
    {
      "auxiliary_loss_clip": 0.01300039,
      "auxiliary_loss_mlp": 0.01047407,
      "balance_loss_clip": 1.08876348,
      "balance_loss_mlp": 1.03289962,
      "epoch": 0.04689472734924548,
      "flos": 28147825198080.0,
      "grad_norm": 1.820766076704463,
      "language_loss": 0.80258238,
      "learning_rate": 3.997027432184792e-06,
      "loss": 0.82605684,
      "num_input_tokens_seen": 8315925,
      "step": 390,
      "time_per_iteration": 2.650614023208618
    },
    {
      "auxiliary_loss_clip": 0.01304077,
      "auxiliary_loss_mlp": 0.01047201,
      "balance_loss_clip": 1.08999467,
      "balance_loss_mlp": 1.0305469,
      "epoch": 0.04701497023988457,
      "flos": 23148952312320.0,
      "grad_norm": 1.860870818974501,
      "language_loss": 0.89635158,
      "learning_rate": 3.99698482597999e-06,
      "loss": 0.9198643,
      "num_input_tokens_seen": 8333605,
      "step": 391,
      "time_per_iteration": 2.5036532878875732
    },
    {
      "auxiliary_loss_clip": 0.01214409,
      "auxiliary_loss_mlp": 0.01006155,
      "balance_loss_clip": 1.07003713,
      "balance_loss_mlp": 1.00150585,
      "epoch": 0.04713521313052366,
      "flos": 64827668764800.0,
      "grad_norm": 0.8663200256425004,
      "language_loss": 0.63894832,
      "learning_rate": 3.99694191683418e-06,
      "loss": 0.66115403,
      "num_input_tokens_seen": 8394405,
      "step": 392,
      "time_per_iteration": 3.1296775341033936
    },
    {
      "auxiliary_loss_clip": 0.01304835,
      "auxiliary_loss_mlp": 0.01060348,
      "balance_loss_clip": 1.0939672,
      "balance_loss_mlp": 1.04332495,
      "epoch": 0.047255456021162746,
      "flos": 18771585477120.0,
      "grad_norm": 2.1932848017695954,
      "language_loss": 0.81711316,
      "learning_rate": 3.996898704753867e-06,
      "loss": 0.840765,
      "num_input_tokens_seen": 8412355,
      "step": 393,
      "time_per_iteration": 2.519263744354248
    },
    {
      "auxiliary_loss_clip": 0.01303552,
      "auxiliary_loss_mlp": 0.01049967,
      "balance_loss_clip": 1.0884918,
      "balance_loss_mlp": 1.03266931,
      "epoch": 0.04737569891180184,
      "flos": 22053820504320.0,
      "grad_norm": 2.2439759125568517,
      "language_loss": 0.8766396,
      "learning_rate": 3.996855189745609e-06,
      "loss": 0.90017486,
      "num_input_tokens_seen": 8431620,
      "step": 394,
      "time_per_iteration": 2.552218198776245
    },
    {
      "auxiliary_loss_clip": 0.01301317,
      "auxiliary_loss_mlp": 0.01053603,
      "balance_loss_clip": 1.08981705,
      "balance_loss_mlp": 1.03820729,
      "epoch": 0.04749594180244093,
      "flos": 29057370410880.0,
      "grad_norm": 1.873707066620662,
      "language_loss": 0.92844695,
      "learning_rate": 3.996811371816007e-06,
      "loss": 0.95199621,
      "num_input_tokens_seen": 8454045,
      "step": 395,
      "time_per_iteration": 2.56040096282959
    },
    {
      "auxiliary_loss_clip": 0.01303635,
      "auxiliary_loss_mlp": 0.01049532,
      "balance_loss_clip": 1.09247947,
      "balance_loss_mlp": 1.03241336,
      "epoch": 0.04761618469308002,
      "flos": 35112268172160.0,
      "grad_norm": 2.1581223253967248,
      "language_loss": 0.77826864,
      "learning_rate": 3.996767250971707e-06,
      "loss": 0.80180031,
      "num_input_tokens_seen": 8476785,
      "step": 396,
      "time_per_iteration": 2.650474786758423
    },
    {
      "auxiliary_loss_clip": 0.01304516,
      "auxiliary_loss_mlp": 0.01053647,
      "balance_loss_clip": 1.09287548,
      "balance_loss_mlp": 1.03682685,
      "epoch": 0.04773642758371911,
      "flos": 25630702796160.0,
      "grad_norm": 8.904443767513166,
      "language_loss": 0.87129539,
      "learning_rate": 3.996722827219403e-06,
      "loss": 0.89487702,
      "num_input_tokens_seen": 8498400,
      "step": 397,
      "time_per_iteration": 2.5519323348999023
    },
    {
      "auxiliary_loss_clip": 0.0130764,
      "auxiliary_loss_mlp": 0.0106136,
      "balance_loss_clip": 1.09413111,
      "balance_loss_mlp": 1.04283524,
      "epoch": 0.0478566704743582,
      "flos": 20631506688000.0,
      "grad_norm": 2.6549321524853005,
      "language_loss": 0.82442391,
      "learning_rate": 3.996678100565833e-06,
      "loss": 0.84811395,
      "num_input_tokens_seen": 8517455,
      "step": 398,
      "time_per_iteration": 2.5456690788269043
    },
    {
      "auxiliary_loss_clip": 0.01300735,
      "auxiliary_loss_mlp": 0.01045699,
      "balance_loss_clip": 1.09053087,
      "balance_loss_mlp": 1.02945113,
      "epoch": 0.04797691336499729,
      "flos": 18835721210880.0,
      "grad_norm": 3.569940572820551,
      "language_loss": 0.88472986,
      "learning_rate": 3.996633071017783e-06,
      "loss": 0.90819418,
      "num_input_tokens_seen": 8534085,
      "step": 399,
      "time_per_iteration": 2.5035767555236816
    },
    {
      "auxiliary_loss_clip": 0.01304135,
      "auxiliary_loss_mlp": 0.01044468,
      "balance_loss_clip": 1.09117079,
      "balance_loss_mlp": 1.028422,
      "epoch": 0.04809715625563638,
      "flos": 21099673578240.0,
      "grad_norm": 7.106574810778126,
      "language_loss": 0.81886029,
      "learning_rate": 3.996587738582084e-06,
      "loss": 0.84234631,
      "num_input_tokens_seen": 8550885,
      "step": 400,
      "time_per_iteration": 2.5291664600372314
    },
    {
      "auxiliary_loss_clip": 0.01300476,
      "auxiliary_loss_mlp": 0.01045793,
      "balance_loss_clip": 1.08953309,
      "balance_loss_mlp": 1.03045106,
      "epoch": 0.04821739914627548,
      "flos": 23805650712960.0,
      "grad_norm": 2.1858759137231982,
      "language_loss": 0.86060905,
      "learning_rate": 3.9965421032656115e-06,
      "loss": 0.88407177,
      "num_input_tokens_seen": 8570815,
      "step": 401,
      "time_per_iteration": 2.5298314094543457
    },
    {
      "auxiliary_loss_clip": 0.01303603,
      "auxiliary_loss_mlp": 0.01043686,
      "balance_loss_clip": 1.0910511,
      "balance_loss_mlp": 1.02719927,
      "epoch": 0.04833764203691457,
      "flos": 22200587475840.0,
      "grad_norm": 3.4827833299518733,
      "language_loss": 0.94582379,
      "learning_rate": 3.99649616507529e-06,
      "loss": 0.96929669,
      "num_input_tokens_seen": 8589910,
      "step": 402,
      "time_per_iteration": 2.496900796890259
    },
    {
      "auxiliary_loss_clip": 0.01210445,
      "auxiliary_loss_mlp": 0.01006452,
      "balance_loss_clip": 1.06811833,
      "balance_loss_mlp": 1.00175512,
      "epoch": 0.04845788492755366,
      "flos": 65904376896000.0,
      "grad_norm": 0.8906472098807581,
      "language_loss": 0.63159406,
      "learning_rate": 3.996449924018088e-06,
      "loss": 0.65376306,
      "num_input_tokens_seen": 8650370,
      "step": 403,
      "time_per_iteration": 3.0230000019073486
    },
    {
      "auxiliary_loss_clip": 0.01300177,
      "auxiliary_loss_mlp": 0.01052974,
      "balance_loss_clip": 1.08999586,
      "balance_loss_mlp": 1.03690505,
      "epoch": 0.04857812781819275,
      "flos": 19281301424640.0,
      "grad_norm": 1.876376947326497,
      "language_loss": 0.79430288,
      "learning_rate": 3.99640338010102e-06,
      "loss": 0.81783438,
      "num_input_tokens_seen": 8669475,
      "step": 404,
      "time_per_iteration": 2.4802281856536865
    },
    {
      "auxiliary_loss_clip": 0.01297923,
      "auxiliary_loss_mlp": 0.01050529,
      "balance_loss_clip": 1.08738911,
      "balance_loss_mlp": 1.03472233,
      "epoch": 0.04869837070883184,
      "flos": 24062376193920.0,
      "grad_norm": 3.0408680190332413,
      "language_loss": 0.78852701,
      "learning_rate": 3.996356533331146e-06,
      "loss": 0.81201154,
      "num_input_tokens_seen": 8691345,
      "step": 405,
      "time_per_iteration": 2.5322022438049316
    },
    {
      "auxiliary_loss_clip": 0.01306199,
      "auxiliary_loss_mlp": 0.01055284,
      "balance_loss_clip": 1.09018242,
      "balance_loss_mlp": 1.03635359,
      "epoch": 0.04881861359947093,
      "flos": 25187169657600.0,
      "grad_norm": 3.3728429723623283,
      "language_loss": 0.62184596,
      "learning_rate": 3.996309383715573e-06,
      "loss": 0.64546084,
      "num_input_tokens_seen": 8710125,
      "step": 406,
      "time_per_iteration": 2.544234275817871
    },
    {
      "auxiliary_loss_clip": 0.01305045,
      "auxiliary_loss_mlp": 0.01054406,
      "balance_loss_clip": 1.09020185,
      "balance_loss_mlp": 1.03675079,
      "epoch": 0.048938856490110025,
      "flos": 16362913213440.0,
      "grad_norm": 2.1460766372378166,
      "language_loss": 0.73604232,
      "learning_rate": 3.996261931261454e-06,
      "loss": 0.75963688,
      "num_input_tokens_seen": 8728705,
      "step": 407,
      "time_per_iteration": 2.513988733291626
    },
    {
      "auxiliary_loss_clip": 0.01305039,
      "auxiliary_loss_mlp": 0.01049638,
      "balance_loss_clip": 1.09272265,
      "balance_loss_mlp": 1.03355658,
      "epoch": 0.049059099380749115,
      "flos": 29895094379520.0,
      "grad_norm": 2.2496197887028835,
      "language_loss": 0.86484158,
      "learning_rate": 3.996214175975987e-06,
      "loss": 0.88838828,
      "num_input_tokens_seen": 8749225,
      "step": 408,
      "time_per_iteration": 2.5560309886932373
    },
    {
      "auxiliary_loss_clip": 0.01304699,
      "auxiliary_loss_mlp": 0.01054176,
      "balance_loss_clip": 1.09303451,
      "balance_loss_mlp": 1.03692627,
      "epoch": 0.049179342271388204,
      "flos": 35918858027520.0,
      "grad_norm": 2.534470054467721,
      "language_loss": 0.79151297,
      "learning_rate": 3.996166117866417e-06,
      "loss": 0.81510174,
      "num_input_tokens_seen": 8771160,
      "step": 409,
      "time_per_iteration": 3.376248359680176
    },
    {
      "auxiliary_loss_clip": 0.01296076,
      "auxiliary_loss_mlp": 0.01055747,
      "balance_loss_clip": 1.0854665,
      "balance_loss_mlp": 1.04101324,
      "epoch": 0.049299585162027294,
      "flos": 14611226659200.0,
      "grad_norm": 1.9696086851391277,
      "language_loss": 0.86797154,
      "learning_rate": 3.996117756940035e-06,
      "loss": 0.89148974,
      "num_input_tokens_seen": 8787845,
      "step": 410,
      "time_per_iteration": 2.473705530166626
    },
    {
      "auxiliary_loss_clip": 0.01302728,
      "auxiliary_loss_mlp": 0.01056083,
      "balance_loss_clip": 1.09131265,
      "balance_loss_mlp": 1.04043055,
      "epoch": 0.049419828052666384,
      "flos": 19567939956480.0,
      "grad_norm": 2.33901952454538,
      "language_loss": 0.97585934,
      "learning_rate": 3.996069093204175e-06,
      "loss": 0.99944741,
      "num_input_tokens_seen": 8803805,
      "step": 411,
      "time_per_iteration": 3.231684923171997
    },
    {
      "auxiliary_loss_clip": 0.0130594,
      "auxiliary_loss_mlp": 0.01056638,
      "balance_loss_clip": 1.09383583,
      "balance_loss_mlp": 1.03970993,
      "epoch": 0.049540070943305474,
      "flos": 13659916907520.0,
      "grad_norm": 3.440668610506445,
      "language_loss": 0.88337725,
      "learning_rate": 3.996020126666221e-06,
      "loss": 0.90700305,
      "num_input_tokens_seen": 8820785,
      "step": 412,
      "time_per_iteration": 3.25516676902771
    },
    {
      "auxiliary_loss_clip": 0.01304753,
      "auxiliary_loss_mlp": 0.01049464,
      "balance_loss_clip": 1.0917604,
      "balance_loss_mlp": 1.03293586,
      "epoch": 0.04966031383394457,
      "flos": 21832035978240.0,
      "grad_norm": 2.2417010280478,
      "language_loss": 0.82295054,
      "learning_rate": 3.995970857333601e-06,
      "loss": 0.84649271,
      "num_input_tokens_seen": 8841195,
      "step": 413,
      "time_per_iteration": 2.5417888164520264
    },
    {
      "auxiliary_loss_clip": 0.01299425,
      "auxiliary_loss_mlp": 0.0105103,
      "balance_loss_clip": 1.08905685,
      "balance_loss_mlp": 1.03412652,
      "epoch": 0.04978055672458366,
      "flos": 28618793349120.0,
      "grad_norm": 2.0079693888675143,
      "language_loss": 0.79779619,
      "learning_rate": 3.995921285213789e-06,
      "loss": 0.82130075,
      "num_input_tokens_seen": 8861455,
      "step": 414,
      "time_per_iteration": 3.3395025730133057
    },
    {
      "auxiliary_loss_clip": 0.01301858,
      "auxiliary_loss_mlp": 0.01045256,
      "balance_loss_clip": 1.09112263,
      "balance_loss_mlp": 1.02948511,
      "epoch": 0.04990079961522275,
      "flos": 19828220883840.0,
      "grad_norm": 2.4524854653135426,
      "language_loss": 0.80663538,
      "learning_rate": 3.995871410314305e-06,
      "loss": 0.8301065,
      "num_input_tokens_seen": 8880015,
      "step": 415,
      "time_per_iteration": 2.4848568439483643
    },
    {
      "auxiliary_loss_clip": 0.01114402,
      "auxiliary_loss_mlp": 0.01005396,
      "balance_loss_clip": 1.05203295,
      "balance_loss_mlp": 1.00036526,
      "epoch": 0.05002104250586184,
      "flos": 62735045293440.0,
      "grad_norm": 0.9071093127485317,
      "language_loss": 0.59621412,
      "learning_rate": 3.995821232642714e-06,
      "loss": 0.61741203,
      "num_input_tokens_seen": 8938420,
      "step": 416,
      "time_per_iteration": 3.637331485748291
    },
    {
      "auxiliary_loss_clip": 0.0123142,
      "auxiliary_loss_mlp": 0.01064681,
      "balance_loss_clip": 1.08116174,
      "balance_loss_mlp": 1.04811072,
      "epoch": 0.05014128539650093,
      "flos": 27928518710400.0,
      "grad_norm": 2.5049563853789505,
      "language_loss": 0.82263327,
      "learning_rate": 3.995770752206629e-06,
      "loss": 0.84559429,
      "num_input_tokens_seen": 8959495,
      "step": 417,
      "time_per_iteration": 3.0834381580352783
    },
    {
      "auxiliary_loss_clip": 0.01252796,
      "auxiliary_loss_mlp": 0.01046552,
      "balance_loss_clip": 1.0809139,
      "balance_loss_mlp": 1.03112602,
      "epoch": 0.05026152828714002,
      "flos": 17705576620800.0,
      "grad_norm": 2.059622790840989,
      "language_loss": 0.9727608,
      "learning_rate": 3.995719969013709e-06,
      "loss": 0.99575412,
      "num_input_tokens_seen": 8976675,
      "step": 418,
      "time_per_iteration": 2.569093704223633
    },
    {
      "auxiliary_loss_clip": 0.01218575,
      "auxiliary_loss_mlp": 0.0105028,
      "balance_loss_clip": 1.06900334,
      "balance_loss_mlp": 1.03325725,
      "epoch": 0.05038177117777912,
      "flos": 19133277477120.0,
      "grad_norm": 2.6844382968837386,
      "language_loss": 0.85673517,
      "learning_rate": 3.995668883071655e-06,
      "loss": 0.87942374,
      "num_input_tokens_seen": 8992900,
      "step": 419,
      "time_per_iteration": 2.5812184810638428
    },
    {
      "auxiliary_loss_clip": 0.01160447,
      "auxiliary_loss_mlp": 0.01050108,
      "balance_loss_clip": 1.07089043,
      "balance_loss_mlp": 1.03174984,
      "epoch": 0.050502014068418206,
      "flos": 20667704618880.0,
      "grad_norm": 2.795913830836951,
      "language_loss": 0.91033769,
      "learning_rate": 3.995617494388219e-06,
      "loss": 0.93244314,
      "num_input_tokens_seen": 9011020,
      "step": 420,
      "time_per_iteration": 2.7269606590270996
    },
    {
      "auxiliary_loss_clip": 0.01208559,
      "auxiliary_loss_mlp": 0.00767294,
      "balance_loss_clip": 1.06865549,
      "balance_loss_mlp": 1.00037038,
      "epoch": 0.050622256959057296,
      "flos": 21361103740800.0,
      "grad_norm": 2.8509581768428855,
      "language_loss": 0.8052063,
      "learning_rate": 3.995565802971196e-06,
      "loss": 0.82496482,
      "num_input_tokens_seen": 9030995,
      "step": 421,
      "time_per_iteration": 2.646571397781372
    },
    {
      "auxiliary_loss_clip": 0.01236172,
      "auxiliary_loss_mlp": 0.01048028,
      "balance_loss_clip": 1.07975626,
      "balance_loss_mlp": 1.03263783,
      "epoch": 0.050742499849696386,
      "flos": 27673588909440.0,
      "grad_norm": 2.309951927326875,
      "language_loss": 0.67802632,
      "learning_rate": 3.995513808828427e-06,
      "loss": 0.70086837,
      "num_input_tokens_seen": 9053790,
      "step": 422,
      "time_per_iteration": 2.6861119270324707
    },
    {
      "auxiliary_loss_clip": 0.01210123,
      "auxiliary_loss_mlp": 0.01046894,
      "balance_loss_clip": 1.07610393,
      "balance_loss_mlp": 1.03114009,
      "epoch": 0.050862742740335476,
      "flos": 19865999013120.0,
      "grad_norm": 1.990078460280154,
      "language_loss": 0.76829374,
      "learning_rate": 3.9954615119678e-06,
      "loss": 0.79086387,
      "num_input_tokens_seen": 9072345,
      "step": 423,
      "time_per_iteration": 2.656407594680786
    },
    {
      "auxiliary_loss_clip": 0.01237031,
      "auxiliary_loss_mlp": 0.00767157,
      "balance_loss_clip": 1.08159781,
      "balance_loss_mlp": 1.00034535,
      "epoch": 0.050982985630974566,
      "flos": 22085098272000.0,
      "grad_norm": 2.238629667851913,
      "language_loss": 0.80836546,
      "learning_rate": 3.995408912397248e-06,
      "loss": 0.82840741,
      "num_input_tokens_seen": 9090240,
      "step": 424,
      "time_per_iteration": 2.6263763904571533
    },
    {
      "auxiliary_loss_clip": 0.01226642,
      "auxiliary_loss_mlp": 0.01061469,
      "balance_loss_clip": 1.07100904,
      "balance_loss_mlp": 1.04331362,
      "epoch": 0.05110322852161366,
      "flos": 20740962407040.0,
      "grad_norm": 2.6517440055687347,
      "language_loss": 0.9344846,
      "learning_rate": 3.99535601012475e-06,
      "loss": 0.95736575,
      "num_input_tokens_seen": 9105570,
      "step": 425,
      "time_per_iteration": 2.602224349975586
    },
    {
      "auxiliary_loss_clip": 0.01256375,
      "auxiliary_loss_mlp": 0.01049227,
      "balance_loss_clip": 1.08618343,
      "balance_loss_mlp": 1.03399825,
      "epoch": 0.05122347141225275,
      "flos": 28547295327360.0,
      "grad_norm": 2.4083709619741436,
      "language_loss": 0.75634116,
      "learning_rate": 3.995302805158333e-06,
      "loss": 0.77939713,
      "num_input_tokens_seen": 9128225,
      "step": 426,
      "time_per_iteration": 2.638531446456909
    },
    {
      "auxiliary_loss_clip": 0.01266379,
      "auxiliary_loss_mlp": 0.01051185,
      "balance_loss_clip": 1.08115911,
      "balance_loss_mlp": 1.03369665,
      "epoch": 0.05134371430289184,
      "flos": 19722679747200.0,
      "grad_norm": 1.7307656434463916,
      "language_loss": 0.8362813,
      "learning_rate": 3.9952492975060665e-06,
      "loss": 0.85945696,
      "num_input_tokens_seen": 9148295,
      "step": 427,
      "time_per_iteration": 2.601634979248047
    },
    {
      "auxiliary_loss_clip": 0.0120967,
      "auxiliary_loss_mlp": 0.01065646,
      "balance_loss_clip": 1.07445526,
      "balance_loss_mlp": 1.04772937,
      "epoch": 0.05146395719353093,
      "flos": 34458945649920.0,
      "grad_norm": 3.0992201953146608,
      "language_loss": 0.85186446,
      "learning_rate": 3.995195487176067e-06,
      "loss": 0.87461758,
      "num_input_tokens_seen": 9168525,
      "step": 428,
      "time_per_iteration": 2.7809383869171143
    },
    {
      "auxiliary_loss_clip": 0.01270446,
      "auxiliary_loss_mlp": 0.01051445,
      "balance_loss_clip": 1.08646715,
      "balance_loss_mlp": 1.03479099,
      "epoch": 0.05158420008417002,
      "flos": 21760286561280.0,
      "grad_norm": 2.0418866719387156,
      "language_loss": 0.85627377,
      "learning_rate": 3.995141374176499e-06,
      "loss": 0.87949264,
      "num_input_tokens_seen": 9186920,
      "step": 429,
      "time_per_iteration": 2.5999112129211426
    },
    {
      "auxiliary_loss_clip": 0.01140738,
      "auxiliary_loss_mlp": 0.01007032,
      "balance_loss_clip": 1.06094885,
      "balance_loss_mlp": 1.00190639,
      "epoch": 0.05170444297480911,
      "flos": 72553956226560.0,
      "grad_norm": 0.8750638748753014,
      "language_loss": 0.6308645,
      "learning_rate": 3.995086958515572e-06,
      "loss": 0.6523422,
      "num_input_tokens_seen": 9244940,
      "step": 430,
      "time_per_iteration": 3.225308656692505
    },
    {
      "auxiliary_loss_clip": 0.01174279,
      "auxiliary_loss_mlp": 0.01014289,
      "balance_loss_clip": 1.06389117,
      "balance_loss_mlp": 1.00954401,
      "epoch": 0.05182468586544821,
      "flos": 62416159326720.0,
      "grad_norm": 0.8610851578880414,
      "language_loss": 0.59962261,
      "learning_rate": 3.995032240201538e-06,
      "loss": 0.62150824,
      "num_input_tokens_seen": 9307335,
      "step": 431,
      "time_per_iteration": 3.089428424835205
    },
    {
      "auxiliary_loss_clip": 0.01187361,
      "auxiliary_loss_mlp": 0.00762118,
      "balance_loss_clip": 1.09302199,
      "balance_loss_mlp": 1.00084293,
      "epoch": 0.0519449287560873,
      "flos": 41225989432320.0,
      "grad_norm": 0.9336745243549834,
      "language_loss": 0.63130003,
      "learning_rate": 3.9949772192427e-06,
      "loss": 0.6507948,
      "num_input_tokens_seen": 9353960,
      "step": 432,
      "time_per_iteration": 2.896350145339966
    },
    {
      "auxiliary_loss_clip": 0.01256648,
      "auxiliary_loss_mlp": 0.01051591,
      "balance_loss_clip": 1.08280468,
      "balance_loss_mlp": 1.03572416,
      "epoch": 0.05206517164672639,
      "flos": 17494530261120.0,
      "grad_norm": 2.1321792369007926,
      "language_loss": 0.79379702,
      "learning_rate": 3.994921895647405e-06,
      "loss": 0.81687945,
      "num_input_tokens_seen": 9372130,
      "step": 433,
      "time_per_iteration": 2.5556108951568604
    },
    {
      "auxiliary_loss_clip": 0.01202641,
      "auxiliary_loss_mlp": 0.01006132,
      "balance_loss_clip": 1.06615281,
      "balance_loss_mlp": 1.00105405,
      "epoch": 0.05218541453736548,
      "flos": 64002762973440.0,
      "grad_norm": 0.8493936524594021,
      "language_loss": 0.55328262,
      "learning_rate": 3.994866269424043e-06,
      "loss": 0.57537031,
      "num_input_tokens_seen": 9428500,
      "step": 434,
      "time_per_iteration": 3.002349376678467
    },
    {
      "auxiliary_loss_clip": 0.0125132,
      "auxiliary_loss_mlp": 0.01052117,
      "balance_loss_clip": 1.07987523,
      "balance_loss_mlp": 1.03604782,
      "epoch": 0.05230565742800457,
      "flos": 19317319787520.0,
      "grad_norm": 2.693989594552271,
      "language_loss": 0.78207004,
      "learning_rate": 3.9948103405810545e-06,
      "loss": 0.80510437,
      "num_input_tokens_seen": 9447450,
      "step": 435,
      "time_per_iteration": 3.3429179191589355
    },
    {
      "auxiliary_loss_clip": 0.01280138,
      "auxiliary_loss_mlp": 0.01044935,
      "balance_loss_clip": 1.08882666,
      "balance_loss_mlp": 1.0291338,
      "epoch": 0.05242590031864366,
      "flos": 25298636538240.0,
      "grad_norm": 2.2660213899908284,
      "language_loss": 0.85673416,
      "learning_rate": 3.994754109126923e-06,
      "loss": 0.87998492,
      "num_input_tokens_seen": 9468945,
      "step": 436,
      "time_per_iteration": 2.6016345024108887
    },
    {
      "auxiliary_loss_clip": 0.01247723,
      "auxiliary_loss_mlp": 0.01052484,
      "balance_loss_clip": 1.08810282,
      "balance_loss_mlp": 1.03661752,
      "epoch": 0.052546143209282754,
      "flos": 26211629456640.0,
      "grad_norm": 2.512179986666604,
      "language_loss": 0.93455243,
      "learning_rate": 3.994697575070181e-06,
      "loss": 0.95755446,
      "num_input_tokens_seen": 9488405,
      "step": 437,
      "time_per_iteration": 3.536594867706299
    },
    {
      "auxiliary_loss_clip": 0.0114198,
      "auxiliary_loss_mlp": 0.01050174,
      "balance_loss_clip": 1.06509554,
      "balance_loss_mlp": 1.0343554,
      "epoch": 0.052666386099921844,
      "flos": 22158140578560.0,
      "grad_norm": 1.8855771063447926,
      "language_loss": 0.91498101,
      "learning_rate": 3.994640738419402e-06,
      "loss": 0.93690258,
      "num_input_tokens_seen": 9507780,
      "step": 438,
      "time_per_iteration": 3.64687442779541
    },
    {
      "auxiliary_loss_clip": 0.01206168,
      "auxiliary_loss_mlp": 0.01046459,
      "balance_loss_clip": 1.07871413,
      "balance_loss_mlp": 1.03122389,
      "epoch": 0.052786628990560934,
      "flos": 23881817502720.0,
      "grad_norm": 1.971873112092528,
      "language_loss": 0.80965984,
      "learning_rate": 3.9945835991832075e-06,
      "loss": 0.8321861,
      "num_input_tokens_seen": 9529665,
      "step": 439,
      "time_per_iteration": 3.869173049926758
    },
    {
      "auxiliary_loss_clip": 0.0117273,
      "auxiliary_loss_mlp": 0.01054022,
      "balance_loss_clip": 1.07338691,
      "balance_loss_mlp": 1.03835833,
      "epoch": 0.052906871881200024,
      "flos": 24605021934720.0,
      "grad_norm": 6.202027442352903,
      "language_loss": 0.92844307,
      "learning_rate": 3.994526157370268e-06,
      "loss": 0.95071065,
      "num_input_tokens_seen": 9548280,
      "step": 440,
      "time_per_iteration": 3.0181357860565186
    },
    {
      "auxiliary_loss_clip": 0.01169952,
      "auxiliary_loss_mlp": 0.01008114,
      "balance_loss_clip": 1.06230927,
      "balance_loss_mlp": 1.00305927,
      "epoch": 0.053027114771839114,
      "flos": 56461631143680.0,
      "grad_norm": 0.8894884337397011,
      "language_loss": 0.59276617,
      "learning_rate": 3.994468412989296e-06,
      "loss": 0.61454678,
      "num_input_tokens_seen": 9609690,
      "step": 441,
      "time_per_iteration": 3.5589215755462646
    },
    {
      "auxiliary_loss_clip": 0.01274163,
      "auxiliary_loss_mlp": 0.01048795,
      "balance_loss_clip": 1.08580136,
      "balance_loss_mlp": 1.03328562,
      "epoch": 0.053147357662478203,
      "flos": 17311098481920.0,
      "grad_norm": 2.1816266914171307,
      "language_loss": 0.92673051,
      "learning_rate": 3.994410366049052e-06,
      "loss": 0.94996005,
      "num_input_tokens_seen": 9627550,
      "step": 442,
      "time_per_iteration": 2.5437710285186768
    },
    {
      "auxiliary_loss_clip": 0.01281509,
      "auxiliary_loss_mlp": 0.01051354,
      "balance_loss_clip": 1.08813059,
      "balance_loss_mlp": 1.03497446,
      "epoch": 0.0532676005531173,
      "flos": 17164977955200.0,
      "grad_norm": 3.0731468559942403,
      "language_loss": 0.8338908,
      "learning_rate": 3.994352016558341e-06,
      "loss": 0.85721946,
      "num_input_tokens_seen": 9644855,
      "step": 443,
      "time_per_iteration": 2.5842201709747314
    },
    {
      "auxiliary_loss_clip": 0.01193939,
      "auxiliary_loss_mlp": 0.01048303,
      "balance_loss_clip": 1.06569755,
      "balance_loss_mlp": 1.03261542,
      "epoch": 0.05338784344375639,
      "flos": 27819960831360.0,
      "grad_norm": 1.958399291325243,
      "language_loss": 0.73676276,
      "learning_rate": 3.994293364526014e-06,
      "loss": 0.75918519,
      "num_input_tokens_seen": 9665740,
      "step": 444,
      "time_per_iteration": 2.764756917953491
    },
    {
      "auxiliary_loss_clip": 0.01280854,
      "auxiliary_loss_mlp": 0.01054063,
      "balance_loss_clip": 1.09247351,
      "balance_loss_mlp": 1.03750491,
      "epoch": 0.05350808633439548,
      "flos": 21507691144320.0,
      "grad_norm": 2.4837368479813966,
      "language_loss": 0.85062075,
      "learning_rate": 3.99423440996097e-06,
      "loss": 0.87396991,
      "num_input_tokens_seen": 9685280,
      "step": 445,
      "time_per_iteration": 2.5753374099731445
    },
    {
      "auxiliary_loss_clip": 0.01254519,
      "auxiliary_loss_mlp": 0.01047818,
      "balance_loss_clip": 1.08634877,
      "balance_loss_mlp": 1.03062212,
      "epoch": 0.05362832922503457,
      "flos": 20084299920000.0,
      "grad_norm": 3.9202729426342553,
      "language_loss": 0.81593239,
      "learning_rate": 3.994175152872152e-06,
      "loss": 0.83895588,
      "num_input_tokens_seen": 9704365,
      "step": 446,
      "time_per_iteration": 2.6268272399902344
    },
    {
      "auxiliary_loss_clip": 0.01190373,
      "auxiliary_loss_mlp": 0.01046547,
      "balance_loss_clip": 1.08218932,
      "balance_loss_mlp": 1.03183651,
      "epoch": 0.05374857211567366,
      "flos": 26137222433280.0,
      "grad_norm": 2.014614401082391,
      "language_loss": 0.78632635,
      "learning_rate": 3.994115593268548e-06,
      "loss": 0.80869555,
      "num_input_tokens_seen": 9724145,
      "step": 447,
      "time_per_iteration": 2.9443037509918213
    },
    {
      "auxiliary_loss_clip": 0.01258089,
      "auxiliary_loss_mlp": 0.01046262,
      "balance_loss_clip": 1.08389807,
      "balance_loss_mlp": 1.03076458,
      "epoch": 0.05386881500631275,
      "flos": 27486817165440.0,
      "grad_norm": 2.199299326474726,
      "language_loss": 0.82338434,
      "learning_rate": 3.994055731159195e-06,
      "loss": 0.84642786,
      "num_input_tokens_seen": 9741615,
      "step": 448,
      "time_per_iteration": 2.9886929988861084
    },
    {
      "auxiliary_loss_clip": 0.01256127,
      "auxiliary_loss_mlp": 0.01048888,
      "balance_loss_clip": 1.08759356,
      "balance_loss_mlp": 1.03396344,
      "epoch": 0.053989057896951846,
      "flos": 23585087249280.0,
      "grad_norm": 1.962071195771186,
      "language_loss": 0.86853313,
      "learning_rate": 3.993995566553172e-06,
      "loss": 0.89158332,
      "num_input_tokens_seen": 9760580,
      "step": 449,
      "time_per_iteration": 2.6234076023101807
    },
    {
      "auxiliary_loss_clip": 0.01259847,
      "auxiliary_loss_mlp": 0.01044335,
      "balance_loss_clip": 1.08487654,
      "balance_loss_mlp": 1.02833104,
      "epoch": 0.054109300787590936,
      "flos": 25228862369280.0,
      "grad_norm": 1.6416338729544615,
      "language_loss": 0.77113676,
      "learning_rate": 3.993935099459607e-06,
      "loss": 0.79417855,
      "num_input_tokens_seen": 9782195,
      "step": 450,
      "time_per_iteration": 2.6660842895507812
    },
    {
      "auxiliary_loss_clip": 0.01211279,
      "auxiliary_loss_mlp": 0.01039668,
      "balance_loss_clip": 1.07864618,
      "balance_loss_mlp": 1.0260725,
      "epoch": 0.054229543678230026,
      "flos": 23841525421440.0,
      "grad_norm": 2.010757094394469,
      "language_loss": 0.73938978,
      "learning_rate": 3.993874329887673e-06,
      "loss": 0.76189929,
      "num_input_tokens_seen": 9800850,
      "step": 451,
      "time_per_iteration": 2.659179210662842
    },
    {
      "auxiliary_loss_clip": 0.01249441,
      "auxiliary_loss_mlp": 0.01043013,
      "balance_loss_clip": 1.08381295,
      "balance_loss_mlp": 1.02742004,
      "epoch": 0.054349786568869116,
      "flos": 16320933192960.0,
      "grad_norm": 3.0436107702573585,
      "language_loss": 0.86208034,
      "learning_rate": 3.993813257846589e-06,
      "loss": 0.88500494,
      "num_input_tokens_seen": 9817605,
      "step": 452,
      "time_per_iteration": 2.5909886360168457
    },
    {
      "auxiliary_loss_clip": 0.01276616,
      "auxiliary_loss_mlp": 0.01050825,
      "balance_loss_clip": 1.08686566,
      "balance_loss_mlp": 1.03355169,
      "epoch": 0.054470029459508205,
      "flos": 18660729127680.0,
      "grad_norm": 3.138765848676793,
      "language_loss": 0.92330682,
      "learning_rate": 3.993751883345619e-06,
      "loss": 0.94658118,
      "num_input_tokens_seen": 9835965,
      "step": 453,
      "time_per_iteration": 2.504728317260742
    },
    {
      "auxiliary_loss_clip": 0.0123318,
      "auxiliary_loss_mlp": 0.00766411,
      "balance_loss_clip": 1.08228993,
      "balance_loss_mlp": 1.000453,
      "epoch": 0.054590272350147295,
      "flos": 17785298856960.0,
      "grad_norm": 2.3706993843543462,
      "language_loss": 0.87683338,
      "learning_rate": 3.993690206394073e-06,
      "loss": 0.89682931,
      "num_input_tokens_seen": 9852265,
      "step": 454,
      "time_per_iteration": 2.5984673500061035
    },
    {
      "auxiliary_loss_clip": 0.01243323,
      "auxiliary_loss_mlp": 0.01053885,
      "balance_loss_clip": 1.08373785,
      "balance_loss_mlp": 1.038692,
      "epoch": 0.054710515240786385,
      "flos": 17785945301760.0,
      "grad_norm": 2.3857264995510272,
      "language_loss": 0.87682879,
      "learning_rate": 3.993628227001307e-06,
      "loss": 0.8998009,
      "num_input_tokens_seen": 9870465,
      "step": 455,
      "time_per_iteration": 2.563446521759033
    },
    {
      "auxiliary_loss_clip": 0.01264285,
      "auxiliary_loss_mlp": 0.01053564,
      "balance_loss_clip": 1.09054887,
      "balance_loss_mlp": 1.03807902,
      "epoch": 0.05483075813142548,
      "flos": 48210900180480.0,
      "grad_norm": 1.9617662888127165,
      "language_loss": 0.71259665,
      "learning_rate": 3.993565945176726e-06,
      "loss": 0.73577511,
      "num_input_tokens_seen": 9891490,
      "step": 456,
      "time_per_iteration": 2.8105499744415283
    },
    {
      "auxiliary_loss_clip": 0.01226273,
      "auxiliary_loss_mlp": 0.01061144,
      "balance_loss_clip": 1.08284688,
      "balance_loss_mlp": 1.04555202,
      "epoch": 0.05495100102206457,
      "flos": 19682244011520.0,
      "grad_norm": 2.3637790191867336,
      "language_loss": 0.84241927,
      "learning_rate": 3.993503360929776e-06,
      "loss": 0.86529344,
      "num_input_tokens_seen": 9910375,
      "step": 457,
      "time_per_iteration": 2.595066785812378
    },
    {
      "auxiliary_loss_clip": 0.01227299,
      "auxiliary_loss_mlp": 0.01052736,
      "balance_loss_clip": 1.08425784,
      "balance_loss_mlp": 1.03632116,
      "epoch": 0.05507124391270366,
      "flos": 26360048453760.0,
      "grad_norm": 1.5958881756019594,
      "language_loss": 0.80926228,
      "learning_rate": 3.99344047426995e-06,
      "loss": 0.8320626,
      "num_input_tokens_seen": 9931635,
      "step": 458,
      "time_per_iteration": 2.6611104011535645
    },
    {
      "auxiliary_loss_clip": 0.01210025,
      "auxiliary_loss_mlp": 0.01043244,
      "balance_loss_clip": 1.08051562,
      "balance_loss_mlp": 1.0267688,
      "epoch": 0.05519148680334275,
      "flos": 22601314581120.0,
      "grad_norm": 3.9846399215403885,
      "language_loss": 0.93404335,
      "learning_rate": 3.993377285206789e-06,
      "loss": 0.95657599,
      "num_input_tokens_seen": 9951420,
      "step": 459,
      "time_per_iteration": 2.6700196266174316
    },
    {
      "auxiliary_loss_clip": 0.0122751,
      "auxiliary_loss_mlp": 0.01050554,
      "balance_loss_clip": 1.07107866,
      "balance_loss_mlp": 1.03378093,
      "epoch": 0.05531172969398184,
      "flos": 40552519380480.0,
      "grad_norm": 1.911391834028679,
      "language_loss": 0.86516666,
      "learning_rate": 3.99331379374988e-06,
      "loss": 0.88794732,
      "num_input_tokens_seen": 9975025,
      "step": 460,
      "time_per_iteration": 2.7920424938201904
    },
    {
      "auxiliary_loss_clip": 0.01241605,
      "auxiliary_loss_mlp": 0.00766997,
      "balance_loss_clip": 1.0785588,
      "balance_loss_mlp": 1.00040674,
      "epoch": 0.05543197258462093,
      "flos": 23477894087040.0,
      "grad_norm": 2.6747055112289497,
      "language_loss": 0.80268514,
      "learning_rate": 3.993249999908852e-06,
      "loss": 0.82277113,
      "num_input_tokens_seen": 9995175,
      "step": 461,
      "time_per_iteration": 3.4245030879974365
    },
    {
      "auxiliary_loss_clip": 0.01226613,
      "auxiliary_loss_mlp": 0.01043482,
      "balance_loss_clip": 1.07716024,
      "balance_loss_mlp": 1.02767467,
      "epoch": 0.05555221547526003,
      "flos": 18624603024000.0,
      "grad_norm": 2.1480471808251904,
      "language_loss": 0.87541503,
      "learning_rate": 3.993185903693384e-06,
      "loss": 0.89811599,
      "num_input_tokens_seen": 10011975,
      "step": 462,
      "time_per_iteration": 3.3644793033599854
    },
    {
      "auxiliary_loss_clip": 0.01236303,
      "auxiliary_loss_mlp": 0.01047087,
      "balance_loss_clip": 1.08421779,
      "balance_loss_mlp": 1.03185177,
      "epoch": 0.05567245836589912,
      "flos": 23587098410880.0,
      "grad_norm": 2.5609241294517853,
      "language_loss": 0.82691872,
      "learning_rate": 3.9931215051131995e-06,
      "loss": 0.8497526,
      "num_input_tokens_seen": 10032620,
      "step": 463,
      "time_per_iteration": 3.4246227741241455
    },
    {
      "auxiliary_loss_clip": 0.01276688,
      "auxiliary_loss_mlp": 0.00767321,
      "balance_loss_clip": 1.08764863,
      "balance_loss_mlp": 1.00036025,
      "epoch": 0.05579270125653821,
      "flos": 27746667129600.0,
      "grad_norm": 2.2578394660885097,
      "language_loss": 0.80087638,
      "learning_rate": 3.993056804178068e-06,
      "loss": 0.82131648,
      "num_input_tokens_seen": 10054165,
      "step": 464,
      "time_per_iteration": 2.6004881858825684
    },
    {
      "auxiliary_loss_clip": 0.01203122,
      "auxiliary_loss_mlp": 0.01045528,
      "balance_loss_clip": 1.08112097,
      "balance_loss_mlp": 1.02854073,
      "epoch": 0.0559129441471773,
      "flos": 27014161075200.0,
      "grad_norm": 2.358361668025362,
      "language_loss": 0.84574807,
      "learning_rate": 3.992991800897803e-06,
      "loss": 0.86823457,
      "num_input_tokens_seen": 10073970,
      "step": 465,
      "time_per_iteration": 3.601606607437134
    },
    {
      "auxiliary_loss_clip": 0.01276152,
      "auxiliary_loss_mlp": 0.01055405,
      "balance_loss_clip": 1.08570004,
      "balance_loss_mlp": 1.03977656,
      "epoch": 0.05603318703781639,
      "flos": 15229787794560.0,
      "grad_norm": 2.2504391793936325,
      "language_loss": 0.90246737,
      "learning_rate": 3.9929264952822665e-06,
      "loss": 0.92578298,
      "num_input_tokens_seen": 10091505,
      "step": 466,
      "time_per_iteration": 2.7533819675445557
    },
    {
      "auxiliary_loss_clip": 0.01246453,
      "auxiliary_loss_mlp": 0.01048344,
      "balance_loss_clip": 1.08123338,
      "balance_loss_mlp": 1.03120208,
      "epoch": 0.05615342992845548,
      "flos": 22266482976000.0,
      "grad_norm": 2.3741926438143874,
      "language_loss": 0.88426805,
      "learning_rate": 3.992860887341366e-06,
      "loss": 0.90721595,
      "num_input_tokens_seen": 10109675,
      "step": 467,
      "time_per_iteration": 2.609844446182251
    },
    {
      "auxiliary_loss_clip": 0.01238398,
      "auxiliary_loss_mlp": 0.01043702,
      "balance_loss_clip": 1.07966685,
      "balance_loss_mlp": 1.02830005,
      "epoch": 0.056273672819094574,
      "flos": 23584979508480.0,
      "grad_norm": 2.8563332913147907,
      "language_loss": 0.80791014,
      "learning_rate": 3.992794977085052e-06,
      "loss": 0.83073115,
      "num_input_tokens_seen": 10127675,
      "step": 468,
      "time_per_iteration": 2.634011745452881
    },
    {
      "auxiliary_loss_clip": 0.01268197,
      "auxiliary_loss_mlp": 0.01050301,
      "balance_loss_clip": 1.08771336,
      "balance_loss_mlp": 1.0328126,
      "epoch": 0.056393915709733664,
      "flos": 19858708552320.0,
      "grad_norm": 2.678566750163164,
      "language_loss": 0.84780383,
      "learning_rate": 3.992728764523326e-06,
      "loss": 0.87098885,
      "num_input_tokens_seen": 10146620,
      "step": 469,
      "time_per_iteration": 2.563322067260742
    },
    {
      "auxiliary_loss_clip": 0.01237799,
      "auxiliary_loss_mlp": 0.01054531,
      "balance_loss_clip": 1.06916285,
      "balance_loss_mlp": 1.03844953,
      "epoch": 0.05651415860037275,
      "flos": 22163779013760.0,
      "grad_norm": 2.314975418169315,
      "language_loss": 0.80913436,
      "learning_rate": 3.99266224966623e-06,
      "loss": 0.8320576,
      "num_input_tokens_seen": 10167535,
      "step": 470,
      "time_per_iteration": 2.6144895553588867
    },
    {
      "auxiliary_loss_clip": 0.0123891,
      "auxiliary_loss_mlp": 0.01042206,
      "balance_loss_clip": 1.07418442,
      "balance_loss_mlp": 1.02649474,
      "epoch": 0.05663440149101184,
      "flos": 19463548055040.0,
      "grad_norm": 2.0420858130770294,
      "language_loss": 0.87748212,
      "learning_rate": 3.992595432523855e-06,
      "loss": 0.90029335,
      "num_input_tokens_seen": 10184825,
      "step": 471,
      "time_per_iteration": 2.5354578495025635
    },
    {
      "auxiliary_loss_clip": 0.01185351,
      "auxiliary_loss_mlp": 0.00766585,
      "balance_loss_clip": 1.06972587,
      "balance_loss_mlp": 1.00031602,
      "epoch": 0.05675464438165093,
      "flos": 22670226823680.0,
      "grad_norm": 2.0098703886254845,
      "language_loss": 0.86132753,
      "learning_rate": 3.992528313106338e-06,
      "loss": 0.88084686,
      "num_input_tokens_seen": 10203025,
      "step": 472,
      "time_per_iteration": 2.775303363800049
    },
    {
      "auxiliary_loss_clip": 0.01259061,
      "auxiliary_loss_mlp": 0.01046173,
      "balance_loss_clip": 1.08925593,
      "balance_loss_mlp": 1.03016353,
      "epoch": 0.05687488727229002,
      "flos": 16901177495040.0,
      "grad_norm": 2.5317671838343547,
      "language_loss": 0.82150793,
      "learning_rate": 3.9924608914238595e-06,
      "loss": 0.84456027,
      "num_input_tokens_seen": 10218020,
      "step": 473,
      "time_per_iteration": 2.5045228004455566
    },
    {
      "auxiliary_loss_clip": 0.01202409,
      "auxiliary_loss_mlp": 0.01044526,
      "balance_loss_clip": 1.07266235,
      "balance_loss_mlp": 1.02842069,
      "epoch": 0.05699513016292912,
      "flos": 29168980945920.0,
      "grad_norm": 2.5114136187242617,
      "language_loss": 0.8446486,
      "learning_rate": 3.992393167486648e-06,
      "loss": 0.867118,
      "num_input_tokens_seen": 10237170,
      "step": 474,
      "time_per_iteration": 2.701730251312256
    },
    {
      "auxiliary_loss_clip": 0.01257061,
      "auxiliary_loss_mlp": 0.01056131,
      "balance_loss_clip": 1.0849793,
      "balance_loss_mlp": 1.03963244,
      "epoch": 0.05711537305356821,
      "flos": 18916197632640.0,
      "grad_norm": 5.841266111761117,
      "language_loss": 0.80942196,
      "learning_rate": 3.992325141304977e-06,
      "loss": 0.83255392,
      "num_input_tokens_seen": 10255125,
      "step": 475,
      "time_per_iteration": 2.525897741317749
    },
    {
      "auxiliary_loss_clip": 0.01252994,
      "auxiliary_loss_mlp": 0.01051118,
      "balance_loss_clip": 1.08248329,
      "balance_loss_mlp": 1.03517985,
      "epoch": 0.0572356159442073,
      "flos": 26758979879040.0,
      "grad_norm": 2.097578066937504,
      "language_loss": 0.86529493,
      "learning_rate": 3.992256812889166e-06,
      "loss": 0.88833606,
      "num_input_tokens_seen": 10271230,
      "step": 476,
      "time_per_iteration": 2.595367193222046
    },
    {
      "auxiliary_loss_clip": 0.01291304,
      "auxiliary_loss_mlp": 0.00766455,
      "balance_loss_clip": 1.08838117,
      "balance_loss_mlp": 1.000337,
      "epoch": 0.05735585883484639,
      "flos": 35116146840960.0,
      "grad_norm": 2.1385234508199993,
      "language_loss": 0.77270502,
      "learning_rate": 3.992188182249582e-06,
      "loss": 0.79328263,
      "num_input_tokens_seen": 10293125,
      "step": 477,
      "time_per_iteration": 2.608881711959839
    },
    {
      "auxiliary_loss_clip": 0.01261541,
      "auxiliary_loss_mlp": 0.01056987,
      "balance_loss_clip": 1.08503449,
      "balance_loss_mlp": 1.03961813,
      "epoch": 0.05747610172548548,
      "flos": 18734381965440.0,
      "grad_norm": 2.4350046070149842,
      "language_loss": 0.90576243,
      "learning_rate": 3.992119249396633e-06,
      "loss": 0.92894769,
      "num_input_tokens_seen": 10311810,
      "step": 478,
      "time_per_iteration": 2.54819917678833
    },
    {
      "auxiliary_loss_clip": 0.01290531,
      "auxiliary_loss_mlp": 0.00767084,
      "balance_loss_clip": 1.08834577,
      "balance_loss_mlp": 1.00032616,
      "epoch": 0.05759634461612457,
      "flos": 27964752554880.0,
      "grad_norm": 2.543246613152425,
      "language_loss": 0.82341719,
      "learning_rate": 3.992050014340778e-06,
      "loss": 0.84399331,
      "num_input_tokens_seen": 10332165,
      "step": 479,
      "time_per_iteration": 2.5502758026123047
    },
    {
      "auxiliary_loss_clip": 0.01176787,
      "auxiliary_loss_mlp": 0.01005521,
      "balance_loss_clip": 1.05974007,
      "balance_loss_mlp": 1.00144422,
      "epoch": 0.057716587506763666,
      "flos": 69292009405440.0,
      "grad_norm": 0.9605372180758126,
      "language_loss": 0.55028367,
      "learning_rate": 3.99198047709252e-06,
      "loss": 0.57210672,
      "num_input_tokens_seen": 10393685,
      "step": 480,
      "time_per_iteration": 3.1895198822021484
    },
    {
      "auxiliary_loss_clip": 0.01215764,
      "auxiliary_loss_mlp": 0.01045465,
      "balance_loss_clip": 1.07609463,
      "balance_loss_mlp": 1.02913976,
      "epoch": 0.057836830397402755,
      "flos": 25009196745600.0,
      "grad_norm": 2.0169778111787675,
      "language_loss": 0.78694874,
      "learning_rate": 3.991910637662408e-06,
      "loss": 0.80956107,
      "num_input_tokens_seen": 10413975,
      "step": 481,
      "time_per_iteration": 2.671320915222168
    },
    {
      "auxiliary_loss_clip": 0.0127244,
      "auxiliary_loss_mlp": 0.01050075,
      "balance_loss_clip": 1.08604336,
      "balance_loss_mlp": 1.03489947,
      "epoch": 0.057957073288041845,
      "flos": 25593894334080.0,
      "grad_norm": 2.451105904561278,
      "language_loss": 0.80809951,
      "learning_rate": 3.9918404960610355e-06,
      "loss": 0.83132458,
      "num_input_tokens_seen": 10433005,
      "step": 482,
      "time_per_iteration": 2.5554568767547607
    },
    {
      "auxiliary_loss_clip": 0.01224892,
      "auxiliary_loss_mlp": 0.01039945,
      "balance_loss_clip": 1.08200693,
      "balance_loss_mlp": 1.02353549,
      "epoch": 0.058077316178680935,
      "flos": 20777411733120.0,
      "grad_norm": 2.388869196914054,
      "language_loss": 0.77393872,
      "learning_rate": 3.991770052299043e-06,
      "loss": 0.79658711,
      "num_input_tokens_seen": 10451235,
      "step": 483,
      "time_per_iteration": 2.650195837020874
    },
    {
      "auxiliary_loss_clip": 0.01263777,
      "auxiliary_loss_mlp": 0.01060068,
      "balance_loss_clip": 1.08404851,
      "balance_loss_mlp": 1.04414773,
      "epoch": 0.058197559069320025,
      "flos": 18916484941440.0,
      "grad_norm": 2.218185048872841,
      "language_loss": 0.87981349,
      "learning_rate": 3.991699306387118e-06,
      "loss": 0.90305197,
      "num_input_tokens_seen": 10469705,
      "step": 484,
      "time_per_iteration": 2.502558946609497
    },
    {
      "auxiliary_loss_clip": 0.01182375,
      "auxiliary_loss_mlp": 0.01051697,
      "balance_loss_clip": 1.07068205,
      "balance_loss_mlp": 1.03578842,
      "epoch": 0.058317801959959115,
      "flos": 24863327614080.0,
      "grad_norm": 2.062308736661351,
      "language_loss": 0.78183645,
      "learning_rate": 3.991628258335991e-06,
      "loss": 0.80417717,
      "num_input_tokens_seen": 10491910,
      "step": 485,
      "time_per_iteration": 2.814039707183838
    },
    {
      "auxiliary_loss_clip": 0.01152094,
      "auxiliary_loss_mlp": 0.01046376,
      "balance_loss_clip": 1.06469619,
      "balance_loss_mlp": 1.02943611,
      "epoch": 0.05843804485059821,
      "flos": 23257977068160.0,
      "grad_norm": 10.545915468225775,
      "language_loss": 0.87879646,
      "learning_rate": 3.991556908156442e-06,
      "loss": 0.90078115,
      "num_input_tokens_seen": 10508435,
      "step": 486,
      "time_per_iteration": 3.944243907928467
    },
    {
      "auxiliary_loss_clip": 0.01250699,
      "auxiliary_loss_mlp": 0.01048708,
      "balance_loss_clip": 1.08262682,
      "balance_loss_mlp": 1.03157735,
      "epoch": 0.0585582877412373,
      "flos": 23150532510720.0,
      "grad_norm": 1.979730639506934,
      "language_loss": 0.87787771,
      "learning_rate": 3.9914852558592914e-06,
      "loss": 0.90087175,
      "num_input_tokens_seen": 10529485,
      "step": 487,
      "time_per_iteration": 3.143906831741333
    },
    {
      "auxiliary_loss_clip": 0.01214232,
      "auxiliary_loss_mlp": 0.01043501,
      "balance_loss_clip": 1.07615662,
      "balance_loss_mlp": 1.02609682,
      "epoch": 0.05867853063187639,
      "flos": 23506406507520.0,
      "grad_norm": 4.11737859734171,
      "language_loss": 0.80857533,
      "learning_rate": 3.991413301455413e-06,
      "loss": 0.83115268,
      "num_input_tokens_seen": 10545935,
      "step": 488,
      "time_per_iteration": 3.4857757091522217
    },
    {
      "auxiliary_loss_clip": 0.01220386,
      "auxiliary_loss_mlp": 0.0076729,
      "balance_loss_clip": 1.07543314,
      "balance_loss_mlp": 1.00043738,
      "epoch": 0.05879877352251548,
      "flos": 29495803818240.0,
      "grad_norm": 2.2827482153682372,
      "language_loss": 0.77917862,
      "learning_rate": 3.991341044955719e-06,
      "loss": 0.79905534,
      "num_input_tokens_seen": 10565690,
      "step": 489,
      "time_per_iteration": 3.631779909133911
    },
    {
      "auxiliary_loss_clip": 0.01173082,
      "auxiliary_loss_mlp": 0.01048957,
      "balance_loss_clip": 1.05808139,
      "balance_loss_mlp": 1.03231514,
      "epoch": 0.05891901641315457,
      "flos": 20157485880960.0,
      "grad_norm": 2.1547499618737427,
      "language_loss": 0.81944275,
      "learning_rate": 3.991268486371172e-06,
      "loss": 0.84166312,
      "num_input_tokens_seen": 10584245,
      "step": 490,
      "time_per_iteration": 2.711778163909912
    },
    {
      "auxiliary_loss_clip": 0.0118316,
      "auxiliary_loss_mlp": 0.01052151,
      "balance_loss_clip": 1.06072116,
      "balance_loss_mlp": 1.03384078,
      "epoch": 0.05903925930379366,
      "flos": 24644200694400.0,
      "grad_norm": 4.884887259152398,
      "language_loss": 0.87850231,
      "learning_rate": 3.991195625712779e-06,
      "loss": 0.90085542,
      "num_input_tokens_seen": 10601210,
      "step": 491,
      "time_per_iteration": 3.4909963607788086
    },
    {
      "auxiliary_loss_clip": 0.01235681,
      "auxiliary_loss_mlp": 0.01045943,
      "balance_loss_clip": 1.077806,
      "balance_loss_mlp": 1.02925396,
      "epoch": 0.05915950219443276,
      "flos": 21250391045760.0,
      "grad_norm": 2.2220755833880372,
      "language_loss": 0.81854105,
      "learning_rate": 3.991122462991592e-06,
      "loss": 0.84135729,
      "num_input_tokens_seen": 10620730,
      "step": 492,
      "time_per_iteration": 2.6232192516326904
    },
    {
      "auxiliary_loss_clip": 0.01243696,
      "auxiliary_loss_mlp": 0.01053419,
      "balance_loss_clip": 1.08493876,
      "balance_loss_mlp": 1.0347507,
      "epoch": 0.05927974508507185,
      "flos": 9902727319680.0,
      "grad_norm": 6.004935194616814,
      "language_loss": 0.81394434,
      "learning_rate": 3.991048998218712e-06,
      "loss": 0.83691549,
      "num_input_tokens_seen": 10634035,
      "step": 493,
      "time_per_iteration": 2.5829579830169678
    },
    {
      "auxiliary_loss_clip": 0.01157749,
      "auxiliary_loss_mlp": 0.010456,
      "balance_loss_clip": 1.06739616,
      "balance_loss_mlp": 1.02892232,
      "epoch": 0.05939998797571094,
      "flos": 18259499232000.0,
      "grad_norm": 2.201559575862782,
      "language_loss": 0.76667523,
      "learning_rate": 3.990975231405281e-06,
      "loss": 0.78870875,
      "num_input_tokens_seen": 10652485,
      "step": 494,
      "time_per_iteration": 2.803849697113037
    },
    {
      "auxiliary_loss_clip": 0.01252903,
      "auxiliary_loss_mlp": 0.01048176,
      "balance_loss_clip": 1.08216941,
      "balance_loss_mlp": 1.03335238,
      "epoch": 0.05952023086635003,
      "flos": 28256598558720.0,
      "grad_norm": 2.0294542105328452,
      "language_loss": 0.79102576,
      "learning_rate": 3.990901162562491e-06,
      "loss": 0.81403655,
      "num_input_tokens_seen": 10673175,
      "step": 495,
      "time_per_iteration": 3.065412759780884
    },
    {
      "auxiliary_loss_clip": 0.01253314,
      "auxiliary_loss_mlp": 0.01040786,
      "balance_loss_clip": 1.0808084,
      "balance_loss_mlp": 1.02570581,
      "epoch": 0.05964047375698912,
      "flos": 14902498045440.0,
      "grad_norm": 2.1540890325407127,
      "language_loss": 0.90537,
      "learning_rate": 3.9908267917015765e-06,
      "loss": 0.92831099,
      "num_input_tokens_seen": 10691235,
      "step": 496,
      "time_per_iteration": 2.543672561645508
    },
    {
      "auxiliary_loss_clip": 0.01253572,
      "auxiliary_loss_mlp": 0.01046395,
      "balance_loss_clip": 1.08350348,
      "balance_loss_mlp": 1.03100502,
      "epoch": 0.059760716647628206,
      "flos": 23185581206400.0,
      "grad_norm": 2.7920509013984107,
      "language_loss": 0.93264681,
      "learning_rate": 3.990752118833821e-06,
      "loss": 0.95564651,
      "num_input_tokens_seen": 10708675,
      "step": 497,
      "time_per_iteration": 2.5661351680755615
    },
    {
      "auxiliary_loss_clip": 0.01183098,
      "auxiliary_loss_mlp": 0.01042734,
      "balance_loss_clip": 1.06842256,
      "balance_loss_mlp": 1.02746332,
      "epoch": 0.0598809595382673,
      "flos": 22746968231040.0,
      "grad_norm": 1.7776741609791218,
      "language_loss": 0.78033984,
      "learning_rate": 3.990677143970553e-06,
      "loss": 0.80259824,
      "num_input_tokens_seen": 10729485,
      "step": 498,
      "time_per_iteration": 2.7075023651123047
    },
    {
      "auxiliary_loss_clip": 0.01227162,
      "auxiliary_loss_mlp": 0.00766837,
      "balance_loss_clip": 1.07367611,
      "balance_loss_mlp": 1.00048184,
      "epoch": 0.06000120242890639,
      "flos": 22127221946880.0,
      "grad_norm": 2.2751021305469896,
      "language_loss": 0.81056559,
      "learning_rate": 3.990601867123144e-06,
      "loss": 0.83050549,
      "num_input_tokens_seen": 10749210,
      "step": 499,
      "time_per_iteration": 2.5975754261016846
    },
    {
      "auxiliary_loss_clip": 0.01264805,
      "auxiliary_loss_mlp": 0.01049214,
      "balance_loss_clip": 1.07945895,
      "balance_loss_mlp": 1.0330081,
      "epoch": 0.06012144531954548,
      "flos": 19171773878400.0,
      "grad_norm": 2.518122184621464,
      "language_loss": 0.85098267,
      "learning_rate": 3.990526288303014e-06,
      "loss": 0.87412286,
      "num_input_tokens_seen": 10768000,
      "step": 500,
      "time_per_iteration": 2.5378878116607666
    },
    {
      "auxiliary_loss_clip": 0.01195565,
      "auxiliary_loss_mlp": 0.01040265,
      "balance_loss_clip": 1.06722784,
      "balance_loss_mlp": 1.02702689,
      "epoch": 0.06024168821018457,
      "flos": 22783345729920.0,
      "grad_norm": 1.783904862406792,
      "language_loss": 0.90786541,
      "learning_rate": 3.9904504075216295e-06,
      "loss": 0.9302237,
      "num_input_tokens_seen": 10788760,
      "step": 501,
      "time_per_iteration": 2.6170411109924316
    },
    {
      "auxiliary_loss_clip": 0.0124367,
      "auxiliary_loss_mlp": 0.0105145,
      "balance_loss_clip": 1.07617188,
      "balance_loss_mlp": 1.0345577,
      "epoch": 0.06036193110082366,
      "flos": 18770687637120.0,
      "grad_norm": 2.2825891827379996,
      "language_loss": 0.93883544,
      "learning_rate": 3.990374224790501e-06,
      "loss": 0.96178663,
      "num_input_tokens_seen": 10806965,
      "step": 502,
      "time_per_iteration": 2.554232597351074
    },
    {
      "auxiliary_loss_clip": 0.01234768,
      "auxiliary_loss_mlp": 0.01057433,
      "balance_loss_clip": 1.07705975,
      "balance_loss_mlp": 1.04203129,
      "epoch": 0.06048217399146275,
      "flos": 17201570935680.0,
      "grad_norm": 2.10814174886202,
      "language_loss": 0.71001184,
      "learning_rate": 3.990297740121185e-06,
      "loss": 0.73293388,
      "num_input_tokens_seen": 10824900,
      "step": 503,
      "time_per_iteration": 2.5278897285461426
    },
    {
      "auxiliary_loss_clip": 0.01177569,
      "auxiliary_loss_mlp": 0.01051072,
      "balance_loss_clip": 1.06949711,
      "balance_loss_mlp": 1.03633761,
      "epoch": 0.06060241688210185,
      "flos": 24024131187840.0,
      "grad_norm": 1.8676010728900043,
      "language_loss": 0.78240573,
      "learning_rate": 3.990220953525284e-06,
      "loss": 0.80469215,
      "num_input_tokens_seen": 10842010,
      "step": 504,
      "time_per_iteration": 2.7730515003204346
    },
    {
      "auxiliary_loss_clip": 0.01268133,
      "auxiliary_loss_mlp": 0.00766634,
      "balance_loss_clip": 1.08273768,
      "balance_loss_mlp": 1.00058746,
      "epoch": 0.06072265977274094,
      "flos": 14611190745600.0,
      "grad_norm": 3.008126198120762,
      "language_loss": 0.74561745,
      "learning_rate": 3.9901438650144465e-06,
      "loss": 0.7659651,
      "num_input_tokens_seen": 10858260,
      "step": 505,
      "time_per_iteration": 2.5028927326202393
    },
    {
      "auxiliary_loss_clip": 0.01252607,
      "auxiliary_loss_mlp": 0.01045506,
      "balance_loss_clip": 1.07825983,
      "balance_loss_mlp": 1.0296278,
      "epoch": 0.06084290266338003,
      "flos": 20558284813440.0,
      "grad_norm": 2.423353036789115,
      "language_loss": 0.91489947,
      "learning_rate": 3.990066474600367e-06,
      "loss": 0.93788064,
      "num_input_tokens_seen": 10876230,
      "step": 506,
      "time_per_iteration": 2.565094232559204
    },
    {
      "auxiliary_loss_clip": 0.01284424,
      "auxiliary_loss_mlp": 0.01047793,
      "balance_loss_clip": 1.08428192,
      "balance_loss_mlp": 1.03300512,
      "epoch": 0.06096314555401912,
      "flos": 22309217182080.0,
      "grad_norm": 1.897936872731947,
      "language_loss": 0.67686951,
      "learning_rate": 3.989988782294786e-06,
      "loss": 0.70019168,
      "num_input_tokens_seen": 10896320,
      "step": 507,
      "time_per_iteration": 2.4962689876556396
    },
    {
      "auxiliary_loss_clip": 0.01190872,
      "auxiliary_loss_mlp": 0.01051939,
      "balance_loss_clip": 1.07110333,
      "balance_loss_mlp": 1.03737164,
      "epoch": 0.06108338844465821,
      "flos": 19131374056320.0,
      "grad_norm": 1.8374167957959031,
      "language_loss": 0.95049453,
      "learning_rate": 3.989910788109489e-06,
      "loss": 0.97292262,
      "num_input_tokens_seen": 10912970,
      "step": 508,
      "time_per_iteration": 2.6346936225891113
    },
    {
      "auxiliary_loss_clip": 0.01255675,
      "auxiliary_loss_mlp": 0.01052471,
      "balance_loss_clip": 1.08447695,
      "balance_loss_mlp": 1.03697431,
      "epoch": 0.0612036313352973,
      "flos": 33584018169600.0,
      "grad_norm": 2.3085056043392007,
      "language_loss": 0.74869269,
      "learning_rate": 3.989832492056307e-06,
      "loss": 0.77177417,
      "num_input_tokens_seen": 10933995,
      "step": 509,
      "time_per_iteration": 2.6467952728271484
    },
    {
      "auxiliary_loss_clip": 0.01225553,
      "auxiliary_loss_mlp": 0.01053929,
      "balance_loss_clip": 1.08022797,
      "balance_loss_mlp": 1.03690577,
      "epoch": 0.06132387422593639,
      "flos": 27490552179840.0,
      "grad_norm": 3.78470637740176,
      "language_loss": 0.80977631,
      "learning_rate": 3.989753894147119e-06,
      "loss": 0.83257115,
      "num_input_tokens_seen": 10954120,
      "step": 510,
      "time_per_iteration": 2.6823770999908447
    },
    {
      "auxiliary_loss_clip": 0.01195006,
      "auxiliary_loss_mlp": 0.01045994,
      "balance_loss_clip": 1.07640743,
      "balance_loss_mlp": 1.03205872,
      "epoch": 0.061444117116575485,
      "flos": 25885057979520.0,
      "grad_norm": 1.880970616875943,
      "language_loss": 0.80032355,
      "learning_rate": 3.989674994393846e-06,
      "loss": 0.82273352,
      "num_input_tokens_seen": 10973595,
      "step": 511,
      "time_per_iteration": 3.4227092266082764
    },
    {
      "auxiliary_loss_clip": 0.01267326,
      "auxiliary_loss_mlp": 0.01048675,
      "balance_loss_clip": 1.08425093,
      "balance_loss_mlp": 1.03408432,
      "epoch": 0.061564360007214575,
      "flos": 28512031150080.0,
      "grad_norm": 2.189516168162112,
      "language_loss": 0.94003594,
      "learning_rate": 3.98959579280846e-06,
      "loss": 0.96319592,
      "num_input_tokens_seen": 10991995,
      "step": 512,
      "time_per_iteration": 2.605437994003296
    },
    {
      "auxiliary_loss_clip": 0.01156239,
      "auxiliary_loss_mlp": 0.01047468,
      "balance_loss_clip": 1.0682683,
      "balance_loss_mlp": 1.03150558,
      "epoch": 0.061684602897853665,
      "flos": 12094355652480.0,
      "grad_norm": 1.9851680454392253,
      "language_loss": 0.8248322,
      "learning_rate": 3.989516289402973e-06,
      "loss": 0.84686929,
      "num_input_tokens_seen": 11007625,
      "step": 513,
      "time_per_iteration": 2.70862078666687
    },
    {
      "auxiliary_loss_clip": 0.01253152,
      "auxiliary_loss_mlp": 0.01043673,
      "balance_loss_clip": 1.08063304,
      "balance_loss_mlp": 1.02945709,
      "epoch": 0.061804845788492754,
      "flos": 19532639865600.0,
      "grad_norm": 2.6019917789153286,
      "language_loss": 0.79912186,
      "learning_rate": 3.989436484189447e-06,
      "loss": 0.82209003,
      "num_input_tokens_seen": 11025570,
      "step": 514,
      "time_per_iteration": 3.5181736946105957
    },
    {
      "auxiliary_loss_clip": 0.0122629,
      "auxiliary_loss_mlp": 0.01049113,
      "balance_loss_clip": 1.07739651,
      "balance_loss_mlp": 1.03359818,
      "epoch": 0.061925088679131844,
      "flos": 15341111020800.0,
      "grad_norm": 3.4676005642022076,
      "language_loss": 0.80776888,
      "learning_rate": 3.9893563771799885e-06,
      "loss": 0.83052295,
      "num_input_tokens_seen": 11042045,
      "step": 515,
      "time_per_iteration": 2.604769229888916
    },
    {
      "auxiliary_loss_clip": 0.01237492,
      "auxiliary_loss_mlp": 0.01053353,
      "balance_loss_clip": 1.08108211,
      "balance_loss_mlp": 1.03698564,
      "epoch": 0.062045331569770934,
      "flos": 25919927107200.0,
      "grad_norm": 2.0556478455793767,
      "language_loss": 0.86337888,
      "learning_rate": 3.989275968386749e-06,
      "loss": 0.88628733,
      "num_input_tokens_seen": 11059955,
      "step": 516,
      "time_per_iteration": 3.3584494590759277
    },
    {
      "auxiliary_loss_clip": 0.01248563,
      "auxiliary_loss_mlp": 0.01054723,
      "balance_loss_clip": 1.07999754,
      "balance_loss_mlp": 1.03995299,
      "epoch": 0.06216557446041003,
      "flos": 28110621686400.0,
      "grad_norm": 2.672464555507555,
      "language_loss": 0.76794553,
      "learning_rate": 3.989195257821926e-06,
      "loss": 0.79097837,
      "num_input_tokens_seen": 11078440,
      "step": 517,
      "time_per_iteration": 3.316333293914795
    },
    {
      "auxiliary_loss_clip": 0.0121883,
      "auxiliary_loss_mlp": 0.00767798,
      "balance_loss_clip": 1.07423937,
      "balance_loss_mlp": 1.00063181,
      "epoch": 0.06228581735104912,
      "flos": 23478181395840.0,
      "grad_norm": 2.0689029675720585,
      "language_loss": 0.84447569,
      "learning_rate": 3.989114245497765e-06,
      "loss": 0.86434197,
      "num_input_tokens_seen": 11098240,
      "step": 518,
      "time_per_iteration": 2.6749613285064697
    },
    {
      "auxiliary_loss_clip": 0.01216369,
      "auxiliary_loss_mlp": 0.01049927,
      "balance_loss_clip": 1.07206762,
      "balance_loss_mlp": 1.03528845,
      "epoch": 0.06240606024168821,
      "flos": 15195205975680.0,
      "grad_norm": 2.7528149948590404,
      "language_loss": 0.94786197,
      "learning_rate": 3.989032931426554e-06,
      "loss": 0.97052491,
      "num_input_tokens_seen": 11115395,
      "step": 519,
      "time_per_iteration": 2.5838866233825684
    },
    {
      "auxiliary_loss_clip": 0.0128239,
      "auxiliary_loss_mlp": 0.0076663,
      "balance_loss_clip": 1.08333528,
      "balance_loss_mlp": 1.00057626,
      "epoch": 0.06252630313232731,
      "flos": 20631829910400.0,
      "grad_norm": 2.2872453841306832,
      "language_loss": 0.86977112,
      "learning_rate": 3.9889513156206295e-06,
      "loss": 0.89026129,
      "num_input_tokens_seen": 11134835,
      "step": 520,
      "time_per_iteration": 2.5275113582611084
    },
    {
      "auxiliary_loss_clip": 0.01264312,
      "auxiliary_loss_mlp": 0.0104982,
      "balance_loss_clip": 1.08162057,
      "balance_loss_mlp": 1.03310692,
      "epoch": 0.06264654602296639,
      "flos": 20778058177920.0,
      "grad_norm": 3.9043468092248297,
      "language_loss": 0.73617291,
      "learning_rate": 3.988869398092371e-06,
      "loss": 0.75931418,
      "num_input_tokens_seen": 11154745,
      "step": 521,
      "time_per_iteration": 2.5170180797576904
    },
    {
      "auxiliary_loss_clip": 0.01215506,
      "auxiliary_loss_mlp": 0.01045043,
      "balance_loss_clip": 1.07604373,
      "balance_loss_mlp": 1.02806759,
      "epoch": 0.06276678891360549,
      "flos": 29605798241280.0,
      "grad_norm": 2.824172027679709,
      "language_loss": 0.79149419,
      "learning_rate": 3.988787178854206e-06,
      "loss": 0.81409967,
      "num_input_tokens_seen": 11174280,
      "step": 522,
      "time_per_iteration": 2.692387580871582
    },
    {
      "auxiliary_loss_clip": 0.01223249,
      "auxiliary_loss_mlp": 0.01044288,
      "balance_loss_clip": 1.06443298,
      "balance_loss_mlp": 1.0270859,
      "epoch": 0.06288703180424457,
      "flos": 22126288193280.0,
      "grad_norm": 3.0274630934057702,
      "language_loss": 0.87513053,
      "learning_rate": 3.988704657918608e-06,
      "loss": 0.89780587,
      "num_input_tokens_seen": 11193340,
      "step": 523,
      "time_per_iteration": 2.565263271331787
    },
    {
      "auxiliary_loss_clip": 0.01217754,
      "auxiliary_loss_mlp": 0.01049328,
      "balance_loss_clip": 1.06728506,
      "balance_loss_mlp": 1.03358018,
      "epoch": 0.06300727469488367,
      "flos": 14976689587200.0,
      "grad_norm": 3.405298001378303,
      "language_loss": 0.79806662,
      "learning_rate": 3.988621835298094e-06,
      "loss": 0.82073748,
      "num_input_tokens_seen": 11210555,
      "step": 524,
      "time_per_iteration": 2.57342529296875
    },
    {
      "auxiliary_loss_clip": 0.01178215,
      "auxiliary_loss_mlp": 0.01040688,
      "balance_loss_clip": 1.06195545,
      "balance_loss_mlp": 1.02634108,
      "epoch": 0.06312751758552275,
      "flos": 24535391420160.0,
      "grad_norm": 1.9690066659543703,
      "language_loss": 0.91605997,
      "learning_rate": 3.988538711005229e-06,
      "loss": 0.93824899,
      "num_input_tokens_seen": 11230010,
      "step": 525,
      "time_per_iteration": 2.6920900344848633
    },
    {
      "auxiliary_loss_clip": 0.01260905,
      "auxiliary_loss_mlp": 0.01043782,
      "balance_loss_clip": 1.08130872,
      "balance_loss_mlp": 1.02985811,
      "epoch": 0.06324776047616185,
      "flos": 21507008785920.0,
      "grad_norm": 2.1676465539137326,
      "language_loss": 0.87840921,
      "learning_rate": 3.988455285052622e-06,
      "loss": 0.90145612,
      "num_input_tokens_seen": 11246190,
      "step": 526,
      "time_per_iteration": 2.5371923446655273
    },
    {
      "auxiliary_loss_clip": 0.01237941,
      "auxiliary_loss_mlp": 0.01054522,
      "balance_loss_clip": 1.07802832,
      "balance_loss_mlp": 1.03810728,
      "epoch": 0.06336800336680094,
      "flos": 21688034353920.0,
      "grad_norm": 2.2568735006104466,
      "language_loss": 0.84159231,
      "learning_rate": 3.98837155745293e-06,
      "loss": 0.86451691,
      "num_input_tokens_seen": 11264230,
      "step": 527,
      "time_per_iteration": 2.5844242572784424
    },
    {
      "auxiliary_loss_clip": 0.01218037,
      "auxiliary_loss_mlp": 0.01049309,
      "balance_loss_clip": 1.07580256,
      "balance_loss_mlp": 1.03285182,
      "epoch": 0.06348824625744003,
      "flos": 19500895221120.0,
      "grad_norm": 2.6940283750290797,
      "language_loss": 0.75969422,
      "learning_rate": 3.988287528218854e-06,
      "loss": 0.78236771,
      "num_input_tokens_seen": 11283015,
      "step": 528,
      "time_per_iteration": 2.606173276901245
    },
    {
      "auxiliary_loss_clip": 0.01247146,
      "auxiliary_loss_mlp": 0.01050144,
      "balance_loss_clip": 1.07723308,
      "balance_loss_mlp": 1.03502834,
      "epoch": 0.06360848914807912,
      "flos": 15481233976320.0,
      "grad_norm": 2.274711536127546,
      "language_loss": 0.90341216,
      "learning_rate": 3.98820319736314e-06,
      "loss": 0.92638505,
      "num_input_tokens_seen": 11299630,
      "step": 529,
      "time_per_iteration": 2.5092389583587646
    },
    {
      "auxiliary_loss_clip": 0.01236862,
      "auxiliary_loss_mlp": 0.01047392,
      "balance_loss_clip": 1.07827437,
      "balance_loss_mlp": 1.03311038,
      "epoch": 0.0637287320387182,
      "flos": 20593369422720.0,
      "grad_norm": 1.972552687859833,
      "language_loss": 0.85549283,
      "learning_rate": 3.988118564898582e-06,
      "loss": 0.8783353,
      "num_input_tokens_seen": 11319170,
      "step": 530,
      "time_per_iteration": 2.5870721340179443
    },
    {
      "auxiliary_loss_clip": 0.01287581,
      "auxiliary_loss_mlp": 0.01045231,
      "balance_loss_clip": 1.08594728,
      "balance_loss_mlp": 1.02910209,
      "epoch": 0.0638489749293573,
      "flos": 17412222245760.0,
      "grad_norm": 2.3861038816015996,
      "language_loss": 0.8946712,
      "learning_rate": 3.988033630838019e-06,
      "loss": 0.91799927,
      "num_input_tokens_seen": 11333210,
      "step": 531,
      "time_per_iteration": 2.4500410556793213
    },
    {
      "auxiliary_loss_clip": 0.01219993,
      "auxiliary_loss_mlp": 0.01054009,
      "balance_loss_clip": 1.08001709,
      "balance_loss_mlp": 1.03911948,
      "epoch": 0.0639692178199964,
      "flos": 23807661874560.0,
      "grad_norm": 1.8475598373847602,
      "language_loss": 0.88320529,
      "learning_rate": 3.987948395194334e-06,
      "loss": 0.9059453,
      "num_input_tokens_seen": 11355590,
      "step": 532,
      "time_per_iteration": 2.659235715866089
    },
    {
      "auxiliary_loss_clip": 0.0118293,
      "auxiliary_loss_mlp": 0.01038745,
      "balance_loss_clip": 1.06315899,
      "balance_loss_mlp": 1.02432108,
      "epoch": 0.06408946071063548,
      "flos": 18477225521280.0,
      "grad_norm": 2.1368710052769013,
      "language_loss": 0.76679677,
      "learning_rate": 3.987862857980458e-06,
      "loss": 0.7890135,
      "num_input_tokens_seen": 11371535,
      "step": 533,
      "time_per_iteration": 2.5942165851593018
    },
    {
      "auxiliary_loss_clip": 0.01286184,
      "auxiliary_loss_mlp": 0.01054553,
      "balance_loss_clip": 1.08654869,
      "balance_loss_mlp": 1.03875208,
      "epoch": 0.06420970360127458,
      "flos": 27162220936320.0,
      "grad_norm": 4.31102957364051,
      "language_loss": 0.7685414,
      "learning_rate": 3.987777019209368e-06,
      "loss": 0.7919488,
      "num_input_tokens_seen": 11392050,
      "step": 534,
      "time_per_iteration": 2.563671112060547
    },
    {
      "auxiliary_loss_clip": 0.01240266,
      "auxiliary_loss_mlp": 0.0104814,
      "balance_loss_clip": 1.07515824,
      "balance_loss_mlp": 1.03369212,
      "epoch": 0.06432994649191366,
      "flos": 23659673840640.0,
      "grad_norm": 2.4593418500623794,
      "language_loss": 0.81301296,
      "learning_rate": 3.987690878894084e-06,
      "loss": 0.83589703,
      "num_input_tokens_seen": 11411765,
      "step": 535,
      "time_per_iteration": 2.5576229095458984
    },
    {
      "auxiliary_loss_clip": 0.01268986,
      "auxiliary_loss_mlp": 0.01049065,
      "balance_loss_clip": 1.08329201,
      "balance_loss_mlp": 1.03243542,
      "epoch": 0.06445018938255276,
      "flos": 23403953940480.0,
      "grad_norm": 2.5684754626869166,
      "language_loss": 0.85329169,
      "learning_rate": 3.987604437047673e-06,
      "loss": 0.87647223,
      "num_input_tokens_seen": 11431565,
      "step": 536,
      "time_per_iteration": 2.5443599224090576
    },
    {
      "auxiliary_loss_clip": 0.01178687,
      "auxiliary_loss_mlp": 0.01043004,
      "balance_loss_clip": 1.0725894,
      "balance_loss_mlp": 1.02815676,
      "epoch": 0.06457043227319184,
      "flos": 19646692525440.0,
      "grad_norm": 2.5965415443085353,
      "language_loss": 0.77718711,
      "learning_rate": 3.987517693683251e-06,
      "loss": 0.79940403,
      "num_input_tokens_seen": 11450140,
      "step": 537,
      "time_per_iteration": 3.641993761062622
    },
    {
      "auxiliary_loss_clip": 0.0124281,
      "auxiliary_loss_mlp": 0.01052818,
      "balance_loss_clip": 1.08217943,
      "balance_loss_mlp": 1.03728533,
      "epoch": 0.06469067516383094,
      "flos": 16978744915200.0,
      "grad_norm": 5.805567376118538,
      "language_loss": 0.96131563,
      "learning_rate": 3.9874306488139745e-06,
      "loss": 0.98427188,
      "num_input_tokens_seen": 11465400,
      "step": 538,
      "time_per_iteration": 2.538276195526123
    },
    {
      "auxiliary_loss_clip": 0.0115567,
      "auxiliary_loss_mlp": 0.01044635,
      "balance_loss_clip": 1.06629944,
      "balance_loss_mlp": 1.02985334,
      "epoch": 0.06481091805447003,
      "flos": 23296401642240.0,
      "grad_norm": 2.4111212945223506,
      "language_loss": 0.88151073,
      "learning_rate": 3.987343302453049e-06,
      "loss": 0.90351373,
      "num_input_tokens_seen": 11486675,
      "step": 539,
      "time_per_iteration": 3.5277140140533447
    },
    {
      "auxiliary_loss_clip": 0.01233672,
      "auxiliary_loss_mlp": 0.00766772,
      "balance_loss_clip": 1.07634377,
      "balance_loss_mlp": 1.00062859,
      "epoch": 0.06493116094510912,
      "flos": 29172356824320.0,
      "grad_norm": 1.667596201872934,
      "language_loss": 0.82873112,
      "learning_rate": 3.987255654613724e-06,
      "loss": 0.84873557,
      "num_input_tokens_seen": 11510440,
      "step": 540,
      "time_per_iteration": 2.6583375930786133
    },
    {
      "auxiliary_loss_clip": 0.01216784,
      "auxiliary_loss_mlp": 0.01053043,
      "balance_loss_clip": 1.06608188,
      "balance_loss_mlp": 1.03802896,
      "epoch": 0.06505140383574821,
      "flos": 19865065259520.0,
      "grad_norm": 2.517263846152075,
      "language_loss": 0.70360148,
      "learning_rate": 3.987167705309296e-06,
      "loss": 0.72629976,
      "num_input_tokens_seen": 11529715,
      "step": 541,
      "time_per_iteration": 3.3921093940734863
    },
    {
      "auxiliary_loss_clip": 0.01244115,
      "auxiliary_loss_mlp": 0.01038144,
      "balance_loss_clip": 1.08180285,
      "balance_loss_mlp": 1.02386308,
      "epoch": 0.0651716467263873,
      "flos": 17924703540480.0,
      "grad_norm": 2.0904701494131555,
      "language_loss": 0.95350158,
      "learning_rate": 3.987079454553108e-06,
      "loss": 0.9763242,
      "num_input_tokens_seen": 11547665,
      "step": 542,
      "time_per_iteration": 3.3429877758026123
    },
    {
      "auxiliary_loss_clip": 0.01235656,
      "auxiliary_loss_mlp": 0.01054219,
      "balance_loss_clip": 1.08217657,
      "balance_loss_mlp": 1.03822064,
      "epoch": 0.0652918896170264,
      "flos": 20842840356480.0,
      "grad_norm": 1.8121398705704936,
      "language_loss": 0.91252637,
      "learning_rate": 3.986990902358546e-06,
      "loss": 0.93542516,
      "num_input_tokens_seen": 11564605,
      "step": 543,
      "time_per_iteration": 2.5890839099884033
    },
    {
      "auxiliary_loss_clip": 0.01267825,
      "auxiliary_loss_mlp": 0.01056563,
      "balance_loss_clip": 1.0790354,
      "balance_loss_mlp": 1.04139924,
      "epoch": 0.06541213250766549,
      "flos": 21872507627520.0,
      "grad_norm": 1.983021610155836,
      "language_loss": 0.93380809,
      "learning_rate": 3.986902048739045e-06,
      "loss": 0.95705199,
      "num_input_tokens_seen": 11584550,
      "step": 544,
      "time_per_iteration": 2.541215658187866
    },
    {
      "auxiliary_loss_clip": 0.01248885,
      "auxiliary_loss_mlp": 0.01051224,
      "balance_loss_clip": 1.07607937,
      "balance_loss_mlp": 1.03482032,
      "epoch": 0.06553237539830457,
      "flos": 23110743219840.0,
      "grad_norm": 3.04130565650579,
      "language_loss": 0.80019277,
      "learning_rate": 3.986812893708082e-06,
      "loss": 0.82319391,
      "num_input_tokens_seen": 11600740,
      "step": 545,
      "time_per_iteration": 2.537158489227295
    },
    {
      "auxiliary_loss_clip": 0.0123414,
      "auxiliary_loss_mlp": 0.01048616,
      "balance_loss_clip": 1.0823729,
      "balance_loss_mlp": 1.03119969,
      "epoch": 0.06565261828894367,
      "flos": 17923769786880.0,
      "grad_norm": 2.8285938694620336,
      "language_loss": 0.8140831,
      "learning_rate": 3.9867234372791826e-06,
      "loss": 0.83691072,
      "num_input_tokens_seen": 11618695,
      "step": 546,
      "time_per_iteration": 2.594949245452881
    },
    {
      "auxiliary_loss_clip": 0.01241496,
      "auxiliary_loss_mlp": 0.01045555,
      "balance_loss_clip": 1.079494,
      "balance_loss_mlp": 1.03104746,
      "epoch": 0.06577286117958275,
      "flos": 22783058421120.0,
      "grad_norm": 1.5606643177908206,
      "language_loss": 0.87352741,
      "learning_rate": 3.986633679465918e-06,
      "loss": 0.89639789,
      "num_input_tokens_seen": 11638850,
      "step": 547,
      "time_per_iteration": 2.5484778881073
    },
    {
      "auxiliary_loss_clip": 0.01167749,
      "auxiliary_loss_mlp": 0.01053121,
      "balance_loss_clip": 1.0613302,
      "balance_loss_mlp": 1.03796935,
      "epoch": 0.06589310407022185,
      "flos": 23696194993920.0,
      "grad_norm": 2.5557813792595576,
      "language_loss": 0.80938148,
      "learning_rate": 3.986543620281904e-06,
      "loss": 0.83159018,
      "num_input_tokens_seen": 11658500,
      "step": 548,
      "time_per_iteration": 2.7898001670837402
    },
    {
      "auxiliary_loss_clip": 0.01183593,
      "auxiliary_loss_mlp": 0.01059305,
      "balance_loss_clip": 1.06186199,
      "balance_loss_mlp": 1.04318821,
      "epoch": 0.06601334696086093,
      "flos": 26864772410880.0,
      "grad_norm": 1.7875625397915023,
      "language_loss": 0.911035,
      "learning_rate": 3.986453259740802e-06,
      "loss": 0.93346405,
      "num_input_tokens_seen": 11676670,
      "step": 549,
      "time_per_iteration": 2.8070292472839355
    },
    {
      "auxiliary_loss_clip": 0.01185939,
      "auxiliary_loss_mlp": 0.01063165,
      "balance_loss_clip": 1.07831633,
      "balance_loss_mlp": 1.04654765,
      "epoch": 0.06613358985150003,
      "flos": 12567694101120.0,
      "grad_norm": 2.366877751665442,
      "language_loss": 0.78959256,
      "learning_rate": 3.986362597856319e-06,
      "loss": 0.8120836,
      "num_input_tokens_seen": 11693170,
      "step": 550,
      "time_per_iteration": 2.6798501014709473
    },
    {
      "auxiliary_loss_clip": 0.01195813,
      "auxiliary_loss_mlp": 0.01047095,
      "balance_loss_clip": 1.07187688,
      "balance_loss_mlp": 1.03056073,
      "epoch": 0.06625383274213913,
      "flos": 18332505624960.0,
      "grad_norm": 2.528845607711333,
      "language_loss": 0.81946075,
      "learning_rate": 3.986271634642211e-06,
      "loss": 0.84188986,
      "num_input_tokens_seen": 11710150,
      "step": 551,
      "time_per_iteration": 2.6199100017547607
    },
    {
      "auxiliary_loss_clip": 0.01195423,
      "auxiliary_loss_mlp": 0.01044969,
      "balance_loss_clip": 1.065624,
      "balance_loss_mlp": 1.03020477,
      "epoch": 0.06637407563277821,
      "flos": 15375585098880.0,
      "grad_norm": 2.1623783497661413,
      "language_loss": 0.81411564,
      "learning_rate": 3.986180370112274e-06,
      "loss": 0.83651948,
      "num_input_tokens_seen": 11726670,
      "step": 552,
      "time_per_iteration": 2.605624198913574
    },
    {
      "auxiliary_loss_clip": 0.01227792,
      "auxiliary_loss_mlp": 0.01042203,
      "balance_loss_clip": 1.07769227,
      "balance_loss_mlp": 1.02814221,
      "epoch": 0.0664943185234173,
      "flos": 24025244509440.0,
      "grad_norm": 1.7460252667341025,
      "language_loss": 0.74728787,
      "learning_rate": 3.986088804280354e-06,
      "loss": 0.76998776,
      "num_input_tokens_seen": 11746400,
      "step": 553,
      "time_per_iteration": 2.6343066692352295
    },
    {
      "auxiliary_loss_clip": 0.01199783,
      "auxiliary_loss_mlp": 0.0104308,
      "balance_loss_clip": 1.07830215,
      "balance_loss_mlp": 1.02819037,
      "epoch": 0.06661456141405639,
      "flos": 20957503547520.0,
      "grad_norm": 2.2837112454116832,
      "language_loss": 0.94128704,
      "learning_rate": 3.985996937160342e-06,
      "loss": 0.96371561,
      "num_input_tokens_seen": 11765590,
      "step": 554,
      "time_per_iteration": 2.659449577331543
    },
    {
      "auxiliary_loss_clip": 0.01246877,
      "auxiliary_loss_mlp": 0.01049031,
      "balance_loss_clip": 1.08122849,
      "balance_loss_mlp": 1.03360581,
      "epoch": 0.06673480430469549,
      "flos": 52223953322880.0,
      "grad_norm": 2.066866785232136,
      "language_loss": 0.68851584,
      "learning_rate": 3.985904768766173e-06,
      "loss": 0.7114749,
      "num_input_tokens_seen": 11788365,
      "step": 555,
      "time_per_iteration": 2.8401646614074707
    },
    {
      "auxiliary_loss_clip": 0.01254319,
      "auxiliary_loss_mlp": 0.010442,
      "balance_loss_clip": 1.08585715,
      "balance_loss_mlp": 1.02796328,
      "epoch": 0.06685504719533458,
      "flos": 16217079995520.0,
      "grad_norm": 3.8244567926633932,
      "language_loss": 0.76217055,
      "learning_rate": 3.98581229911183e-06,
      "loss": 0.78515577,
      "num_input_tokens_seen": 11807285,
      "step": 556,
      "time_per_iteration": 2.567657709121704
    },
    {
      "auxiliary_loss_clip": 0.01248573,
      "auxiliary_loss_mlp": 0.01042053,
      "balance_loss_clip": 1.07676435,
      "balance_loss_mlp": 1.02691948,
      "epoch": 0.06697529008597367,
      "flos": 22491535639680.0,
      "grad_norm": 2.1660946588903247,
      "language_loss": 0.92225415,
      "learning_rate": 3.985719528211341e-06,
      "loss": 0.94516039,
      "num_input_tokens_seen": 11826655,
      "step": 557,
      "time_per_iteration": 2.5995259284973145
    },
    {
      "auxiliary_loss_clip": 0.01193936,
      "auxiliary_loss_mlp": 0.01007546,
      "balance_loss_clip": 1.07528758,
      "balance_loss_mlp": 1.00392234,
      "epoch": 0.06709553297661276,
      "flos": 62688216936960.0,
      "grad_norm": 0.8381759435356868,
      "language_loss": 0.63007897,
      "learning_rate": 3.985626456078777e-06,
      "loss": 0.65209383,
      "num_input_tokens_seen": 11891310,
      "step": 558,
      "time_per_iteration": 3.2146925926208496
    },
    {
      "auxiliary_loss_clip": 0.01269051,
      "auxiliary_loss_mlp": 0.01048994,
      "balance_loss_clip": 1.08803296,
      "balance_loss_mlp": 1.03446245,
      "epoch": 0.06721577586725185,
      "flos": 11216590997760.0,
      "grad_norm": 2.1451021927980305,
      "language_loss": 0.86321807,
      "learning_rate": 3.985533082728259e-06,
      "loss": 0.88639855,
      "num_input_tokens_seen": 11906965,
      "step": 559,
      "time_per_iteration": 2.497136116027832
    },
    {
      "auxiliary_loss_clip": 0.01192788,
      "auxiliary_loss_mlp": 0.01042611,
      "balance_loss_clip": 1.07067084,
      "balance_loss_mlp": 1.02651775,
      "epoch": 0.06733601875789094,
      "flos": 25922189664000.0,
      "grad_norm": 1.941463555387104,
      "language_loss": 0.74692959,
      "learning_rate": 3.985439408173951e-06,
      "loss": 0.76928359,
      "num_input_tokens_seen": 11927190,
      "step": 560,
      "time_per_iteration": 2.6921513080596924
    },
    {
      "auxiliary_loss_clip": 0.01216333,
      "auxiliary_loss_mlp": 0.01051607,
      "balance_loss_clip": 1.07988381,
      "balance_loss_mlp": 1.03599072,
      "epoch": 0.06745626164853002,
      "flos": 20813645577600.0,
      "grad_norm": 2.758169354872853,
      "language_loss": 0.71018106,
      "learning_rate": 3.9853454324300634e-06,
      "loss": 0.73286045,
      "num_input_tokens_seen": 11946400,
      "step": 561,
      "time_per_iteration": 2.617393970489502
    },
    {
      "auxiliary_loss_clip": 0.01242385,
      "auxiliary_loss_mlp": 0.01042458,
      "balance_loss_clip": 1.08219314,
      "balance_loss_mlp": 1.02707994,
      "epoch": 0.06757650453916912,
      "flos": 19829262378240.0,
      "grad_norm": 1.9235018309936267,
      "language_loss": 0.7796424,
      "learning_rate": 3.985251155510852e-06,
      "loss": 0.80249083,
      "num_input_tokens_seen": 11965430,
      "step": 562,
      "time_per_iteration": 2.560236692428589
    },
    {
      "auxiliary_loss_clip": 0.01243602,
      "auxiliary_loss_mlp": 0.01045731,
      "balance_loss_clip": 1.07825685,
      "balance_loss_mlp": 1.03095508,
      "epoch": 0.06769674742980822,
      "flos": 25739224761600.0,
      "grad_norm": 5.263481089547647,
      "language_loss": 0.80386722,
      "learning_rate": 3.98515657743062e-06,
      "loss": 0.82676059,
      "num_input_tokens_seen": 11984895,
      "step": 563,
      "time_per_iteration": 3.423363447189331
    },
    {
      "auxiliary_loss_clip": 0.01148917,
      "auxiliary_loss_mlp": 0.01043096,
      "balance_loss_clip": 1.05992234,
      "balance_loss_mlp": 1.02758694,
      "epoch": 0.0678169903204473,
      "flos": 13074788355840.0,
      "grad_norm": 2.3115278246754216,
      "language_loss": 0.77863348,
      "learning_rate": 3.985061698203711e-06,
      "loss": 0.80055356,
      "num_input_tokens_seen": 12002010,
      "step": 564,
      "time_per_iteration": 2.6908326148986816
    },
    {
      "auxiliary_loss_clip": 0.01095754,
      "auxiliary_loss_mlp": 0.01004671,
      "balance_loss_clip": 1.05807853,
      "balance_loss_mlp": 1.00133288,
      "epoch": 0.0679372332110864,
      "flos": 70865830788480.0,
      "grad_norm": 0.8808394977245944,
      "language_loss": 0.63781232,
      "learning_rate": 3.984966517844523e-06,
      "loss": 0.65881658,
      "num_input_tokens_seen": 12057255,
      "step": 565,
      "time_per_iteration": 4.151827573776245
    },
    {
      "auxiliary_loss_clip": 0.01239532,
      "auxiliary_loss_mlp": 0.01049811,
      "balance_loss_clip": 1.07903838,
      "balance_loss_mlp": 1.03265691,
      "epoch": 0.06805747610172548,
      "flos": 28256418990720.0,
      "grad_norm": 2.2390611681183477,
      "language_loss": 0.8078416,
      "learning_rate": 3.984871036367492e-06,
      "loss": 0.83073509,
      "num_input_tokens_seen": 12077280,
      "step": 566,
      "time_per_iteration": 2.9077959060668945
    },
    {
      "auxiliary_loss_clip": 0.01174384,
      "auxiliary_loss_mlp": 0.0104535,
      "balance_loss_clip": 1.05928206,
      "balance_loss_mlp": 1.02975774,
      "epoch": 0.06817771899236458,
      "flos": 20120533764480.0,
      "grad_norm": 2.20772925865805,
      "language_loss": 0.8294003,
      "learning_rate": 3.984775253787102e-06,
      "loss": 0.85159767,
      "num_input_tokens_seen": 12095570,
      "step": 567,
      "time_per_iteration": 3.4738125801086426
    },
    {
      "auxiliary_loss_clip": 0.01239331,
      "auxiliary_loss_mlp": 0.01039366,
      "balance_loss_clip": 1.08252883,
      "balance_loss_mlp": 1.02385688,
      "epoch": 0.06829796188300366,
      "flos": 17930629284480.0,
      "grad_norm": 2.603594558703699,
      "language_loss": 0.87950253,
      "learning_rate": 3.984679170117885e-06,
      "loss": 0.90228951,
      "num_input_tokens_seen": 12111775,
      "step": 568,
      "time_per_iteration": 3.5776493549346924
    },
    {
      "auxiliary_loss_clip": 0.01225117,
      "auxiliary_loss_mlp": 0.0076693,
      "balance_loss_clip": 1.07384968,
      "balance_loss_mlp": 1.00090706,
      "epoch": 0.06841820477364276,
      "flos": 14501627285760.0,
      "grad_norm": 2.8193778604223394,
      "language_loss": 0.78842568,
      "learning_rate": 3.984582785374415e-06,
      "loss": 0.80834615,
      "num_input_tokens_seen": 12129215,
      "step": 569,
      "time_per_iteration": 2.641599416732788
    },
    {
      "auxiliary_loss_clip": 0.01241662,
      "auxiliary_loss_mlp": 0.00767089,
      "balance_loss_clip": 1.08216381,
      "balance_loss_mlp": 1.00082874,
      "epoch": 0.06853844766428185,
      "flos": 21938474954880.0,
      "grad_norm": 3.4517540222270684,
      "language_loss": 0.80883962,
      "learning_rate": 3.9844860995713155e-06,
      "loss": 0.8289271,
      "num_input_tokens_seen": 12148755,
      "step": 570,
      "time_per_iteration": 2.6378378868103027
    },
    {
      "auxiliary_loss_clip": 0.01233904,
      "auxiliary_loss_mlp": 0.01045893,
      "balance_loss_clip": 1.07917786,
      "balance_loss_mlp": 1.0315696,
      "epoch": 0.06865869055492094,
      "flos": 16800628348800.0,
      "grad_norm": 2.5986501137780964,
      "language_loss": 0.83066285,
      "learning_rate": 3.9843891127232524e-06,
      "loss": 0.85346085,
      "num_input_tokens_seen": 12166290,
      "step": 571,
      "time_per_iteration": 2.5830204486846924
    },
    {
      "auxiliary_loss_clip": 0.01276354,
      "auxiliary_loss_mlp": 0.01046917,
      "balance_loss_clip": 1.08412361,
      "balance_loss_mlp": 1.03285003,
      "epoch": 0.06877893344556003,
      "flos": 19937281553280.0,
      "grad_norm": 2.6613942561571884,
      "language_loss": 0.67062712,
      "learning_rate": 3.984291824844938e-06,
      "loss": 0.69385982,
      "num_input_tokens_seen": 12181385,
      "step": 572,
      "time_per_iteration": 2.500764846801758
    },
    {
      "auxiliary_loss_clip": 0.01250171,
      "auxiliary_loss_mlp": 0.01038574,
      "balance_loss_clip": 1.0828073,
      "balance_loss_mlp": 1.02261174,
      "epoch": 0.06889917633619912,
      "flos": 23039388852480.0,
      "grad_norm": 3.0890994285056648,
      "language_loss": 0.85107583,
      "learning_rate": 3.984194235951132e-06,
      "loss": 0.87396324,
      "num_input_tokens_seen": 12197530,
      "step": 573,
      "time_per_iteration": 2.547579288482666
    },
    {
      "auxiliary_loss_clip": 0.01245684,
      "auxiliary_loss_mlp": 0.01044906,
      "balance_loss_clip": 1.07287645,
      "balance_loss_mlp": 1.02920628,
      "epoch": 0.06901941922683821,
      "flos": 20960556203520.0,
      "grad_norm": 2.810158689030013,
      "language_loss": 0.84467685,
      "learning_rate": 3.9840963460566375e-06,
      "loss": 0.8675828,
      "num_input_tokens_seen": 12216310,
      "step": 574,
      "time_per_iteration": 2.5444529056549072
    },
    {
      "auxiliary_loss_clip": 0.01212242,
      "auxiliary_loss_mlp": 0.01041294,
      "balance_loss_clip": 1.07370138,
      "balance_loss_mlp": 1.02665555,
      "epoch": 0.06913966211747731,
      "flos": 24821850384000.0,
      "grad_norm": 1.655594832083946,
      "language_loss": 0.89707005,
      "learning_rate": 3.983998155176305e-06,
      "loss": 0.91960543,
      "num_input_tokens_seen": 12236670,
      "step": 575,
      "time_per_iteration": 2.656935930252075
    },
    {
      "auxiliary_loss_clip": 0.01198924,
      "auxiliary_loss_mlp": 0.01006411,
      "balance_loss_clip": 1.07235169,
      "balance_loss_mlp": 1.00302529,
      "epoch": 0.06925990500811639,
      "flos": 58367446957440.0,
      "grad_norm": 0.8167493505374961,
      "language_loss": 0.57100546,
      "learning_rate": 3.9838996633250305e-06,
      "loss": 0.59305882,
      "num_input_tokens_seen": 12297185,
      "step": 576,
      "time_per_iteration": 3.065204620361328
    },
    {
      "auxiliary_loss_clip": 0.0124233,
      "auxiliary_loss_mlp": 0.01040838,
      "balance_loss_clip": 1.07935667,
      "balance_loss_mlp": 1.02641988,
      "epoch": 0.06938014789875549,
      "flos": 12749940731520.0,
      "grad_norm": 2.1206654073879374,
      "language_loss": 0.88233972,
      "learning_rate": 3.983800870517753e-06,
      "loss": 0.90517139,
      "num_input_tokens_seen": 12313975,
      "step": 577,
      "time_per_iteration": 2.530272960662842
    },
    {
      "auxiliary_loss_clip": 0.0124767,
      "auxiliary_loss_mlp": 0.01037255,
      "balance_loss_clip": 1.08566284,
      "balance_loss_mlp": 1.02307522,
      "epoch": 0.06950039078939457,
      "flos": 22820226019200.0,
      "grad_norm": 5.2308610533059525,
      "language_loss": 0.78212428,
      "learning_rate": 3.983701776769463e-06,
      "loss": 0.80497354,
      "num_input_tokens_seen": 12331385,
      "step": 578,
      "time_per_iteration": 2.6114859580993652
    },
    {
      "auxiliary_loss_clip": 0.01236976,
      "auxiliary_loss_mlp": 0.01052105,
      "balance_loss_clip": 1.08223057,
      "balance_loss_mlp": 1.03782415,
      "epoch": 0.06962063368003367,
      "flos": 21941348042880.0,
      "grad_norm": 2.2343328502672817,
      "language_loss": 0.85820395,
      "learning_rate": 3.9836023820951885e-06,
      "loss": 0.88109475,
      "num_input_tokens_seen": 12350600,
      "step": 579,
      "time_per_iteration": 2.5820112228393555
    },
    {
      "auxiliary_loss_clip": 0.01239176,
      "auxiliary_loss_mlp": 0.01047856,
      "balance_loss_clip": 1.07738304,
      "balance_loss_mlp": 1.03341365,
      "epoch": 0.06974087657067275,
      "flos": 20706021452160.0,
      "grad_norm": 2.58727666265256,
      "language_loss": 0.6858412,
      "learning_rate": 3.983502686510011e-06,
      "loss": 0.7087115,
      "num_input_tokens_seen": 12371430,
      "step": 580,
      "time_per_iteration": 2.5942752361297607
    },
    {
      "auxiliary_loss_clip": 0.01276696,
      "auxiliary_loss_mlp": 0.01042991,
      "balance_loss_clip": 1.08430266,
      "balance_loss_mlp": 1.0275768,
      "epoch": 0.06986111946131185,
      "flos": 22638230784000.0,
      "grad_norm": 1.9007316777114827,
      "language_loss": 0.73651177,
      "learning_rate": 3.9834026900290525e-06,
      "loss": 0.75970864,
      "num_input_tokens_seen": 12390825,
      "step": 581,
      "time_per_iteration": 2.523956298828125
    },
    {
      "auxiliary_loss_clip": 0.01242631,
      "auxiliary_loss_mlp": 0.0104829,
      "balance_loss_clip": 1.08110642,
      "balance_loss_mlp": 1.03285241,
      "epoch": 0.06998136235195095,
      "flos": 26943453152640.0,
      "grad_norm": 1.9920089789324555,
      "language_loss": 1.00210655,
      "learning_rate": 3.983302392667482e-06,
      "loss": 1.02501571,
      "num_input_tokens_seen": 12411670,
      "step": 582,
      "time_per_iteration": 2.639482259750366
    },
    {
      "auxiliary_loss_clip": 0.01213124,
      "auxiliary_loss_mlp": 0.01043124,
      "balance_loss_clip": 1.08046651,
      "balance_loss_mlp": 1.02781808,
      "epoch": 0.07010160524259003,
      "flos": 22492505306880.0,
      "grad_norm": 2.0453542383782453,
      "language_loss": 0.93526417,
      "learning_rate": 3.983201794440517e-06,
      "loss": 0.95782661,
      "num_input_tokens_seen": 12431245,
      "step": 583,
      "time_per_iteration": 2.6006150245666504
    },
    {
      "auxiliary_loss_clip": 0.01236952,
      "auxiliary_loss_mlp": 0.01048269,
      "balance_loss_clip": 1.08419204,
      "balance_loss_mlp": 1.03436351,
      "epoch": 0.07022184813322913,
      "flos": 18332541538560.0,
      "grad_norm": 2.194306684696329,
      "language_loss": 0.67416131,
      "learning_rate": 3.9831008953634165e-06,
      "loss": 0.6970135,
      "num_input_tokens_seen": 12450535,
      "step": 584,
      "time_per_iteration": 2.589796543121338
    },
    {
      "auxiliary_loss_clip": 0.01202433,
      "auxiliary_loss_mlp": 0.0104555,
      "balance_loss_clip": 1.07442331,
      "balance_loss_mlp": 1.03078616,
      "epoch": 0.07034209102386821,
      "flos": 24675550289280.0,
      "grad_norm": 37.7090390485317,
      "language_loss": 0.81432027,
      "learning_rate": 3.9829996954514864e-06,
      "loss": 0.8368001,
      "num_input_tokens_seen": 12469675,
      "step": 585,
      "time_per_iteration": 2.6647837162017822
    },
    {
      "auxiliary_loss_clip": 0.0125941,
      "auxiliary_loss_mlp": 0.00765939,
      "balance_loss_clip": 1.08506036,
      "balance_loss_mlp": 1.00083709,
      "epoch": 0.0704623339145073,
      "flos": 25995878415360.0,
      "grad_norm": 1.9691080743472145,
      "language_loss": 0.8444221,
      "learning_rate": 3.982898194720079e-06,
      "loss": 0.86467552,
      "num_input_tokens_seen": 12490405,
      "step": 586,
      "time_per_iteration": 2.6676113605499268
    },
    {
      "auxiliary_loss_clip": 0.01206716,
      "auxiliary_loss_mlp": 0.01052071,
      "balance_loss_clip": 1.07591677,
      "balance_loss_mlp": 1.03664505,
      "epoch": 0.0705825768051464,
      "flos": 25338318088320.0,
      "grad_norm": 2.07720934382009,
      "language_loss": 0.826976,
      "learning_rate": 3.982796393184592e-06,
      "loss": 0.8495639,
      "num_input_tokens_seen": 12509485,
      "step": 587,
      "time_per_iteration": 2.648129463195801
    },
    {
      "auxiliary_loss_clip": 0.0117931,
      "auxiliary_loss_mlp": 0.01007528,
      "balance_loss_clip": 1.06798196,
      "balance_loss_mlp": 1.00392807,
      "epoch": 0.07070281969578548,
      "flos": 66047552507520.0,
      "grad_norm": 0.7931773189524897,
      "language_loss": 0.62615216,
      "learning_rate": 3.98269429086047e-06,
      "loss": 0.64802051,
      "num_input_tokens_seen": 12567325,
      "step": 588,
      "time_per_iteration": 3.046891927719116
    },
    {
      "auxiliary_loss_clip": 0.01284172,
      "auxiliary_loss_mlp": 0.01047701,
      "balance_loss_clip": 1.08804035,
      "balance_loss_mlp": 1.03126228,
      "epoch": 0.07082306258642458,
      "flos": 23653568528640.0,
      "grad_norm": 2.7890172388862853,
      "language_loss": 0.86819124,
      "learning_rate": 3.982591887763199e-06,
      "loss": 0.89150995,
      "num_input_tokens_seen": 12584785,
      "step": 589,
      "time_per_iteration": 3.27860951423645
    },
    {
      "auxiliary_loss_clip": 0.01261714,
      "auxiliary_loss_mlp": 0.01045195,
      "balance_loss_clip": 1.0819931,
      "balance_loss_mlp": 1.03062773,
      "epoch": 0.07094330547706366,
      "flos": 13880049408000.0,
      "grad_norm": 2.3234032101705053,
      "language_loss": 0.81785154,
      "learning_rate": 3.982489183908316e-06,
      "loss": 0.84092057,
      "num_input_tokens_seen": 12601205,
      "step": 590,
      "time_per_iteration": 3.245854139328003
    },
    {
      "auxiliary_loss_clip": 0.01135806,
      "auxiliary_loss_mlp": 0.01043473,
      "balance_loss_clip": 1.06261027,
      "balance_loss_mlp": 1.02976978,
      "epoch": 0.07106354836770276,
      "flos": 24645098534400.0,
      "grad_norm": 1.7585913938003184,
      "language_loss": 0.84410793,
      "learning_rate": 3.982386179311399e-06,
      "loss": 0.8659007,
      "num_input_tokens_seen": 12621725,
      "step": 591,
      "time_per_iteration": 2.8197999000549316
    },
    {
      "auxiliary_loss_clip": 0.0125133,
      "auxiliary_loss_mlp": 0.0104483,
      "balance_loss_clip": 1.08335447,
      "balance_loss_mlp": 1.02858746,
      "epoch": 0.07118379125834184,
      "flos": 16217223649920.0,
      "grad_norm": 3.2762404625596493,
      "language_loss": 0.87671924,
      "learning_rate": 3.982282873988075e-06,
      "loss": 0.89968085,
      "num_input_tokens_seen": 12639600,
      "step": 592,
      "time_per_iteration": 2.814849615097046
    },
    {
      "auxiliary_loss_clip": 0.01262241,
      "auxiliary_loss_mlp": 0.01034629,
      "balance_loss_clip": 1.08487558,
      "balance_loss_mlp": 1.02037108,
      "epoch": 0.07130403414898094,
      "flos": 19719986227200.0,
      "grad_norm": 1.7017158837395339,
      "language_loss": 0.86930197,
      "learning_rate": 3.982179267954016e-06,
      "loss": 0.89227068,
      "num_input_tokens_seen": 12660030,
      "step": 593,
      "time_per_iteration": 3.3470213413238525
    },
    {
      "auxiliary_loss_clip": 0.01199116,
      "auxiliary_loss_mlp": 0.01048463,
      "balance_loss_clip": 1.0600481,
      "balance_loss_mlp": 1.03283501,
      "epoch": 0.07142427703962004,
      "flos": 21871933009920.0,
      "grad_norm": 2.4315618417317375,
      "language_loss": 0.96389061,
      "learning_rate": 3.982075361224937e-06,
      "loss": 0.98636645,
      "num_input_tokens_seen": 12678395,
      "step": 594,
      "time_per_iteration": 3.430814743041992
    },
    {
      "auxiliary_loss_clip": 0.01228407,
      "auxiliary_loss_mlp": 0.0105112,
      "balance_loss_clip": 1.0786581,
      "balance_loss_mlp": 1.03593886,
      "epoch": 0.07154451993025912,
      "flos": 18296595002880.0,
      "grad_norm": 2.8114301107512825,
      "language_loss": 0.87912166,
      "learning_rate": 3.981971153816602e-06,
      "loss": 0.90191692,
      "num_input_tokens_seen": 12696000,
      "step": 595,
      "time_per_iteration": 2.5545411109924316
    },
    {
      "auxiliary_loss_clip": 0.01081894,
      "auxiliary_loss_mlp": 0.00766207,
      "balance_loss_clip": 1.04674459,
      "balance_loss_mlp": 1.00135827,
      "epoch": 0.07166476282089822,
      "flos": 22160690444160.0,
      "grad_norm": 1.635146056584503,
      "language_loss": 0.9610914,
      "learning_rate": 3.981866645744819e-06,
      "loss": 0.97957242,
      "num_input_tokens_seen": 12716715,
      "step": 596,
      "time_per_iteration": 3.2321794033050537
    },
    {
      "auxiliary_loss_clip": 0.01266742,
      "auxiliary_loss_mlp": 0.01038345,
      "balance_loss_clip": 1.08034909,
      "balance_loss_mlp": 1.02356303,
      "epoch": 0.0717850057115373,
      "flos": 14136343925760.0,
      "grad_norm": 3.059058282940376,
      "language_loss": 0.81561458,
      "learning_rate": 3.9817618370254416e-06,
      "loss": 0.83866549,
      "num_input_tokens_seen": 12733370,
      "step": 597,
      "time_per_iteration": 3.0622196197509766
    },
    {
      "auxiliary_loss_clip": 0.01235527,
      "auxiliary_loss_mlp": 0.01042061,
      "balance_loss_clip": 1.08069265,
      "balance_loss_mlp": 1.02683771,
      "epoch": 0.0719052486021764,
      "flos": 30917794412160.0,
      "grad_norm": 2.1497108243026513,
      "language_loss": 0.87479627,
      "learning_rate": 3.9816567276743684e-06,
      "loss": 0.89757222,
      "num_input_tokens_seen": 12753235,
      "step": 598,
      "time_per_iteration": 2.6984753608703613
    },
    {
      "auxiliary_loss_clip": 0.01241589,
      "auxiliary_loss_mlp": 0.01046277,
      "balance_loss_clip": 1.08110809,
      "balance_loss_mlp": 1.02982605,
      "epoch": 0.0720254914928155,
      "flos": 21287019939840.0,
      "grad_norm": 1.972379307293337,
      "language_loss": 0.77684665,
      "learning_rate": 3.9815513177075466e-06,
      "loss": 0.79972529,
      "num_input_tokens_seen": 12772020,
      "step": 599,
      "time_per_iteration": 2.581003427505493
    },
    {
      "auxiliary_loss_clip": 0.01212322,
      "auxiliary_loss_mlp": 0.01040096,
      "balance_loss_clip": 1.0766952,
      "balance_loss_mlp": 1.02690518,
      "epoch": 0.07214573438345458,
      "flos": 27819170732160.0,
      "grad_norm": 1.5259347279786224,
      "language_loss": 0.70251083,
      "learning_rate": 3.9814456071409646e-06,
      "loss": 0.72503507,
      "num_input_tokens_seen": 12792555,
      "step": 600,
      "time_per_iteration": 2.6900675296783447
    },
    {
      "auxiliary_loss_clip": 0.01186904,
      "auxiliary_loss_mlp": 0.01051624,
      "balance_loss_clip": 1.06950283,
      "balance_loss_mlp": 1.0334084,
      "epoch": 0.07226597727409367,
      "flos": 25483576688640.0,
      "grad_norm": 2.6017697965718747,
      "language_loss": 0.85200006,
      "learning_rate": 3.981339595990659e-06,
      "loss": 0.87438536,
      "num_input_tokens_seen": 12811085,
      "step": 601,
      "time_per_iteration": 2.688391923904419
    },
    {
      "auxiliary_loss_clip": 0.01184954,
      "auxiliary_loss_mlp": 0.00767097,
      "balance_loss_clip": 1.06208491,
      "balance_loss_mlp": 1.00134885,
      "epoch": 0.07238622016473276,
      "flos": 23513840622720.0,
      "grad_norm": 2.119869340435436,
      "language_loss": 0.81179631,
      "learning_rate": 3.981233284272713e-06,
      "loss": 0.83131683,
      "num_input_tokens_seen": 12830830,
      "step": 602,
      "time_per_iteration": 2.655672788619995
    },
    {
      "auxiliary_loss_clip": 0.01236671,
      "auxiliary_loss_mlp": 0.01042285,
      "balance_loss_clip": 1.07966614,
      "balance_loss_mlp": 1.02889156,
      "epoch": 0.07250646305537185,
      "flos": 25453519983360.0,
      "grad_norm": 1.6005062461459285,
      "language_loss": 0.90102696,
      "learning_rate": 3.981126672003253e-06,
      "loss": 0.92381656,
      "num_input_tokens_seen": 12853505,
      "step": 603,
      "time_per_iteration": 2.6580076217651367
    },
    {
      "auxiliary_loss_clip": 0.01234837,
      "auxiliary_loss_mlp": 0.01041543,
      "balance_loss_clip": 1.07395899,
      "balance_loss_mlp": 1.02732086,
      "epoch": 0.07262670594601094,
      "flos": 27155038216320.0,
      "grad_norm": 2.5634052191796233,
      "language_loss": 0.78097296,
      "learning_rate": 3.981019759198451e-06,
      "loss": 0.80373675,
      "num_input_tokens_seen": 12872455,
      "step": 604,
      "time_per_iteration": 2.5939154624938965
    },
    {
      "auxiliary_loss_clip": 0.0124211,
      "auxiliary_loss_mlp": 0.0104543,
      "balance_loss_clip": 1.07976699,
      "balance_loss_mlp": 1.0314703,
      "epoch": 0.07274694883665003,
      "flos": 26651607148800.0,
      "grad_norm": 2.9302431753920994,
      "language_loss": 0.84390652,
      "learning_rate": 3.980912545874528e-06,
      "loss": 0.86678195,
      "num_input_tokens_seen": 12892620,
      "step": 605,
      "time_per_iteration": 2.597200870513916
    },
    {
      "auxiliary_loss_clip": 0.01203865,
      "auxiliary_loss_mlp": 0.0104228,
      "balance_loss_clip": 1.06834507,
      "balance_loss_mlp": 1.02574539,
      "epoch": 0.07286719172728913,
      "flos": 29862344154240.0,
      "grad_norm": 3.114712733753547,
      "language_loss": 0.8541494,
      "learning_rate": 3.980805032047746e-06,
      "loss": 0.87661088,
      "num_input_tokens_seen": 12914090,
      "step": 606,
      "time_per_iteration": 2.687706470489502
    },
    {
      "auxiliary_loss_clip": 0.01280696,
      "auxiliary_loss_mlp": 0.01045895,
      "balance_loss_clip": 1.085747,
      "balance_loss_mlp": 1.03039801,
      "epoch": 0.07298743461792821,
      "flos": 17382057799680.0,
      "grad_norm": 1.87783437643763,
      "language_loss": 0.81104207,
      "learning_rate": 3.980697217734415e-06,
      "loss": 0.83430803,
      "num_input_tokens_seen": 12931830,
      "step": 607,
      "time_per_iteration": 2.4798269271850586
    },
    {
      "auxiliary_loss_clip": 0.0120923,
      "auxiliary_loss_mlp": 0.01045297,
      "balance_loss_clip": 1.0744288,
      "balance_loss_mlp": 1.02942467,
      "epoch": 0.07310767750856731,
      "flos": 19498201701120.0,
      "grad_norm": 2.0865497191079654,
      "language_loss": 0.91850829,
      "learning_rate": 3.980589102950891e-06,
      "loss": 0.94105357,
      "num_input_tokens_seen": 12949995,
      "step": 608,
      "time_per_iteration": 2.591695547103882
    },
    {
      "auxiliary_loss_clip": 0.01215767,
      "auxiliary_loss_mlp": 0.0104453,
      "balance_loss_clip": 1.07636571,
      "balance_loss_mlp": 1.02927721,
      "epoch": 0.07322792039920639,
      "flos": 29168693637120.0,
      "grad_norm": 2.420670452012399,
      "language_loss": 0.75966442,
      "learning_rate": 3.9804806877135755e-06,
      "loss": 0.78226733,
      "num_input_tokens_seen": 12968040,
      "step": 609,
      "time_per_iteration": 2.6427059173583984
    },
    {
      "auxiliary_loss_clip": 0.01245438,
      "auxiliary_loss_mlp": 0.01048329,
      "balance_loss_clip": 1.07740319,
      "balance_loss_mlp": 1.0312469,
      "epoch": 0.07334816328984549,
      "flos": 23477822259840.0,
      "grad_norm": 3.7421650927216334,
      "language_loss": 0.860834,
      "learning_rate": 3.980371972038915e-06,
      "loss": 0.88377172,
      "num_input_tokens_seen": 12988530,
      "step": 610,
      "time_per_iteration": 2.565213203430176
    },
    {
      "auxiliary_loss_clip": 0.01183724,
      "auxiliary_loss_mlp": 0.01050688,
      "balance_loss_clip": 1.06211925,
      "balance_loss_mlp": 1.03410053,
      "epoch": 0.07346840618048459,
      "flos": 22962467877120.0,
      "grad_norm": 2.3145890978218095,
      "language_loss": 0.84260511,
      "learning_rate": 3.980262955943399e-06,
      "loss": 0.86494929,
      "num_input_tokens_seen": 13008195,
      "step": 611,
      "time_per_iteration": 2.645430088043213
    },
    {
      "auxiliary_loss_clip": 0.01239868,
      "auxiliary_loss_mlp": 0.00766503,
      "balance_loss_clip": 1.08041739,
      "balance_loss_mlp": 1.00127006,
      "epoch": 0.07358864907112367,
      "flos": 17673903803520.0,
      "grad_norm": 2.4794645660837804,
      "language_loss": 0.87277627,
      "learning_rate": 3.980153639443569e-06,
      "loss": 0.89283997,
      "num_input_tokens_seen": 13024180,
      "step": 612,
      "time_per_iteration": 2.5388715267181396
    },
    {
      "auxiliary_loss_clip": 0.01254307,
      "auxiliary_loss_mlp": 0.01051526,
      "balance_loss_clip": 1.08563602,
      "balance_loss_mlp": 1.03552818,
      "epoch": 0.07370889196176277,
      "flos": 24097029840000.0,
      "grad_norm": 3.2080115781781715,
      "language_loss": 0.80201423,
      "learning_rate": 3.980044022556005e-06,
      "loss": 0.82507253,
      "num_input_tokens_seen": 13043865,
      "step": 613,
      "time_per_iteration": 2.6146039962768555
    },
    {
      "auxiliary_loss_clip": 0.01174113,
      "auxiliary_loss_mlp": 0.01059448,
      "balance_loss_clip": 1.05519176,
      "balance_loss_mlp": 1.04364038,
      "epoch": 0.07382913485240185,
      "flos": 25885919905920.0,
      "grad_norm": 2.3068080527016064,
      "language_loss": 0.73307502,
      "learning_rate": 3.9799341052973375e-06,
      "loss": 0.75541067,
      "num_input_tokens_seen": 13063700,
      "step": 614,
      "time_per_iteration": 2.6965079307556152
    },
    {
      "auxiliary_loss_clip": 0.01192267,
      "auxiliary_loss_mlp": 0.01053516,
      "balance_loss_clip": 1.07572567,
      "balance_loss_mlp": 1.03670752,
      "epoch": 0.07394937774304094,
      "flos": 16873850223360.0,
      "grad_norm": 2.6780365073593755,
      "language_loss": 0.7520749,
      "learning_rate": 3.979823887684241e-06,
      "loss": 0.77453274,
      "num_input_tokens_seen": 13082640,
      "step": 615,
      "time_per_iteration": 3.7841646671295166
    },
    {
      "auxiliary_loss_clip": 0.01230909,
      "auxiliary_loss_mlp": 0.01052059,
      "balance_loss_clip": 1.077384,
      "balance_loss_mlp": 1.03716993,
      "epoch": 0.07406962063368003,
      "flos": 20703471586560.0,
      "grad_norm": 2.7508048771826084,
      "language_loss": 0.8488096,
      "learning_rate": 3.979713369733434e-06,
      "loss": 0.87163925,
      "num_input_tokens_seen": 13100505,
      "step": 616,
      "time_per_iteration": 3.5927348136901855
    },
    {
      "auxiliary_loss_clip": 0.01268454,
      "auxiliary_loss_mlp": 0.0076733,
      "balance_loss_clip": 1.08554077,
      "balance_loss_mlp": 1.00089693,
      "epoch": 0.07418986352431912,
      "flos": 21430985650560.0,
      "grad_norm": 2.3798002377408096,
      "language_loss": 0.84755027,
      "learning_rate": 3.979602551461683e-06,
      "loss": 0.86790812,
      "num_input_tokens_seen": 13121285,
      "step": 617,
      "time_per_iteration": 2.5586040019989014
    },
    {
      "auxiliary_loss_clip": 0.0117858,
      "auxiliary_loss_mlp": 0.0105198,
      "balance_loss_clip": 1.07326531,
      "balance_loss_mlp": 1.03577387,
      "epoch": 0.07431010641495822,
      "flos": 12021133777920.0,
      "grad_norm": 2.4502075290203345,
      "language_loss": 0.91751099,
      "learning_rate": 3.979491432885799e-06,
      "loss": 0.93981659,
      "num_input_tokens_seen": 13137550,
      "step": 618,
      "time_per_iteration": 2.7065064907073975
    },
    {
      "auxiliary_loss_clip": 0.01252658,
      "auxiliary_loss_mlp": 0.01041947,
      "balance_loss_clip": 1.07711267,
      "balance_loss_mlp": 1.0277319,
      "epoch": 0.0744303493055973,
      "flos": 20957575374720.0,
      "grad_norm": 8.169014160110951,
      "language_loss": 0.82887888,
      "learning_rate": 3.97938001402264e-06,
      "loss": 0.85182494,
      "num_input_tokens_seen": 13156675,
      "step": 619,
      "time_per_iteration": 3.351684808731079
    },
    {
      "auxiliary_loss_clip": 0.01281154,
      "auxiliary_loss_mlp": 0.01053609,
      "balance_loss_clip": 1.08559763,
      "balance_loss_mlp": 1.03813577,
      "epoch": 0.0745505921962364,
      "flos": 16253134272000.0,
      "grad_norm": 2.8034279431387588,
      "language_loss": 0.80045718,
      "learning_rate": 3.979268294889105e-06,
      "loss": 0.8238048,
      "num_input_tokens_seen": 13172225,
      "step": 620,
      "time_per_iteration": 3.167755603790283
    },
    {
      "auxiliary_loss_clip": 0.012369,
      "auxiliary_loss_mlp": 0.01050249,
      "balance_loss_clip": 1.0795989,
      "balance_loss_mlp": 1.03537154,
      "epoch": 0.07467083508687548,
      "flos": 50944635550080.0,
      "grad_norm": 2.092820584829215,
      "language_loss": 0.7394129,
      "learning_rate": 3.979156275502143e-06,
      "loss": 0.7622844,
      "num_input_tokens_seen": 13195885,
      "step": 621,
      "time_per_iteration": 2.8253026008605957
    },
    {
      "auxiliary_loss_clip": 0.01230035,
      "auxiliary_loss_mlp": 0.01044479,
      "balance_loss_clip": 1.07786489,
      "balance_loss_mlp": 1.02806377,
      "epoch": 0.07479107797751458,
      "flos": 17529686697600.0,
      "grad_norm": 2.536132353065517,
      "language_loss": 0.91568762,
      "learning_rate": 3.979043955878749e-06,
      "loss": 0.93843275,
      "num_input_tokens_seen": 13213730,
      "step": 622,
      "time_per_iteration": 2.5638885498046875
    },
    {
      "auxiliary_loss_clip": 0.0123383,
      "auxiliary_loss_mlp": 0.01052641,
      "balance_loss_clip": 1.0816648,
      "balance_loss_mlp": 1.03644085,
      "epoch": 0.07491132086815366,
      "flos": 23473943591040.0,
      "grad_norm": 2.5018627002386835,
      "language_loss": 0.83507299,
      "learning_rate": 3.978931336035959e-06,
      "loss": 0.85793775,
      "num_input_tokens_seen": 13232540,
      "step": 623,
      "time_per_iteration": 2.619553565979004
    },
    {
      "auxiliary_loss_clip": 0.01230271,
      "auxiliary_loss_mlp": 0.01055942,
      "balance_loss_clip": 1.07978106,
      "balance_loss_mlp": 1.04031336,
      "epoch": 0.07503156375879276,
      "flos": 20157557708160.0,
      "grad_norm": 2.4787330114297355,
      "language_loss": 0.82427526,
      "learning_rate": 3.9788184159908595e-06,
      "loss": 0.84713733,
      "num_input_tokens_seen": 13249670,
      "step": 624,
      "time_per_iteration": 2.5722475051879883
    },
    {
      "auxiliary_loss_clip": 0.0124288,
      "auxiliary_loss_mlp": 0.01051912,
      "balance_loss_clip": 1.07836187,
      "balance_loss_mlp": 1.03786862,
      "epoch": 0.07515180664943186,
      "flos": 15115519653120.0,
      "grad_norm": 3.168768873325545,
      "language_loss": 0.82297802,
      "learning_rate": 3.97870519576058e-06,
      "loss": 0.84592593,
      "num_input_tokens_seen": 13266095,
      "step": 625,
      "time_per_iteration": 2.5070040225982666
    },
    {
      "auxiliary_loss_clip": 0.01198456,
      "auxiliary_loss_mlp": 0.01042081,
      "balance_loss_clip": 1.07643592,
      "balance_loss_mlp": 1.02609515,
      "epoch": 0.07527204954007094,
      "flos": 21287702298240.0,
      "grad_norm": 2.5575161717589303,
      "language_loss": 0.80944955,
      "learning_rate": 3.978591675362295e-06,
      "loss": 0.83185494,
      "num_input_tokens_seen": 13284810,
      "step": 626,
      "time_per_iteration": 2.672092914581299
    },
    {
      "auxiliary_loss_clip": 0.01242019,
      "auxiliary_loss_mlp": 0.00766852,
      "balance_loss_clip": 1.08344269,
      "balance_loss_mlp": 1.00107563,
      "epoch": 0.07539229243071004,
      "flos": 21324187537920.0,
      "grad_norm": 1.9266768724374617,
      "language_loss": 0.87477702,
      "learning_rate": 3.978477854813226e-06,
      "loss": 0.89486569,
      "num_input_tokens_seen": 13304150,
      "step": 627,
      "time_per_iteration": 2.5666065216064453
    },
    {
      "auxiliary_loss_clip": 0.01223835,
      "auxiliary_loss_mlp": 0.01050127,
      "balance_loss_clip": 1.0670259,
      "balance_loss_mlp": 1.03501129,
      "epoch": 0.07551253532134912,
      "flos": 13042540920960.0,
      "grad_norm": 2.551984454781684,
      "language_loss": 0.82996821,
      "learning_rate": 3.97836373413064e-06,
      "loss": 0.85270786,
      "num_input_tokens_seen": 13322205,
      "step": 628,
      "time_per_iteration": 2.5666394233703613
    },
    {
      "auxiliary_loss_clip": 0.0125727,
      "auxiliary_loss_mlp": 0.01042894,
      "balance_loss_clip": 1.07959139,
      "balance_loss_mlp": 1.02740908,
      "epoch": 0.07563277821198822,
      "flos": 19208761908480.0,
      "grad_norm": 2.1318259326306035,
      "language_loss": 0.7491402,
      "learning_rate": 3.978249313331848e-06,
      "loss": 0.77214187,
      "num_input_tokens_seen": 13340435,
      "step": 629,
      "time_per_iteration": 2.485793352127075
    },
    {
      "auxiliary_loss_clip": 0.01213272,
      "auxiliary_loss_mlp": 0.01053378,
      "balance_loss_clip": 1.07703292,
      "balance_loss_mlp": 1.03745115,
      "epoch": 0.07575302110262731,
      "flos": 19537200892800.0,
      "grad_norm": 3.4803141095646954,
      "language_loss": 0.62019914,
      "learning_rate": 3.978134592434208e-06,
      "loss": 0.64286566,
      "num_input_tokens_seen": 13358185,
      "step": 630,
      "time_per_iteration": 2.618222713470459
    },
    {
      "auxiliary_loss_clip": 0.01169133,
      "auxiliary_loss_mlp": 0.01008808,
      "balance_loss_clip": 1.0636878,
      "balance_loss_mlp": 1.00506449,
      "epoch": 0.0758732639932664,
      "flos": 67961808017280.0,
      "grad_norm": 1.0191677992111439,
      "language_loss": 0.59370601,
      "learning_rate": 3.978019571455123e-06,
      "loss": 0.61548543,
      "num_input_tokens_seen": 13410130,
      "step": 631,
      "time_per_iteration": 3.1424264907836914
    },
    {
      "auxiliary_loss_clip": 0.01247305,
      "auxiliary_loss_mlp": 0.01054208,
      "balance_loss_clip": 1.08161163,
      "balance_loss_mlp": 1.04040337,
      "epoch": 0.07599350688390549,
      "flos": 18989204025600.0,
      "grad_norm": 2.3727535573168153,
      "language_loss": 0.84030771,
      "learning_rate": 3.977904250412042e-06,
      "loss": 0.86332285,
      "num_input_tokens_seen": 13429085,
      "step": 632,
      "time_per_iteration": 2.5692529678344727
    },
    {
      "auxiliary_loss_clip": 0.01220415,
      "auxiliary_loss_mlp": 0.00766088,
      "balance_loss_clip": 1.0791167,
      "balance_loss_mlp": 1.00092983,
      "epoch": 0.07611374977454458,
      "flos": 21069006341760.0,
      "grad_norm": 2.187478602508997,
      "language_loss": 0.85545057,
      "learning_rate": 3.97778862932246e-06,
      "loss": 0.87531561,
      "num_input_tokens_seen": 13446250,
      "step": 633,
      "time_per_iteration": 2.6271727085113525
    },
    {
      "auxiliary_loss_clip": 0.01229989,
      "auxiliary_loss_mlp": 0.01049354,
      "balance_loss_clip": 1.0769341,
      "balance_loss_mlp": 1.03336823,
      "epoch": 0.07623399266518367,
      "flos": 18514536773760.0,
      "grad_norm": 2.4239364394743594,
      "language_loss": 0.94105154,
      "learning_rate": 3.9776727082039144e-06,
      "loss": 0.96384501,
      "num_input_tokens_seen": 13463220,
      "step": 634,
      "time_per_iteration": 2.5959224700927734
    },
    {
      "auxiliary_loss_clip": 0.01092389,
      "auxiliary_loss_mlp": 0.01004113,
      "balance_loss_clip": 1.05517733,
      "balance_loss_mlp": 1.00022697,
      "epoch": 0.07635423555582276,
      "flos": 44663036077440.0,
      "grad_norm": 0.8279002678594178,
      "language_loss": 0.55464602,
      "learning_rate": 3.977556487073991e-06,
      "loss": 0.575611,
      "num_input_tokens_seen": 13517775,
      "step": 635,
      "time_per_iteration": 3.184572219848633
    },
    {
      "auxiliary_loss_clip": 0.01231141,
      "auxiliary_loss_mlp": 0.00766248,
      "balance_loss_clip": 1.07538617,
      "balance_loss_mlp": 1.00104117,
      "epoch": 0.07647447844646185,
      "flos": 21761148487680.0,
      "grad_norm": 1.8121354504329674,
      "language_loss": 0.81740677,
      "learning_rate": 3.97743996595032e-06,
      "loss": 0.83738065,
      "num_input_tokens_seen": 13537815,
      "step": 636,
      "time_per_iteration": 2.644584894180298
    },
    {
      "auxiliary_loss_clip": 0.0124269,
      "auxiliary_loss_mlp": 0.01045528,
      "balance_loss_clip": 1.07798207,
      "balance_loss_mlp": 1.03015554,
      "epoch": 0.07659472133710095,
      "flos": 23806799948160.0,
      "grad_norm": 2.159696360037236,
      "language_loss": 0.81692743,
      "learning_rate": 3.9773231448505804e-06,
      "loss": 0.83980966,
      "num_input_tokens_seen": 13559605,
      "step": 637,
      "time_per_iteration": 2.5915029048919678
    },
    {
      "auxiliary_loss_clip": 0.01205569,
      "auxiliary_loss_mlp": 0.01047854,
      "balance_loss_clip": 1.07365894,
      "balance_loss_mlp": 1.03166533,
      "epoch": 0.07671496422774003,
      "flos": 21469984842240.0,
      "grad_norm": 1.9369664300449505,
      "language_loss": 0.78289557,
      "learning_rate": 3.977206023792491e-06,
      "loss": 0.80542982,
      "num_input_tokens_seen": 13579495,
      "step": 638,
      "time_per_iteration": 2.7078781127929688
    },
    {
      "auxiliary_loss_clip": 0.01221634,
      "auxiliary_loss_mlp": 0.00767311,
      "balance_loss_clip": 1.07454228,
      "balance_loss_mlp": 1.00085735,
      "epoch": 0.07683520711837913,
      "flos": 16980971558400.0,
      "grad_norm": 3.1177912031084056,
      "language_loss": 0.81092072,
      "learning_rate": 3.97708860279382e-06,
      "loss": 0.83081013,
      "num_input_tokens_seen": 13597605,
      "step": 639,
      "time_per_iteration": 2.565375328063965
    },
    {
      "auxiliary_loss_clip": 0.01209584,
      "auxiliary_loss_mlp": 0.01053481,
      "balance_loss_clip": 1.0725615,
      "balance_loss_mlp": 1.03895545,
      "epoch": 0.07695545000901821,
      "flos": 23476744851840.0,
      "grad_norm": 1.7292302990193626,
      "language_loss": 0.78343052,
      "learning_rate": 3.97697088187238e-06,
      "loss": 0.80606115,
      "num_input_tokens_seen": 13618120,
      "step": 640,
      "time_per_iteration": 3.3966495990753174
    },
    {
      "auxiliary_loss_clip": 0.01261614,
      "auxiliary_loss_mlp": 0.01045256,
      "balance_loss_clip": 1.08434248,
      "balance_loss_mlp": 1.03048015,
      "epoch": 0.07707569289965731,
      "flos": 17634258167040.0,
      "grad_norm": 4.394800687913413,
      "language_loss": 0.91678715,
      "learning_rate": 3.976852861046029e-06,
      "loss": 0.93985587,
      "num_input_tokens_seen": 13634735,
      "step": 641,
      "time_per_iteration": 2.523347854614258
    },
    {
      "auxiliary_loss_clip": 0.01256427,
      "auxiliary_loss_mlp": 0.01039347,
      "balance_loss_clip": 1.08262515,
      "balance_loss_mlp": 1.02523255,
      "epoch": 0.0771959357902964,
      "flos": 25775674087680.0,
      "grad_norm": 1.749358859574003,
      "language_loss": 0.80343997,
      "learning_rate": 3.97673454033267e-06,
      "loss": 0.82639766,
      "num_input_tokens_seen": 13656835,
      "step": 642,
      "time_per_iteration": 3.486616611480713
    },
    {
      "auxiliary_loss_clip": 0.01242974,
      "auxiliary_loss_mlp": 0.00766359,
      "balance_loss_clip": 1.08104897,
      "balance_loss_mlp": 1.00081122,
      "epoch": 0.07731617868093549,
      "flos": 19828651847040.0,
      "grad_norm": 1.9870373644815003,
      "language_loss": 0.83003801,
      "learning_rate": 3.976615919750254e-06,
      "loss": 0.85013127,
      "num_input_tokens_seen": 13674535,
      "step": 643,
      "time_per_iteration": 2.5797760486602783
    },
    {
      "auxiliary_loss_clip": 0.01204549,
      "auxiliary_loss_mlp": 0.01052631,
      "balance_loss_clip": 1.06188583,
      "balance_loss_mlp": 1.03608513,
      "epoch": 0.07743642157157458,
      "flos": 21324654414720.0,
      "grad_norm": 2.1341183782696636,
      "language_loss": 0.87103212,
      "learning_rate": 3.976496999316775e-06,
      "loss": 0.89360392,
      "num_input_tokens_seen": 13693290,
      "step": 644,
      "time_per_iteration": 3.420043468475342
    },
    {
      "auxiliary_loss_clip": 0.01193657,
      "auxiliary_loss_mlp": 0.01050423,
      "balance_loss_clip": 1.06853545,
      "balance_loss_mlp": 1.03489554,
      "epoch": 0.07755666446221367,
      "flos": 19969133938560.0,
      "grad_norm": 2.0269959076703397,
      "language_loss": 0.84154165,
      "learning_rate": 3.976377779050271e-06,
      "loss": 0.86398244,
      "num_input_tokens_seen": 13711420,
      "step": 645,
      "time_per_iteration": 2.597649335861206
    },
    {
      "auxiliary_loss_clip": 0.01231713,
      "auxiliary_loss_mlp": 0.0104262,
      "balance_loss_clip": 1.07569778,
      "balance_loss_mlp": 1.02726638,
      "epoch": 0.07767690735285276,
      "flos": 23623224514560.0,
      "grad_norm": 2.300492457996893,
      "language_loss": 0.84550393,
      "learning_rate": 3.976258258968831e-06,
      "loss": 0.86824727,
      "num_input_tokens_seen": 13729965,
      "step": 646,
      "time_per_iteration": 3.404906749725342
    },
    {
      "auxiliary_loss_clip": 0.01248373,
      "auxiliary_loss_mlp": 0.01037038,
      "balance_loss_clip": 1.07943082,
      "balance_loss_mlp": 1.02317381,
      "epoch": 0.07779715024349185,
      "flos": 22236246702720.0,
      "grad_norm": 2.4888158860088585,
      "language_loss": 0.74483252,
      "learning_rate": 3.976138439090583e-06,
      "loss": 0.76768661,
      "num_input_tokens_seen": 13748045,
      "step": 647,
      "time_per_iteration": 2.552244186401367
    },
    {
      "auxiliary_loss_clip": 0.01140549,
      "auxiliary_loss_mlp": 0.01049641,
      "balance_loss_clip": 1.06165314,
      "balance_loss_mlp": 1.0345434,
      "epoch": 0.07791739313413094,
      "flos": 20955097336320.0,
      "grad_norm": 2.2112096500019933,
      "language_loss": 0.85192668,
      "learning_rate": 3.976018319433706e-06,
      "loss": 0.87382853,
      "num_input_tokens_seen": 13765590,
      "step": 648,
      "time_per_iteration": 2.7398879528045654
    },
    {
      "auxiliary_loss_clip": 0.01261144,
      "auxiliary_loss_mlp": 0.01052379,
      "balance_loss_clip": 1.08423924,
      "balance_loss_mlp": 1.03590417,
      "epoch": 0.07803763602477004,
      "flos": 19312327797120.0,
      "grad_norm": 6.192774691371939,
      "language_loss": 0.9163512,
      "learning_rate": 3.9758979000164205e-06,
      "loss": 0.9394865,
      "num_input_tokens_seen": 13782410,
      "step": 649,
      "time_per_iteration": 2.5440762042999268
    },
    {
      "auxiliary_loss_clip": 0.01219742,
      "auxiliary_loss_mlp": 0.01049669,
      "balance_loss_clip": 1.07482374,
      "balance_loss_mlp": 1.03388548,
      "epoch": 0.07815787891540912,
      "flos": 22710806213760.0,
      "grad_norm": 1.7457581768674681,
      "language_loss": 0.72045922,
      "learning_rate": 3.975777180856995e-06,
      "loss": 0.74315339,
      "num_input_tokens_seen": 13801530,
      "step": 650,
      "time_per_iteration": 2.6364293098449707
    },
    {
      "auxiliary_loss_clip": 0.01230247,
      "auxiliary_loss_mlp": 0.01046802,
      "balance_loss_clip": 1.07393837,
      "balance_loss_mlp": 1.03136992,
      "epoch": 0.07827812180604822,
      "flos": 22711129436160.0,
      "grad_norm": 2.4577496404138146,
      "language_loss": 0.86311078,
      "learning_rate": 3.975656161973742e-06,
      "loss": 0.88588125,
      "num_input_tokens_seen": 13820615,
      "step": 651,
      "time_per_iteration": 2.5760366916656494
    },
    {
      "auxiliary_loss_clip": 0.01228356,
      "auxiliary_loss_mlp": 0.0105071,
      "balance_loss_clip": 1.07640934,
      "balance_loss_mlp": 1.03403318,
      "epoch": 0.0783983646966873,
      "flos": 21725597001600.0,
      "grad_norm": 4.272970745663678,
      "language_loss": 0.88790321,
      "learning_rate": 3.9755348433850194e-06,
      "loss": 0.91069388,
      "num_input_tokens_seen": 13835955,
      "step": 652,
      "time_per_iteration": 2.6253082752227783
    },
    {
      "auxiliary_loss_clip": 0.01122349,
      "auxiliary_loss_mlp": 0.01005189,
      "balance_loss_clip": 1.05726814,
      "balance_loss_mlp": 1.00166082,
      "epoch": 0.0785186075873264,
      "flos": 60640877537280.0,
      "grad_norm": 0.9593780850681785,
      "language_loss": 0.6362325,
      "learning_rate": 3.975413225109232e-06,
      "loss": 0.6575079,
      "num_input_tokens_seen": 13896505,
      "step": 653,
      "time_per_iteration": 3.2686941623687744
    },
    {
      "auxiliary_loss_clip": 0.0125568,
      "auxiliary_loss_mlp": 0.01052728,
      "balance_loss_clip": 1.07990646,
      "balance_loss_mlp": 1.03631258,
      "epoch": 0.0786388504779655,
      "flos": 23877902920320.0,
      "grad_norm": 10.239619805885095,
      "language_loss": 0.93988347,
      "learning_rate": 3.975291307164829e-06,
      "loss": 0.96296751,
      "num_input_tokens_seen": 13915150,
      "step": 654,
      "time_per_iteration": 2.5503876209259033
    },
    {
      "auxiliary_loss_clip": 0.0122698,
      "auxiliary_loss_mlp": 0.01044813,
      "balance_loss_clip": 1.07411909,
      "balance_loss_mlp": 1.03144956,
      "epoch": 0.07875909336860458,
      "flos": 15158684822400.0,
      "grad_norm": 1.945864336209593,
      "language_loss": 0.85308766,
      "learning_rate": 3.975169089570306e-06,
      "loss": 0.87580556,
      "num_input_tokens_seen": 13933525,
      "step": 655,
      "time_per_iteration": 2.557859182357788
    },
    {
      "auxiliary_loss_clip": 0.0127009,
      "auxiliary_loss_mlp": 0.01053424,
      "balance_loss_clip": 1.0818125,
      "balance_loss_mlp": 1.03838611,
      "epoch": 0.07887933625924368,
      "flos": 22236857233920.0,
      "grad_norm": 2.014600057002975,
      "language_loss": 0.9170264,
      "learning_rate": 3.975046572344202e-06,
      "loss": 0.94026154,
      "num_input_tokens_seen": 13949985,
      "step": 656,
      "time_per_iteration": 2.483473300933838
    },
    {
      "auxiliary_loss_clip": 0.01239895,
      "auxiliary_loss_mlp": 0.01037346,
      "balance_loss_clip": 1.0768913,
      "balance_loss_mlp": 1.02325535,
      "epoch": 0.07899957914988276,
      "flos": 20777734955520.0,
      "grad_norm": 2.0228625820250454,
      "language_loss": 0.7130177,
      "learning_rate": 3.974923755505103e-06,
      "loss": 0.73579013,
      "num_input_tokens_seen": 13969215,
      "step": 657,
      "time_per_iteration": 2.5664467811584473
    },
    {
      "auxiliary_loss_clip": 0.01184116,
      "auxiliary_loss_mlp": 0.00766681,
      "balance_loss_clip": 1.06833649,
      "balance_loss_mlp": 1.0013541,
      "epoch": 0.07911982204052186,
      "flos": 23003047267200.0,
      "grad_norm": 1.739460998611984,
      "language_loss": 0.9111495,
      "learning_rate": 3.974800639071641e-06,
      "loss": 0.93065745,
      "num_input_tokens_seen": 13989935,
      "step": 658,
      "time_per_iteration": 2.652122974395752
    },
    {
      "auxiliary_loss_clip": 0.01206668,
      "auxiliary_loss_mlp": 0.01041156,
      "balance_loss_clip": 1.07138193,
      "balance_loss_mlp": 1.02622473,
      "epoch": 0.07924006493116094,
      "flos": 23111389664640.0,
      "grad_norm": 3.6061210997633846,
      "language_loss": 1.00692689,
      "learning_rate": 3.974677223062492e-06,
      "loss": 1.02940512,
      "num_input_tokens_seen": 14007150,
      "step": 659,
      "time_per_iteration": 2.5890872478485107
    },
    {
      "auxiliary_loss_clip": 0.01218717,
      "auxiliary_loss_mlp": 0.0104705,
      "balance_loss_clip": 1.07366228,
      "balance_loss_mlp": 1.03155863,
      "epoch": 0.07936030782180004,
      "flos": 16472153450880.0,
      "grad_norm": 2.144650015755838,
      "language_loss": 0.74459255,
      "learning_rate": 3.974553507496378e-06,
      "loss": 0.76725018,
      "num_input_tokens_seen": 14025725,
      "step": 660,
      "time_per_iteration": 2.573432683944702
    },
    {
      "auxiliary_loss_clip": 0.01234107,
      "auxiliary_loss_mlp": 0.01051554,
      "balance_loss_clip": 1.08084738,
      "balance_loss_mlp": 1.03494203,
      "epoch": 0.07948055071243913,
      "flos": 23733290764800.0,
      "grad_norm": 2.0845066468914255,
      "language_loss": 0.89022744,
      "learning_rate": 3.974429492392068e-06,
      "loss": 0.91308403,
      "num_input_tokens_seen": 14045750,
      "step": 661,
      "time_per_iteration": 2.650777578353882
    },
    {
      "auxiliary_loss_clip": 0.01200084,
      "auxiliary_loss_mlp": 0.01046748,
      "balance_loss_clip": 1.07334614,
      "balance_loss_mlp": 1.03170383,
      "epoch": 0.07960079360307822,
      "flos": 19573326996480.0,
      "grad_norm": 2.0683426237374856,
      "language_loss": 0.91035306,
      "learning_rate": 3.974305177768373e-06,
      "loss": 0.93282139,
      "num_input_tokens_seen": 14063960,
      "step": 662,
      "time_per_iteration": 2.5832550525665283
    },
    {
      "auxiliary_loss_clip": 0.01194332,
      "auxiliary_loss_mlp": 0.01046025,
      "balance_loss_clip": 1.07856369,
      "balance_loss_mlp": 1.03121924,
      "epoch": 0.07972103649371731,
      "flos": 23513409659520.0,
      "grad_norm": 2.223492827701987,
      "language_loss": 0.86764997,
      "learning_rate": 3.974180563644152e-06,
      "loss": 0.89005351,
      "num_input_tokens_seen": 14082525,
      "step": 663,
      "time_per_iteration": 2.6662862300872803
    },
    {
      "auxiliary_loss_clip": 0.01239232,
      "auxiliary_loss_mlp": 0.0104583,
      "balance_loss_clip": 1.07632709,
      "balance_loss_mlp": 1.03167963,
      "epoch": 0.0798412793843564,
      "flos": 16726867770240.0,
      "grad_norm": 7.953891721936555,
      "language_loss": 0.89391327,
      "learning_rate": 3.97405565003831e-06,
      "loss": 0.9167639,
      "num_input_tokens_seen": 14098610,
      "step": 664,
      "time_per_iteration": 2.5527279376983643
    },
    {
      "auxiliary_loss_clip": 0.01241285,
      "auxiliary_loss_mlp": 0.0076673,
      "balance_loss_clip": 1.07825255,
      "balance_loss_mlp": 1.00111842,
      "epoch": 0.07996152227499549,
      "flos": 18223337214720.0,
      "grad_norm": 2.0940268156389057,
      "language_loss": 0.7814858,
      "learning_rate": 3.973930436969794e-06,
      "loss": 0.801566,
      "num_input_tokens_seen": 14117065,
      "step": 665,
      "time_per_iteration": 2.5579237937927246
    },
    {
      "auxiliary_loss_clip": 0.01239652,
      "auxiliary_loss_mlp": 0.01046207,
      "balance_loss_clip": 1.08128047,
      "balance_loss_mlp": 1.03052533,
      "epoch": 0.08008176516563459,
      "flos": 20594877793920.0,
      "grad_norm": 1.9868430607440686,
      "language_loss": 0.85691506,
      "learning_rate": 3.973804924457602e-06,
      "loss": 0.87977368,
      "num_input_tokens_seen": 14135145,
      "step": 666,
      "time_per_iteration": 3.2670774459838867
    },
    {
      "auxiliary_loss_clip": 0.01243565,
      "auxiliary_loss_mlp": 0.01054271,
      "balance_loss_clip": 1.07820547,
      "balance_loss_mlp": 1.03895307,
      "epoch": 0.08020200805627367,
      "flos": 31834306863360.0,
      "grad_norm": 1.8325869684422673,
      "language_loss": 0.85781634,
      "learning_rate": 3.973679112520771e-06,
      "loss": 0.8807947,
      "num_input_tokens_seen": 14156860,
      "step": 667,
      "time_per_iteration": 3.4012863636016846
    },
    {
      "auxiliary_loss_clip": 0.01251622,
      "auxiliary_loss_mlp": 0.01045388,
      "balance_loss_clip": 1.08033061,
      "balance_loss_mlp": 1.03212035,
      "epoch": 0.08032225094691277,
      "flos": 17783503176960.0,
      "grad_norm": 2.2030684009514934,
      "language_loss": 0.99296087,
      "learning_rate": 3.973553001178389e-06,
      "loss": 1.01593101,
      "num_input_tokens_seen": 14174365,
      "step": 668,
      "time_per_iteration": 2.50698184967041
    },
    {
      "auxiliary_loss_clip": 0.01223822,
      "auxiliary_loss_mlp": 0.0103656,
      "balance_loss_clip": 1.07672215,
      "balance_loss_mlp": 1.02246404,
      "epoch": 0.08044249383755185,
      "flos": 24061693835520.0,
      "grad_norm": 1.934502052816849,
      "language_loss": 0.7556572,
      "learning_rate": 3.973426590449585e-06,
      "loss": 0.77826107,
      "num_input_tokens_seen": 14192320,
      "step": 669,
      "time_per_iteration": 2.589428424835205
    },
    {
      "auxiliary_loss_clip": 0.01230119,
      "auxiliary_loss_mlp": 0.01045429,
      "balance_loss_clip": 1.08195257,
      "balance_loss_mlp": 1.03044462,
      "epoch": 0.08056273672819095,
      "flos": 18223624523520.0,
      "grad_norm": 2.4111894416678266,
      "language_loss": 0.75416905,
      "learning_rate": 3.9732998803535364e-06,
      "loss": 0.77692455,
      "num_input_tokens_seen": 14210380,
      "step": 670,
      "time_per_iteration": 3.3234329223632812
    },
    {
      "auxiliary_loss_clip": 0.01185255,
      "auxiliary_loss_mlp": 0.01050435,
      "balance_loss_clip": 1.07154059,
      "balance_loss_mlp": 1.03515267,
      "epoch": 0.08068297961883003,
      "flos": 19676856971520.0,
      "grad_norm": 3.473201888237786,
      "language_loss": 0.85511529,
      "learning_rate": 3.973172870909465e-06,
      "loss": 0.87747216,
      "num_input_tokens_seen": 14225145,
      "step": 671,
      "time_per_iteration": 2.589503049850464
    },
    {
      "auxiliary_loss_clip": 0.01197769,
      "auxiliary_loss_mlp": 0.0104432,
      "balance_loss_clip": 1.06903648,
      "balance_loss_mlp": 1.02710617,
      "epoch": 0.08080322250946913,
      "flos": 23148736830720.0,
      "grad_norm": 2.6898586669656104,
      "language_loss": 0.80986875,
      "learning_rate": 3.973045562136638e-06,
      "loss": 0.8322897,
      "num_input_tokens_seen": 14241960,
      "step": 672,
      "time_per_iteration": 3.4119491577148438
    },
    {
      "auxiliary_loss_clip": 0.01194618,
      "auxiliary_loss_mlp": 0.01043454,
      "balance_loss_clip": 1.07553148,
      "balance_loss_mlp": 1.03020346,
      "epoch": 0.08092346540010822,
      "flos": 21763626526080.0,
      "grad_norm": 2.2834853763583003,
      "language_loss": 0.91786838,
      "learning_rate": 3.972917954054368e-06,
      "loss": 0.94024909,
      "num_input_tokens_seen": 14260515,
      "step": 673,
      "time_per_iteration": 2.644604444503784
    },
    {
      "auxiliary_loss_clip": 0.01242847,
      "auxiliary_loss_mlp": 0.01054989,
      "balance_loss_clip": 1.08462906,
      "balance_loss_mlp": 1.037781,
      "epoch": 0.08104370829074731,
      "flos": 21032485188480.0,
      "grad_norm": 2.9348035135582706,
      "language_loss": 0.82085836,
      "learning_rate": 3.972790046682013e-06,
      "loss": 0.84383667,
      "num_input_tokens_seen": 14279190,
      "step": 674,
      "time_per_iteration": 2.5637195110321045
    },
    {
      "auxiliary_loss_clip": 0.01252189,
      "auxiliary_loss_mlp": 0.01042487,
      "balance_loss_clip": 1.07832634,
      "balance_loss_mlp": 1.0272398,
      "epoch": 0.0811639511813864,
      "flos": 20083186598400.0,
      "grad_norm": 1.9259232046300385,
      "language_loss": 0.79088616,
      "learning_rate": 3.972661840038977e-06,
      "loss": 0.81383294,
      "num_input_tokens_seen": 14299480,
      "step": 675,
      "time_per_iteration": 2.541774034500122
    },
    {
      "auxiliary_loss_clip": 0.01212159,
      "auxiliary_loss_mlp": 0.01048587,
      "balance_loss_clip": 1.07049024,
      "balance_loss_mlp": 1.03406715,
      "epoch": 0.08128419407202549,
      "flos": 16836718538880.0,
      "grad_norm": 2.2811293640471866,
      "language_loss": 0.83817053,
      "learning_rate": 3.972533334144707e-06,
      "loss": 0.86077797,
      "num_input_tokens_seen": 14316405,
      "step": 676,
      "time_per_iteration": 2.574030637741089
    },
    {
      "auxiliary_loss_clip": 0.01228272,
      "auxiliary_loss_mlp": 0.01046658,
      "balance_loss_clip": 1.0662241,
      "balance_loss_mlp": 1.03047502,
      "epoch": 0.08140443696266458,
      "flos": 23769273214080.0,
      "grad_norm": 2.0458081884228574,
      "language_loss": 0.78517991,
      "learning_rate": 3.972404529018699e-06,
      "loss": 0.80792928,
      "num_input_tokens_seen": 14336265,
      "step": 677,
      "time_per_iteration": 2.599534273147583
    },
    {
      "auxiliary_loss_clip": 0.01205538,
      "auxiliary_loss_mlp": 0.01047149,
      "balance_loss_clip": 1.06309295,
      "balance_loss_mlp": 1.03129387,
      "epoch": 0.08152467985330367,
      "flos": 24390132819840.0,
      "grad_norm": 2.1022058442155105,
      "language_loss": 0.85373461,
      "learning_rate": 3.972275424680493e-06,
      "loss": 0.87626147,
      "num_input_tokens_seen": 14356375,
      "step": 678,
      "time_per_iteration": 2.598515748977661
    },
    {
      "auxiliary_loss_clip": 0.01193104,
      "auxiliary_loss_mlp": 0.00767155,
      "balance_loss_clip": 1.07225966,
      "balance_loss_mlp": 1.00139666,
      "epoch": 0.08164492274394276,
      "flos": 19317750750720.0,
      "grad_norm": 2.2028335488459376,
      "language_loss": 0.92054021,
      "learning_rate": 3.972146021149673e-06,
      "loss": 0.94014281,
      "num_input_tokens_seen": 14374650,
      "step": 679,
      "time_per_iteration": 2.6196205615997314
    },
    {
      "auxiliary_loss_clip": 0.01220555,
      "auxiliary_loss_mlp": 0.01043763,
      "balance_loss_clip": 1.06800354,
      "balance_loss_mlp": 1.02964306,
      "epoch": 0.08176516563458186,
      "flos": 14830461319680.0,
      "grad_norm": 2.389038849384458,
      "language_loss": 0.78443956,
      "learning_rate": 3.972016318445868e-06,
      "loss": 0.80708277,
      "num_input_tokens_seen": 14392650,
      "step": 680,
      "time_per_iteration": 2.541459083557129
    },
    {
      "auxiliary_loss_clip": 0.01230557,
      "auxiliary_loss_mlp": 0.01042868,
      "balance_loss_clip": 1.07580996,
      "balance_loss_mlp": 1.02672124,
      "epoch": 0.08188540852522094,
      "flos": 22602320161920.0,
      "grad_norm": 2.6055189903163285,
      "language_loss": 0.92351383,
      "learning_rate": 3.971886316588757e-06,
      "loss": 0.94624805,
      "num_input_tokens_seen": 14413155,
      "step": 681,
      "time_per_iteration": 2.59188175201416
    },
    {
      "auxiliary_loss_clip": 0.01237769,
      "auxiliary_loss_mlp": 0.01047987,
      "balance_loss_clip": 1.07688951,
      "balance_loss_mlp": 1.03158355,
      "epoch": 0.08200565141586004,
      "flos": 19463727623040.0,
      "grad_norm": 3.7449731751964794,
      "language_loss": 0.73465747,
      "learning_rate": 3.9717560155980595e-06,
      "loss": 0.75751501,
      "num_input_tokens_seen": 14428805,
      "step": 682,
      "time_per_iteration": 2.523861885070801
    },
    {
      "auxiliary_loss_clip": 0.01255346,
      "auxiliary_loss_mlp": 0.01047476,
      "balance_loss_clip": 1.08033836,
      "balance_loss_mlp": 1.03206825,
      "epoch": 0.08212589430649912,
      "flos": 20594662312320.0,
      "grad_norm": 3.6366075726139666,
      "language_loss": 0.91898233,
      "learning_rate": 3.971625415493542e-06,
      "loss": 0.94201052,
      "num_input_tokens_seen": 14447125,
      "step": 683,
      "time_per_iteration": 2.5447025299072266
    },
    {
      "auxiliary_loss_clip": 0.01191933,
      "auxiliary_loss_mlp": 0.01045766,
      "balance_loss_clip": 1.07112908,
      "balance_loss_mlp": 1.03049493,
      "epoch": 0.08224613719713822,
      "flos": 25953611086080.0,
      "grad_norm": 2.073697656325714,
      "language_loss": 0.87755108,
      "learning_rate": 3.971494516295017e-06,
      "loss": 0.89992809,
      "num_input_tokens_seen": 14466575,
      "step": 684,
      "time_per_iteration": 2.626819133758545
    },
    {
      "auxiliary_loss_clip": 0.01225943,
      "auxiliary_loss_mlp": 0.00767185,
      "balance_loss_clip": 1.07856536,
      "balance_loss_mlp": 1.00162649,
      "epoch": 0.08236638008777732,
      "flos": 23768734510080.0,
      "grad_norm": 1.9866273237227268,
      "language_loss": 0.85371292,
      "learning_rate": 3.971363318022341e-06,
      "loss": 0.87364411,
      "num_input_tokens_seen": 14487915,
      "step": 685,
      "time_per_iteration": 2.5899293422698975
    },
    {
      "auxiliary_loss_clip": 0.01165015,
      "auxiliary_loss_mlp": 0.01054909,
      "balance_loss_clip": 1.0627135,
      "balance_loss_mlp": 1.03863716,
      "epoch": 0.0824866229784164,
      "flos": 38799144887040.0,
      "grad_norm": 2.2631584912800755,
      "language_loss": 0.68467104,
      "learning_rate": 3.971231820695417e-06,
      "loss": 0.70687032,
      "num_input_tokens_seen": 14511530,
      "step": 686,
      "time_per_iteration": 2.7963240146636963
    },
    {
      "auxiliary_loss_clip": 0.01227272,
      "auxiliary_loss_mlp": 0.01052801,
      "balance_loss_clip": 1.0766716,
      "balance_loss_mlp": 1.03744698,
      "epoch": 0.0826068658690555,
      "flos": 23107762391040.0,
      "grad_norm": 3.3029878549421574,
      "language_loss": 0.81275541,
      "learning_rate": 3.971100024334193e-06,
      "loss": 0.83555615,
      "num_input_tokens_seen": 14529050,
      "step": 687,
      "time_per_iteration": 2.5794918537139893
    },
    {
      "auxiliary_loss_clip": 0.01222221,
      "auxiliary_loss_mlp": 0.01045337,
      "balance_loss_clip": 1.07394135,
      "balance_loss_mlp": 1.03116298,
      "epoch": 0.08272710875969458,
      "flos": 21136374299520.0,
      "grad_norm": 2.3931138005481603,
      "language_loss": 0.85989535,
      "learning_rate": 3.970967928958663e-06,
      "loss": 0.88257098,
      "num_input_tokens_seen": 14546165,
      "step": 688,
      "time_per_iteration": 2.545147657394409
    },
    {
      "auxiliary_loss_clip": 0.01149435,
      "auxiliary_loss_mlp": 0.01044665,
      "balance_loss_clip": 1.0684557,
      "balance_loss_mlp": 1.03145099,
      "epoch": 0.08284735165033368,
      "flos": 19063000517760.0,
      "grad_norm": 2.6004630653289307,
      "language_loss": 0.8382858,
      "learning_rate": 3.970835534588865e-06,
      "loss": 0.86022681,
      "num_input_tokens_seen": 14563660,
      "step": 689,
      "time_per_iteration": 2.779228448867798
    },
    {
      "auxiliary_loss_clip": 0.01276759,
      "auxiliary_loss_mlp": 0.00767314,
      "balance_loss_clip": 1.08954775,
      "balance_loss_mlp": 1.0015533,
      "epoch": 0.08296759454097276,
      "flos": 16727442387840.0,
      "grad_norm": 2.4850601515859587,
      "language_loss": 0.86021924,
      "learning_rate": 3.970702841244883e-06,
      "loss": 0.88066,
      "num_input_tokens_seen": 14581980,
      "step": 690,
      "time_per_iteration": 3.038844108581543
    },
    {
      "auxiliary_loss_clip": 0.01274521,
      "auxiliary_loss_mlp": 0.00767039,
      "balance_loss_clip": 1.08365357,
      "balance_loss_mlp": 1.00146961,
      "epoch": 0.08308783743161186,
      "flos": 18004928567040.0,
      "grad_norm": 3.029865712896052,
      "language_loss": 0.8304944,
      "learning_rate": 3.970569848946847e-06,
      "loss": 0.85090995,
      "num_input_tokens_seen": 14601795,
      "step": 691,
      "time_per_iteration": 2.495129346847534
    },
    {
      "auxiliary_loss_clip": 0.01270795,
      "auxiliary_loss_mlp": 0.01043745,
      "balance_loss_clip": 1.08260441,
      "balance_loss_mlp": 1.02794337,
      "epoch": 0.08320808032225095,
      "flos": 15079788599040.0,
      "grad_norm": 2.720718934133009,
      "language_loss": 0.83219421,
      "learning_rate": 3.970436557714932e-06,
      "loss": 0.85533959,
      "num_input_tokens_seen": 14618315,
      "step": 692,
      "time_per_iteration": 2.4555258750915527
    },
    {
      "auxiliary_loss_clip": 0.01257444,
      "auxiliary_loss_mlp": 0.01056057,
      "balance_loss_clip": 1.07711017,
      "balance_loss_mlp": 1.04001713,
      "epoch": 0.08332832321289003,
      "flos": 22383085501440.0,
      "grad_norm": 2.447114955183074,
      "language_loss": 0.86434436,
      "learning_rate": 3.970302967569358e-06,
      "loss": 0.88747942,
      "num_input_tokens_seen": 14636905,
      "step": 693,
      "time_per_iteration": 3.271531581878662
    },
    {
      "auxiliary_loss_clip": 0.01223164,
      "auxiliary_loss_mlp": 0.0105364,
      "balance_loss_clip": 1.07604265,
      "balance_loss_mlp": 1.03749895,
      "epoch": 0.08344856610352913,
      "flos": 24717386655360.0,
      "grad_norm": 2.1852781614116865,
      "language_loss": 0.68405747,
      "learning_rate": 3.9701690785303896e-06,
      "loss": 0.70682549,
      "num_input_tokens_seen": 14656100,
      "step": 694,
      "time_per_iteration": 3.344320774078369
    },
    {
      "auxiliary_loss_clip": 0.01227668,
      "auxiliary_loss_mlp": 0.01047635,
      "balance_loss_clip": 1.07275534,
      "balance_loss_mlp": 1.0330255,
      "epoch": 0.08356880899416821,
      "flos": 25370206387200.0,
      "grad_norm": 2.1871867798642453,
      "language_loss": 0.88191873,
      "learning_rate": 3.970034890618339e-06,
      "loss": 0.90467179,
      "num_input_tokens_seen": 14675790,
      "step": 695,
      "time_per_iteration": 2.562988042831421
    },
    {
      "auxiliary_loss_clip": 0.0117614,
      "auxiliary_loss_mlp": 0.01048251,
      "balance_loss_clip": 1.06556368,
      "balance_loss_mlp": 1.03426182,
      "epoch": 0.08368905188480731,
      "flos": 24353072962560.0,
      "grad_norm": 2.068109411421455,
      "language_loss": 0.88097453,
      "learning_rate": 3.969900403853562e-06,
      "loss": 0.90321845,
      "num_input_tokens_seen": 14694830,
      "step": 696,
      "time_per_iteration": 2.7289562225341797
    },
    {
      "auxiliary_loss_clip": 0.0116695,
      "auxiliary_loss_mlp": 0.01047332,
      "balance_loss_clip": 1.0698061,
      "balance_loss_mlp": 1.03232932,
      "epoch": 0.08380929477544641,
      "flos": 18037319656320.0,
      "grad_norm": 1.7565254641934598,
      "language_loss": 0.78135312,
      "learning_rate": 3.96976561825646e-06,
      "loss": 0.80349594,
      "num_input_tokens_seen": 14711920,
      "step": 697,
      "time_per_iteration": 3.360534906387329
    },
    {
      "auxiliary_loss_clip": 0.01175383,
      "auxiliary_loss_mlp": 0.01042987,
      "balance_loss_clip": 1.06855869,
      "balance_loss_mlp": 1.02828264,
      "epoch": 0.08392953766608549,
      "flos": 26286287875200.0,
      "grad_norm": 1.9604082056317704,
      "language_loss": 0.87168002,
      "learning_rate": 3.969630533847479e-06,
      "loss": 0.89386368,
      "num_input_tokens_seen": 14730880,
      "step": 698,
      "time_per_iteration": 3.4612324237823486
    },
    {
      "auxiliary_loss_clip": 0.01181354,
      "auxiliary_loss_mlp": 0.0104187,
      "balance_loss_clip": 1.06823778,
      "balance_loss_mlp": 1.02773786,
      "epoch": 0.08404978055672459,
      "flos": 22492146170880.0,
      "grad_norm": 2.2477995368729253,
      "language_loss": 0.84391701,
      "learning_rate": 3.969495150647113e-06,
      "loss": 0.86614931,
      "num_input_tokens_seen": 14749050,
      "step": 699,
      "time_per_iteration": 2.647876501083374
    },
    {
      "auxiliary_loss_clip": 0.01179875,
      "auxiliary_loss_mlp": 0.01044799,
      "balance_loss_clip": 1.06990266,
      "balance_loss_mlp": 1.02956462,
      "epoch": 0.08417002344736367,
      "flos": 24826878288000.0,
      "grad_norm": 1.94542137671069,
      "language_loss": 0.76872075,
      "learning_rate": 3.969359468675899e-06,
      "loss": 0.79096746,
      "num_input_tokens_seen": 14769180,
      "step": 700,
      "time_per_iteration": 2.666329860687256
    },
    {
      "auxiliary_loss_clip": 0.01122193,
      "auxiliary_loss_mlp": 0.01044911,
      "balance_loss_clip": 1.04754472,
      "balance_loss_mlp": 1.03091609,
      "epoch": 0.08429026633800277,
      "flos": 16945922862720.0,
      "grad_norm": 2.0533545829586743,
      "language_loss": 0.89401311,
      "learning_rate": 3.969223487954418e-06,
      "loss": 0.91568416,
      "num_input_tokens_seen": 14786640,
      "step": 701,
      "time_per_iteration": 2.650880813598633
    },
    {
      "auxiliary_loss_clip": 0.01236265,
      "auxiliary_loss_mlp": 0.01044843,
      "balance_loss_clip": 1.08102012,
      "balance_loss_mlp": 1.03122306,
      "epoch": 0.08441050922864185,
      "flos": 23841920471040.0,
      "grad_norm": 2.4347167211674137,
      "language_loss": 0.82704133,
      "learning_rate": 3.969087208503301e-06,
      "loss": 0.84985244,
      "num_input_tokens_seen": 14806720,
      "step": 702,
      "time_per_iteration": 2.5636966228485107
    },
    {
      "auxiliary_loss_clip": 0.01229564,
      "auxiliary_loss_mlp": 0.01045972,
      "balance_loss_clip": 1.07486939,
      "balance_loss_mlp": 1.02954471,
      "epoch": 0.08453075211928095,
      "flos": 25520205582720.0,
      "grad_norm": 2.235625872215779,
      "language_loss": 0.84768283,
      "learning_rate": 3.968950630343219e-06,
      "loss": 0.8704381,
      "num_input_tokens_seen": 14823705,
      "step": 703,
      "time_per_iteration": 2.6197447776794434
    },
    {
      "auxiliary_loss_clip": 0.01145709,
      "auxiliary_loss_mlp": 0.0104251,
      "balance_loss_clip": 1.06249774,
      "balance_loss_mlp": 1.02812791,
      "epoch": 0.08465099500992004,
      "flos": 19532496211200.0,
      "grad_norm": 2.026489904998371,
      "language_loss": 0.93534517,
      "learning_rate": 3.968813753494892e-06,
      "loss": 0.95722735,
      "num_input_tokens_seen": 14841865,
      "step": 704,
      "time_per_iteration": 2.7536427974700928
    },
    {
      "auxiliary_loss_clip": 0.0126704,
      "auxiliary_loss_mlp": 0.01041142,
      "balance_loss_clip": 1.0801059,
      "balance_loss_mlp": 1.0274148,
      "epoch": 0.08477123790055913,
      "flos": 29351299403520.0,
      "grad_norm": 2.5151690303328897,
      "language_loss": 0.75329542,
      "learning_rate": 3.968676577979084e-06,
      "loss": 0.7763772,
      "num_input_tokens_seen": 14861415,
      "step": 705,
      "time_per_iteration": 3.0466361045837402
    },
    {
      "auxiliary_loss_clip": 0.01191218,
      "auxiliary_loss_mlp": 0.01047274,
      "balance_loss_clip": 1.07646,
      "balance_loss_mlp": 1.0322299,
      "epoch": 0.08489148079119822,
      "flos": 18624495283200.0,
      "grad_norm": 2.0738396601390274,
      "language_loss": 0.78325796,
      "learning_rate": 3.968539103816605e-06,
      "loss": 0.80564296,
      "num_input_tokens_seen": 14879215,
      "step": 706,
      "time_per_iteration": 2.6455962657928467
    },
    {
      "auxiliary_loss_clip": 0.01168272,
      "auxiliary_loss_mlp": 0.01049296,
      "balance_loss_clip": 1.06703246,
      "balance_loss_mlp": 1.03279757,
      "epoch": 0.0850117236818373,
      "flos": 23471393725440.0,
      "grad_norm": 2.5196357178803988,
      "language_loss": 0.89031583,
      "learning_rate": 3.9684013310283085e-06,
      "loss": 0.9124915,
      "num_input_tokens_seen": 14897900,
      "step": 707,
      "time_per_iteration": 2.698777914047241
    },
    {
      "auxiliary_loss_clip": 0.01218293,
      "auxiliary_loss_mlp": 0.01040371,
      "balance_loss_clip": 1.07963693,
      "balance_loss_mlp": 1.02724028,
      "epoch": 0.0851319665724764,
      "flos": 40625058896640.0,
      "grad_norm": 2.645564497964321,
      "language_loss": 0.63914335,
      "learning_rate": 3.9682632596350956e-06,
      "loss": 0.66172993,
      "num_input_tokens_seen": 14919065,
      "step": 708,
      "time_per_iteration": 2.7866086959838867
    },
    {
      "auxiliary_loss_clip": 0.01131109,
      "auxiliary_loss_mlp": 0.01044056,
      "balance_loss_clip": 1.05563939,
      "balance_loss_mlp": 1.02984679,
      "epoch": 0.0852522094631155,
      "flos": 15879554870400.0,
      "grad_norm": 2.864424449080321,
      "language_loss": 0.78690016,
      "learning_rate": 3.968124889657911e-06,
      "loss": 0.8086518,
      "num_input_tokens_seen": 14934165,
      "step": 709,
      "time_per_iteration": 2.6647024154663086
    },
    {
      "auxiliary_loss_clip": 0.0118981,
      "auxiliary_loss_mlp": 0.0104612,
      "balance_loss_clip": 1.07335103,
      "balance_loss_mlp": 1.03019977,
      "epoch": 0.08537245235375458,
      "flos": 14567091822720.0,
      "grad_norm": 2.464604972866228,
      "language_loss": 0.90538895,
      "learning_rate": 3.967986221117746e-06,
      "loss": 0.92774826,
      "num_input_tokens_seen": 14950105,
      "step": 710,
      "time_per_iteration": 2.638901710510254
    },
    {
      "auxiliary_loss_clip": 0.01253458,
      "auxiliary_loss_mlp": 0.01039413,
      "balance_loss_clip": 1.0826695,
      "balance_loss_mlp": 1.02554297,
      "epoch": 0.08549269524439368,
      "flos": 26468929555200.0,
      "grad_norm": 3.0869673731444744,
      "language_loss": 0.8668623,
      "learning_rate": 3.967847254035635e-06,
      "loss": 0.88979101,
      "num_input_tokens_seen": 14969490,
      "step": 711,
      "time_per_iteration": 2.574925422668457
    },
    {
      "auxiliary_loss_clip": 0.01236527,
      "auxiliary_loss_mlp": 0.01052913,
      "balance_loss_clip": 1.07913756,
      "balance_loss_mlp": 1.03727317,
      "epoch": 0.08561293813503276,
      "flos": 13590214565760.0,
      "grad_norm": 2.6471889221857197,
      "language_loss": 0.8584044,
      "learning_rate": 3.967707988432661e-06,
      "loss": 0.88129884,
      "num_input_tokens_seen": 14987195,
      "step": 712,
      "time_per_iteration": 2.5512101650238037
    },
    {
      "auxiliary_loss_clip": 0.01192259,
      "auxiliary_loss_mlp": 0.01043574,
      "balance_loss_clip": 1.06649971,
      "balance_loss_mlp": 1.02938831,
      "epoch": 0.08573318102567186,
      "flos": 26943524979840.0,
      "grad_norm": 2.373743542433335,
      "language_loss": 0.87839121,
      "learning_rate": 3.967568424329949e-06,
      "loss": 0.90074956,
      "num_input_tokens_seen": 15007620,
      "step": 713,
      "time_per_iteration": 2.653925895690918
    },
    {
      "auxiliary_loss_clip": 0.01129228,
      "auxiliary_loss_mlp": 0.01006301,
      "balance_loss_clip": 1.05518663,
      "balance_loss_mlp": 1.00301063,
      "epoch": 0.08585342391631094,
      "flos": 67302739319040.0,
      "grad_norm": 0.8441373430069209,
      "language_loss": 0.55493963,
      "learning_rate": 3.967428561748671e-06,
      "loss": 0.5762949,
      "num_input_tokens_seen": 15075590,
      "step": 714,
      "time_per_iteration": 3.325444459915161
    },
    {
      "auxiliary_loss_clip": 0.01255358,
      "auxiliary_loss_mlp": 0.01046014,
      "balance_loss_clip": 1.08063459,
      "balance_loss_mlp": 1.02971864,
      "epoch": 0.08597366680695004,
      "flos": 22456594684800.0,
      "grad_norm": 2.3610817387380743,
      "language_loss": 0.87531114,
      "learning_rate": 3.967288400710045e-06,
      "loss": 0.89832485,
      "num_input_tokens_seen": 15095055,
      "step": 715,
      "time_per_iteration": 2.588609218597412
    },
    {
      "auxiliary_loss_clip": 0.0125231,
      "auxiliary_loss_mlp": 0.01045269,
      "balance_loss_clip": 1.0832721,
      "balance_loss_mlp": 1.03129172,
      "epoch": 0.08609390969758914,
      "flos": 23550505430400.0,
      "grad_norm": 1.9524902754840854,
      "language_loss": 0.88727474,
      "learning_rate": 3.9671479412353335e-06,
      "loss": 0.91025054,
      "num_input_tokens_seen": 15113520,
      "step": 716,
      "time_per_iteration": 2.5760631561279297
    },
    {
      "auxiliary_loss_clip": 0.01208942,
      "auxiliary_loss_mlp": 0.00767141,
      "balance_loss_clip": 1.07256091,
      "balance_loss_mlp": 1.00116587,
      "epoch": 0.08621415258822822,
      "flos": 25885848078720.0,
      "grad_norm": 2.549273886470715,
      "language_loss": 0.73993552,
      "learning_rate": 3.967007183345843e-06,
      "loss": 0.75969625,
      "num_input_tokens_seen": 15133375,
      "step": 717,
      "time_per_iteration": 2.6701900959014893
    },
    {
      "auxiliary_loss_clip": 0.01128195,
      "auxiliary_loss_mlp": 0.01044067,
      "balance_loss_clip": 1.06279278,
      "balance_loss_mlp": 1.03063238,
      "epoch": 0.08633439547886732,
      "flos": 13589568120960.0,
      "grad_norm": 2.5718521163876553,
      "language_loss": 0.89773321,
      "learning_rate": 3.966866127062927e-06,
      "loss": 0.91945577,
      "num_input_tokens_seen": 15150500,
      "step": 718,
      "time_per_iteration": 2.685286045074463
    },
    {
      "auxiliary_loss_clip": 0.0116117,
      "auxiliary_loss_mlp": 0.01004047,
      "balance_loss_clip": 1.05756569,
      "balance_loss_mlp": 1.00073338,
      "epoch": 0.0864546383695064,
      "flos": 57767342434560.0,
      "grad_norm": 0.864040322660292,
      "language_loss": 0.62718141,
      "learning_rate": 3.966724772407982e-06,
      "loss": 0.64883351,
      "num_input_tokens_seen": 15208015,
      "step": 719,
      "time_per_iteration": 4.64807391166687
    },
    {
      "auxiliary_loss_clip": 0.01241119,
      "auxiliary_loss_mlp": 0.0104507,
      "balance_loss_clip": 1.07879364,
      "balance_loss_mlp": 1.03133702,
      "epoch": 0.0865748812601455,
      "flos": 20046952753920.0,
      "grad_norm": 2.077746333449827,
      "language_loss": 0.89024746,
      "learning_rate": 3.966583119402454e-06,
      "loss": 0.9131093,
      "num_input_tokens_seen": 15224780,
      "step": 720,
      "time_per_iteration": 2.526698112487793
    },
    {
      "auxiliary_loss_clip": 0.01266378,
      "auxiliary_loss_mlp": 0.0076709,
      "balance_loss_clip": 1.08100724,
      "balance_loss_mlp": 1.00151038,
      "epoch": 0.08669512415078459,
      "flos": 35262446935680.0,
      "grad_norm": 1.764562586009185,
      "language_loss": 0.81892633,
      "learning_rate": 3.9664411680678305e-06,
      "loss": 0.83926105,
      "num_input_tokens_seen": 15246535,
      "step": 721,
      "time_per_iteration": 2.6572265625
    },
    {
      "auxiliary_loss_clip": 0.01140375,
      "auxiliary_loss_mlp": 0.01003818,
      "balance_loss_clip": 1.05272794,
      "balance_loss_mlp": 1.00038457,
      "epoch": 0.08681536704142367,
      "flos": 65654870048640.0,
      "grad_norm": 0.8475234413304435,
      "language_loss": 0.61411476,
      "learning_rate": 3.966298918425644e-06,
      "loss": 0.6355567,
      "num_input_tokens_seen": 15304025,
      "step": 722,
      "time_per_iteration": 3.759632110595703
    },
    {
      "auxiliary_loss_clip": 0.0120536,
      "auxiliary_loss_mlp": 0.01045982,
      "balance_loss_clip": 1.07191384,
      "balance_loss_mlp": 1.0313549,
      "epoch": 0.08693560993206277,
      "flos": 34529940881280.0,
      "grad_norm": 7.0171952663069295,
      "language_loss": 0.82821423,
      "learning_rate": 3.966156370497476e-06,
      "loss": 0.85072762,
      "num_input_tokens_seen": 15327635,
      "step": 723,
      "time_per_iteration": 3.5169429779052734
    },
    {
      "auxiliary_loss_clip": 0.01108008,
      "auxiliary_loss_mlp": 0.01042055,
      "balance_loss_clip": 1.04780424,
      "balance_loss_mlp": 1.02705812,
      "epoch": 0.08705585282270185,
      "flos": 23149419189120.0,
      "grad_norm": 2.2375645942105913,
      "language_loss": 0.89012945,
      "learning_rate": 3.96601352430495e-06,
      "loss": 0.91163003,
      "num_input_tokens_seen": 15347405,
      "step": 724,
      "time_per_iteration": 2.7331695556640625
    },
    {
      "auxiliary_loss_clip": 0.01200128,
      "auxiliary_loss_mlp": 0.01041385,
      "balance_loss_clip": 1.06737065,
      "balance_loss_mlp": 1.02800941,
      "epoch": 0.08717609571334095,
      "flos": 29497599498240.0,
      "grad_norm": 1.6549441860799188,
      "language_loss": 0.83244896,
      "learning_rate": 3.965870379869735e-06,
      "loss": 0.854864,
      "num_input_tokens_seen": 15369450,
      "step": 725,
      "time_per_iteration": 2.6662535667419434
    },
    {
      "auxiliary_loss_clip": 0.01236348,
      "auxiliary_loss_mlp": 0.007664,
      "balance_loss_clip": 1.07607019,
      "balance_loss_mlp": 1.00115991,
      "epoch": 0.08729633860398003,
      "flos": 20667489137280.0,
      "grad_norm": 3.2733566536715144,
      "language_loss": 0.86996913,
      "learning_rate": 3.965726937213547e-06,
      "loss": 0.88999665,
      "num_input_tokens_seen": 15388085,
      "step": 726,
      "time_per_iteration": 2.592111587524414
    },
    {
      "auxiliary_loss_clip": 0.01166173,
      "auxiliary_loss_mlp": 0.01046527,
      "balance_loss_clip": 1.06768823,
      "balance_loss_mlp": 1.02990913,
      "epoch": 0.08741658149461913,
      "flos": 18369493655040.0,
      "grad_norm": 3.2573500073715476,
      "language_loss": 0.81227118,
      "learning_rate": 3.965583196358144e-06,
      "loss": 0.83439821,
      "num_input_tokens_seen": 15407120,
      "step": 727,
      "time_per_iteration": 2.7909481525421143
    },
    {
      "auxiliary_loss_clip": 0.01239047,
      "auxiliary_loss_mlp": 0.01038973,
      "balance_loss_clip": 1.07776093,
      "balance_loss_mlp": 1.02416134,
      "epoch": 0.08753682438525823,
      "flos": 18729677283840.0,
      "grad_norm": 2.57498713665269,
      "language_loss": 0.73709893,
      "learning_rate": 3.965439157325335e-06,
      "loss": 0.75987905,
      "num_input_tokens_seen": 15424485,
      "step": 728,
      "time_per_iteration": 2.5364513397216797
    },
    {
      "auxiliary_loss_clip": 0.01182435,
      "auxiliary_loss_mlp": 0.01042164,
      "balance_loss_clip": 1.066208,
      "balance_loss_mlp": 1.02701211,
      "epoch": 0.08765706727589731,
      "flos": 27776113303680.0,
      "grad_norm": 2.600660067102919,
      "language_loss": 0.76129299,
      "learning_rate": 3.965294820136968e-06,
      "loss": 0.783539,
      "num_input_tokens_seen": 15446285,
      "step": 729,
      "time_per_iteration": 2.766899585723877
    },
    {
      "auxiliary_loss_clip": 0.01268801,
      "auxiliary_loss_mlp": 0.00767072,
      "balance_loss_clip": 1.08160448,
      "balance_loss_mlp": 1.00148785,
      "epoch": 0.08777731016653641,
      "flos": 24389127239040.0,
      "grad_norm": 2.049706448374551,
      "language_loss": 0.8718307,
      "learning_rate": 3.965150184814938e-06,
      "loss": 0.89218944,
      "num_input_tokens_seen": 15465770,
      "step": 730,
      "time_per_iteration": 2.5266761779785156
    },
    {
      "auxiliary_loss_clip": 0.01195745,
      "auxiliary_loss_mlp": 0.01044724,
      "balance_loss_clip": 1.07618034,
      "balance_loss_mlp": 1.03124714,
      "epoch": 0.08789755305717549,
      "flos": 21981855605760.0,
      "grad_norm": 2.3087884028675485,
      "language_loss": 0.76679701,
      "learning_rate": 3.965005251381189e-06,
      "loss": 0.78920168,
      "num_input_tokens_seen": 15483705,
      "step": 731,
      "time_per_iteration": 2.6932802200317383
    },
    {
      "auxiliary_loss_clip": 0.01155745,
      "auxiliary_loss_mlp": 0.00760738,
      "balance_loss_clip": 1.05383468,
      "balance_loss_mlp": 1.00165045,
      "epoch": 0.08801779594781459,
      "flos": 58360120583040.0,
      "grad_norm": 0.8847691361505425,
      "language_loss": 0.64553809,
      "learning_rate": 3.964860019857705e-06,
      "loss": 0.66470295,
      "num_input_tokens_seen": 15548620,
      "step": 732,
      "time_per_iteration": 3.142707347869873
    },
    {
      "auxiliary_loss_clip": 0.01215793,
      "auxiliary_loss_mlp": 0.01043868,
      "balance_loss_clip": 1.07526445,
      "balance_loss_mlp": 1.03084421,
      "epoch": 0.08813803883845367,
      "flos": 23294785530240.0,
      "grad_norm": 2.0675181323369958,
      "language_loss": 0.84339797,
      "learning_rate": 3.964714490266518e-06,
      "loss": 0.86599451,
      "num_input_tokens_seen": 15569265,
      "step": 733,
      "time_per_iteration": 2.5784547328948975
    },
    {
      "auxiliary_loss_clip": 0.01125321,
      "auxiliary_loss_mlp": 0.01003859,
      "balance_loss_clip": 1.04134154,
      "balance_loss_mlp": 1.00035417,
      "epoch": 0.08825828172909277,
      "flos": 63424924882560.0,
      "grad_norm": 0.8771371424439008,
      "language_loss": 0.64573491,
      "learning_rate": 3.964568662629706e-06,
      "loss": 0.66702676,
      "num_input_tokens_seen": 15630570,
      "step": 734,
      "time_per_iteration": 3.0632221698760986
    },
    {
      "auxiliary_loss_clip": 0.01151524,
      "auxiliary_loss_mlp": 0.01042071,
      "balance_loss_clip": 1.06313014,
      "balance_loss_mlp": 1.02734256,
      "epoch": 0.08837852461973186,
      "flos": 26720986268160.0,
      "grad_norm": 2.1481277286858615,
      "language_loss": 0.84080982,
      "learning_rate": 3.9644225369693895e-06,
      "loss": 0.86274576,
      "num_input_tokens_seen": 15650870,
      "step": 735,
      "time_per_iteration": 2.7498767375946045
    },
    {
      "auxiliary_loss_clip": 0.01180699,
      "auxiliary_loss_mlp": 0.01047785,
      "balance_loss_clip": 1.06290483,
      "balance_loss_mlp": 1.03412962,
      "epoch": 0.08849876751037095,
      "flos": 27265427688960.0,
      "grad_norm": 2.617618708497164,
      "language_loss": 0.8703962,
      "learning_rate": 3.964276113307735e-06,
      "loss": 0.892681,
      "num_input_tokens_seen": 15670835,
      "step": 736,
      "time_per_iteration": 2.6631391048431396
    },
    {
      "auxiliary_loss_clip": 0.01199251,
      "auxiliary_loss_mlp": 0.01044988,
      "balance_loss_clip": 1.06788087,
      "balance_loss_mlp": 1.0300746,
      "epoch": 0.08861901040101004,
      "flos": 19828759587840.0,
      "grad_norm": 1.8588889908256907,
      "language_loss": 0.8078872,
      "learning_rate": 3.9641293916669574e-06,
      "loss": 0.83032966,
      "num_input_tokens_seen": 15689795,
      "step": 737,
      "time_per_iteration": 2.58668851852417
    },
    {
      "auxiliary_loss_clip": 0.01188021,
      "auxiliary_loss_mlp": 0.01043988,
      "balance_loss_clip": 1.06686807,
      "balance_loss_mlp": 1.02863991,
      "epoch": 0.08873925329164913,
      "flos": 23658704173440.0,
      "grad_norm": 9.81952813942635,
      "language_loss": 0.82996601,
      "learning_rate": 3.9639823720693115e-06,
      "loss": 0.8522861,
      "num_input_tokens_seen": 15711650,
      "step": 738,
      "time_per_iteration": 2.6975646018981934
    },
    {
      "auxiliary_loss_clip": 0.01117316,
      "auxiliary_loss_mlp": 0.01012712,
      "balance_loss_clip": 1.04835534,
      "balance_loss_mlp": 1.00920677,
      "epoch": 0.08885949618228822,
      "flos": 71831541893760.0,
      "grad_norm": 0.8479021469843174,
      "language_loss": 0.60003775,
      "learning_rate": 3.963835054537102e-06,
      "loss": 0.62133807,
      "num_input_tokens_seen": 15780615,
      "step": 739,
      "time_per_iteration": 3.2563209533691406
    },
    {
      "auxiliary_loss_clip": 0.01244069,
      "auxiliary_loss_mlp": 0.00767002,
      "balance_loss_clip": 1.07676721,
      "balance_loss_mlp": 1.00129974,
      "epoch": 0.08897973907292732,
      "flos": 22346169298560.0,
      "grad_norm": 2.1413084940455334,
      "language_loss": 0.61024469,
      "learning_rate": 3.963687439092676e-06,
      "loss": 0.63035536,
      "num_input_tokens_seen": 15801300,
      "step": 740,
      "time_per_iteration": 2.541802406311035
    },
    {
      "auxiliary_loss_clip": 0.01223994,
      "auxiliary_loss_mlp": 0.01049079,
      "balance_loss_clip": 1.07482135,
      "balance_loss_mlp": 1.03460097,
      "epoch": 0.0890999819635664,
      "flos": 21251827589760.0,
      "grad_norm": 2.2709528467729787,
      "language_loss": 0.80766582,
      "learning_rate": 3.963539525758427e-06,
      "loss": 0.83039653,
      "num_input_tokens_seen": 15820860,
      "step": 741,
      "time_per_iteration": 2.5682733058929443
    },
    {
      "auxiliary_loss_clip": 0.0124628,
      "auxiliary_loss_mlp": 0.0104038,
      "balance_loss_clip": 1.07555175,
      "balance_loss_mlp": 1.02644479,
      "epoch": 0.0892202248542055,
      "flos": 25370888745600.0,
      "grad_norm": 2.3912804021992375,
      "language_loss": 0.68065679,
      "learning_rate": 3.9633913145567925e-06,
      "loss": 0.70352346,
      "num_input_tokens_seen": 15841350,
      "step": 742,
      "time_per_iteration": 2.556364059448242
    },
    {
      "auxiliary_loss_clip": 0.01222557,
      "auxiliary_loss_mlp": 0.01047016,
      "balance_loss_clip": 1.07235849,
      "balance_loss_mlp": 1.0338192,
      "epoch": 0.08934046774484458,
      "flos": 24457895827200.0,
      "grad_norm": 2.0346917454446265,
      "language_loss": 0.81347185,
      "learning_rate": 3.9632428055102575e-06,
      "loss": 0.83616757,
      "num_input_tokens_seen": 15861360,
      "step": 743,
      "time_per_iteration": 2.591404914855957
    },
    {
      "auxiliary_loss_clip": 0.01215165,
      "auxiliary_loss_mlp": 0.00767082,
      "balance_loss_clip": 1.07336116,
      "balance_loss_mlp": 1.0011065,
      "epoch": 0.08946071063548368,
      "flos": 35772773414400.0,
      "grad_norm": 2.6581705765878136,
      "language_loss": 0.66926658,
      "learning_rate": 3.9630939986413495e-06,
      "loss": 0.68908906,
      "num_input_tokens_seen": 15883160,
      "step": 744,
      "time_per_iteration": 2.701741933822632
    },
    {
      "auxiliary_loss_clip": 0.0126285,
      "auxiliary_loss_mlp": 0.01047401,
      "balance_loss_clip": 1.07936943,
      "balance_loss_mlp": 1.03386509,
      "epoch": 0.08958095352612276,
      "flos": 14356584167040.0,
      "grad_norm": 2.30030672330948,
      "language_loss": 0.78174043,
      "learning_rate": 3.962944893972643e-06,
      "loss": 0.80484289,
      "num_input_tokens_seen": 15901610,
      "step": 745,
      "time_per_iteration": 3.4856297969818115
    },
    {
      "auxiliary_loss_clip": 0.0124022,
      "auxiliary_loss_mlp": 0.01048594,
      "balance_loss_clip": 1.07760775,
      "balance_loss_mlp": 1.03334665,
      "epoch": 0.08970119641676186,
      "flos": 17853277345920.0,
      "grad_norm": 2.6291849707244808,
      "language_loss": 0.90577567,
      "learning_rate": 3.962795491526756e-06,
      "loss": 0.92866385,
      "num_input_tokens_seen": 15918770,
      "step": 746,
      "time_per_iteration": 2.5330982208251953
    },
    {
      "auxiliary_loss_clip": 0.01167707,
      "auxiliary_loss_mlp": 0.01057039,
      "balance_loss_clip": 1.06104994,
      "balance_loss_mlp": 1.04160118,
      "epoch": 0.08982143930740095,
      "flos": 20811670329600.0,
      "grad_norm": 3.2209867461860533,
      "language_loss": 0.89119196,
      "learning_rate": 3.962645791326354e-06,
      "loss": 0.91343939,
      "num_input_tokens_seen": 15938025,
      "step": 747,
      "time_per_iteration": 2.6915342807769775
    },
    {
      "auxiliary_loss_clip": 0.01216388,
      "auxiliary_loss_mlp": 0.01042405,
      "balance_loss_clip": 1.07210398,
      "balance_loss_mlp": 1.02946472,
      "epoch": 0.08994168219804004,
      "flos": 24097712198400.0,
      "grad_norm": 2.199008997067869,
      "language_loss": 0.83334494,
      "learning_rate": 3.962495793394146e-06,
      "loss": 0.85593283,
      "num_input_tokens_seen": 15957215,
      "step": 748,
      "time_per_iteration": 3.6112358570098877
    },
    {
      "auxiliary_loss_clip": 0.01143081,
      "auxiliary_loss_mlp": 0.01012884,
      "balance_loss_clip": 1.05303764,
      "balance_loss_mlp": 1.00952244,
      "epoch": 0.09006192508867913,
      "flos": 57188893812480.0,
      "grad_norm": 0.743775651828821,
      "language_loss": 0.61205208,
      "learning_rate": 3.9623454977528864e-06,
      "loss": 0.6336118,
      "num_input_tokens_seen": 16015870,
      "step": 749,
      "time_per_iteration": 3.758213996887207
    },
    {
      "auxiliary_loss_clip": 0.01189721,
      "auxiliary_loss_mlp": 0.0105372,
      "balance_loss_clip": 1.07194805,
      "balance_loss_mlp": 1.03713202,
      "epoch": 0.09018216797931822,
      "flos": 20487505063680.0,
      "grad_norm": 1.7236816400719739,
      "language_loss": 0.84912258,
      "learning_rate": 3.962194904425375e-06,
      "loss": 0.871557,
      "num_input_tokens_seen": 16036500,
      "step": 750,
      "time_per_iteration": 2.6984381675720215
    },
    {
      "auxiliary_loss_clip": 0.01186991,
      "auxiliary_loss_mlp": 0.01048407,
      "balance_loss_clip": 1.06662202,
      "balance_loss_mlp": 1.03467417,
      "epoch": 0.09030241086995731,
      "flos": 22638123043200.0,
      "grad_norm": 2.8054574469343,
      "language_loss": 0.68185604,
      "learning_rate": 3.9620440134344566e-06,
      "loss": 0.70421004,
      "num_input_tokens_seen": 16054655,
      "step": 751,
      "time_per_iteration": 2.6511313915252686
    },
    {
      "auxiliary_loss_clip": 0.0119253,
      "auxiliary_loss_mlp": 0.01051957,
      "balance_loss_clip": 1.0702405,
      "balance_loss_mlp": 1.035923,
      "epoch": 0.09042265376059641,
      "flos": 21871502046720.0,
      "grad_norm": 2.2048213332982507,
      "language_loss": 0.82467532,
      "learning_rate": 3.9618928248030215e-06,
      "loss": 0.84712017,
      "num_input_tokens_seen": 16074165,
      "step": 752,
      "time_per_iteration": 2.6521387100219727
    },
    {
      "auxiliary_loss_clip": 0.01185901,
      "auxiliary_loss_mlp": 0.00766386,
      "balance_loss_clip": 1.06686378,
      "balance_loss_mlp": 1.00100803,
      "epoch": 0.0905428966512355,
      "flos": 24316192673280.0,
      "grad_norm": 2.3335550253989403,
      "language_loss": 0.82953143,
      "learning_rate": 3.961741338554005e-06,
      "loss": 0.84905428,
      "num_input_tokens_seen": 16092505,
      "step": 753,
      "time_per_iteration": 2.632967948913574
    },
    {
      "auxiliary_loss_clip": 0.01219762,
      "auxiliary_loss_mlp": 0.01047258,
      "balance_loss_clip": 1.06672692,
      "balance_loss_mlp": 1.03104591,
      "epoch": 0.09066313954187459,
      "flos": 35845061535360.0,
      "grad_norm": 2.2653300701095116,
      "language_loss": 0.75736266,
      "learning_rate": 3.9615895547103865e-06,
      "loss": 0.78003293,
      "num_input_tokens_seen": 16116150,
      "step": 754,
      "time_per_iteration": 2.6760807037353516
    },
    {
      "auxiliary_loss_clip": 0.0123829,
      "auxiliary_loss_mlp": 0.0076621,
      "balance_loss_clip": 1.07299578,
      "balance_loss_mlp": 1.00114274,
      "epoch": 0.09078338243251367,
      "flos": 29168729550720.0,
      "grad_norm": 3.2282881491951643,
      "language_loss": 0.78077644,
      "learning_rate": 3.961437473295193e-06,
      "loss": 0.80082142,
      "num_input_tokens_seen": 16136295,
      "step": 755,
      "time_per_iteration": 2.5685806274414062
    },
    {
      "auxiliary_loss_clip": 0.01243988,
      "auxiliary_loss_mlp": 0.01041116,
      "balance_loss_clip": 1.07485151,
      "balance_loss_mlp": 1.02930284,
      "epoch": 0.09090362532315277,
      "flos": 21907699977600.0,
      "grad_norm": 2.359387687557653,
      "language_loss": 0.72393084,
      "learning_rate": 3.961285094331495e-06,
      "loss": 0.74678195,
      "num_input_tokens_seen": 16154210,
      "step": 756,
      "time_per_iteration": 2.5175397396087646
    },
    {
      "auxiliary_loss_clip": 0.01208292,
      "auxiliary_loss_mlp": 0.01041418,
      "balance_loss_clip": 1.07060695,
      "balance_loss_mlp": 1.02872205,
      "epoch": 0.09102386821379185,
      "flos": 27344503480320.0,
      "grad_norm": 1.7455479398343465,
      "language_loss": 0.85913342,
      "learning_rate": 3.961132417842406e-06,
      "loss": 0.88163054,
      "num_input_tokens_seen": 16173995,
      "step": 757,
      "time_per_iteration": 2.616971254348755
    },
    {
      "auxiliary_loss_clip": 0.01186318,
      "auxiliary_loss_mlp": 0.01046742,
      "balance_loss_clip": 1.06551576,
      "balance_loss_mlp": 1.0325979,
      "epoch": 0.09114411110443095,
      "flos": 20813501923200.0,
      "grad_norm": 3.341271762344445,
      "language_loss": 0.75179011,
      "learning_rate": 3.960979443851089e-06,
      "loss": 0.77412069,
      "num_input_tokens_seen": 16191020,
      "step": 758,
      "time_per_iteration": 2.6407594680786133
    },
    {
      "auxiliary_loss_clip": 0.01236558,
      "auxiliary_loss_mlp": 0.01042804,
      "balance_loss_clip": 1.07737184,
      "balance_loss_mlp": 1.02811766,
      "epoch": 0.09126435399507005,
      "flos": 26145949438080.0,
      "grad_norm": 1.8936578057276716,
      "language_loss": 0.78769797,
      "learning_rate": 3.96082617238075e-06,
      "loss": 0.81049156,
      "num_input_tokens_seen": 16213645,
      "step": 759,
      "time_per_iteration": 2.6455516815185547
    },
    {
      "auxiliary_loss_clip": 0.01211643,
      "auxiliary_loss_mlp": 0.01040183,
      "balance_loss_clip": 1.06813049,
      "balance_loss_mlp": 1.02547228,
      "epoch": 0.09138459688570913,
      "flos": 24388911757440.0,
      "grad_norm": 2.4825044427215843,
      "language_loss": 0.79988486,
      "learning_rate": 3.960672603454639e-06,
      "loss": 0.82240307,
      "num_input_tokens_seen": 16233625,
      "step": 760,
      "time_per_iteration": 2.6197621822357178
    },
    {
      "auxiliary_loss_clip": 0.01226793,
      "auxiliary_loss_mlp": 0.01048104,
      "balance_loss_clip": 1.07334375,
      "balance_loss_mlp": 1.03330433,
      "epoch": 0.09150483977634823,
      "flos": 21032664756480.0,
      "grad_norm": 2.9135142805740832,
      "language_loss": 0.7654494,
      "learning_rate": 3.960518737096054e-06,
      "loss": 0.78819835,
      "num_input_tokens_seen": 16253255,
      "step": 761,
      "time_per_iteration": 2.5446224212646484
    },
    {
      "auxiliary_loss_clip": 0.01204962,
      "auxiliary_loss_mlp": 0.01049049,
      "balance_loss_clip": 1.0700475,
      "balance_loss_mlp": 1.03634691,
      "epoch": 0.09162508266698731,
      "flos": 22856998567680.0,
      "grad_norm": 2.3690840686053893,
      "language_loss": 0.73701286,
      "learning_rate": 3.960364573328334e-06,
      "loss": 0.75955302,
      "num_input_tokens_seen": 16272580,
      "step": 762,
      "time_per_iteration": 2.5874831676483154
    },
    {
      "auxiliary_loss_clip": 0.012437,
      "auxiliary_loss_mlp": 0.00766857,
      "balance_loss_clip": 1.07276773,
      "balance_loss_mlp": 1.00113761,
      "epoch": 0.0917453255576264,
      "flos": 21724411852800.0,
      "grad_norm": 4.981778113625512,
      "language_loss": 0.88729858,
      "learning_rate": 3.9602101121748675e-06,
      "loss": 0.90740418,
      "num_input_tokens_seen": 16293075,
      "step": 763,
      "time_per_iteration": 2.557673454284668
    },
    {
      "auxiliary_loss_clip": 0.0122974,
      "auxiliary_loss_mlp": 0.01037922,
      "balance_loss_clip": 1.07261777,
      "balance_loss_mlp": 1.02533388,
      "epoch": 0.0918655684482655,
      "flos": 14609215497600.0,
      "grad_norm": 1.9420615090813345,
      "language_loss": 0.7272374,
      "learning_rate": 3.960055353659085e-06,
      "loss": 0.74991399,
      "num_input_tokens_seen": 16310185,
      "step": 764,
      "time_per_iteration": 2.564389944076538
    },
    {
      "auxiliary_loss_clip": 0.01231921,
      "auxiliary_loss_mlp": 0.01038255,
      "balance_loss_clip": 1.07738495,
      "balance_loss_mlp": 1.02533877,
      "epoch": 0.09198581133890459,
      "flos": 23435016226560.0,
      "grad_norm": 2.7959111450544993,
      "language_loss": 0.83774817,
      "learning_rate": 3.959900297804465e-06,
      "loss": 0.86044991,
      "num_input_tokens_seen": 16330355,
      "step": 765,
      "time_per_iteration": 2.5811045169830322
    },
    {
      "auxiliary_loss_clip": 0.0121606,
      "auxiliary_loss_mlp": 0.01039277,
      "balance_loss_clip": 1.07051039,
      "balance_loss_mlp": 1.02730203,
      "epoch": 0.09210605422954368,
      "flos": 16795887753600.0,
      "grad_norm": 1.962167330473937,
      "language_loss": 0.77322936,
      "learning_rate": 3.9597449446345276e-06,
      "loss": 0.7957828,
      "num_input_tokens_seen": 16347600,
      "step": 766,
      "time_per_iteration": 2.5622661113739014
    },
    {
      "auxiliary_loss_clip": 0.0119242,
      "auxiliary_loss_mlp": 0.0105314,
      "balance_loss_clip": 1.06690705,
      "balance_loss_mlp": 1.03964531,
      "epoch": 0.09222629712018277,
      "flos": 22674249146880.0,
      "grad_norm": 2.6023096291100005,
      "language_loss": 0.83673471,
      "learning_rate": 3.95958929417284e-06,
      "loss": 0.85919034,
      "num_input_tokens_seen": 16365755,
      "step": 767,
      "time_per_iteration": 2.5809075832366943
    },
    {
      "auxiliary_loss_clip": 0.01113246,
      "auxiliary_loss_mlp": 0.01004135,
      "balance_loss_clip": 1.04652715,
      "balance_loss_mlp": 1.00101173,
      "epoch": 0.09234654001082186,
      "flos": 69976756327680.0,
      "grad_norm": 0.7308785602710113,
      "language_loss": 0.58741891,
      "learning_rate": 3.9594333464430145e-06,
      "loss": 0.60859275,
      "num_input_tokens_seen": 16435245,
      "step": 768,
      "time_per_iteration": 3.2980408668518066
    },
    {
      "auxiliary_loss_clip": 0.01248506,
      "auxiliary_loss_mlp": 0.0104267,
      "balance_loss_clip": 1.07748842,
      "balance_loss_mlp": 1.03064823,
      "epoch": 0.09246678290146094,
      "flos": 20011437181440.0,
      "grad_norm": 1.9756165256389286,
      "language_loss": 0.88660276,
      "learning_rate": 3.959277101468709e-06,
      "loss": 0.90951449,
      "num_input_tokens_seen": 16454795,
      "step": 769,
      "time_per_iteration": 2.5281784534454346
    },
    {
      "auxiliary_loss_clip": 0.01198514,
      "auxiliary_loss_mlp": 0.00765491,
      "balance_loss_clip": 1.07170105,
      "balance_loss_mlp": 1.00111222,
      "epoch": 0.09258702579210004,
      "flos": 17747448900480.0,
      "grad_norm": 3.26667901316601,
      "language_loss": 0.78805035,
      "learning_rate": 3.959120559273624e-06,
      "loss": 0.80769032,
      "num_input_tokens_seen": 16472580,
      "step": 770,
      "time_per_iteration": 2.596433401107788
    },
    {
      "auxiliary_loss_clip": 0.01222013,
      "auxiliary_loss_mlp": 0.01045717,
      "balance_loss_clip": 1.07704353,
      "balance_loss_mlp": 1.03221679,
      "epoch": 0.09270726868273914,
      "flos": 20886544229760.0,
      "grad_norm": 1.8861862648186174,
      "language_loss": 0.83483803,
      "learning_rate": 3.958963719881509e-06,
      "loss": 0.85751539,
      "num_input_tokens_seen": 16490670,
      "step": 771,
      "time_per_iteration": 4.19177770614624
    },
    {
      "auxiliary_loss_clip": 0.01188031,
      "auxiliary_loss_mlp": 0.01046073,
      "balance_loss_clip": 1.06754732,
      "balance_loss_mlp": 1.03253055,
      "epoch": 0.09282751157337822,
      "flos": 17015697031680.0,
      "grad_norm": 1.9135333226696736,
      "language_loss": 0.93973565,
      "learning_rate": 3.958806583316154e-06,
      "loss": 0.96207666,
      "num_input_tokens_seen": 16508640,
      "step": 772,
      "time_per_iteration": 2.5961780548095703
    },
    {
      "auxiliary_loss_clip": 0.0116582,
      "auxiliary_loss_mlp": 0.01047855,
      "balance_loss_clip": 1.06468916,
      "balance_loss_mlp": 1.03414571,
      "epoch": 0.09294775446401732,
      "flos": 32523647748480.0,
      "grad_norm": 1.8787170822934909,
      "language_loss": 0.78839511,
      "learning_rate": 3.9586491496013985e-06,
      "loss": 0.81053185,
      "num_input_tokens_seen": 16531035,
      "step": 773,
      "time_per_iteration": 2.874605178833008
    },
    {
      "auxiliary_loss_clip": 0.01216033,
      "auxiliary_loss_mlp": 0.01047692,
      "balance_loss_clip": 1.074072,
      "balance_loss_mlp": 1.03356004,
      "epoch": 0.0930679973546564,
      "flos": 18259750627200.0,
      "grad_norm": 1.9839670888424736,
      "language_loss": 0.83097124,
      "learning_rate": 3.958491418761124e-06,
      "loss": 0.85360855,
      "num_input_tokens_seen": 16548605,
      "step": 774,
      "time_per_iteration": 3.5654897689819336
    },
    {
      "auxiliary_loss_clip": 0.01265308,
      "auxiliary_loss_mlp": 0.01045654,
      "balance_loss_clip": 1.08003497,
      "balance_loss_mlp": 1.03134322,
      "epoch": 0.0931882402452955,
      "flos": 21099745405440.0,
      "grad_norm": 2.372864223362955,
      "language_loss": 0.72668278,
      "learning_rate": 3.958333390819258e-06,
      "loss": 0.74979246,
      "num_input_tokens_seen": 16565535,
      "step": 775,
      "time_per_iteration": 3.210710287094116
    },
    {
      "auxiliary_loss_clip": 0.01228138,
      "auxiliary_loss_mlp": 0.01043754,
      "balance_loss_clip": 1.07685566,
      "balance_loss_mlp": 1.03059363,
      "epoch": 0.0933084831359346,
      "flos": 24207275658240.0,
      "grad_norm": 2.058193389764022,
      "language_loss": 0.80374873,
      "learning_rate": 3.9581750657997754e-06,
      "loss": 0.82646763,
      "num_input_tokens_seen": 16584900,
      "step": 776,
      "time_per_iteration": 2.632889747619629
    },
    {
      "auxiliary_loss_clip": 0.01178782,
      "auxiliary_loss_mlp": 0.0103475,
      "balance_loss_clip": 1.05556488,
      "balance_loss_mlp": 1.02189898,
      "epoch": 0.09342872602657368,
      "flos": 25480272637440.0,
      "grad_norm": 1.783989709531195,
      "language_loss": 0.89502543,
      "learning_rate": 3.95801644372669e-06,
      "loss": 0.91716075,
      "num_input_tokens_seen": 16604805,
      "step": 777,
      "time_per_iteration": 2.6362831592559814
    },
    {
      "auxiliary_loss_clip": 0.01231538,
      "auxiliary_loss_mlp": 0.01035284,
      "balance_loss_clip": 1.07442069,
      "balance_loss_mlp": 1.02197456,
      "epoch": 0.09354896891721277,
      "flos": 23149060053120.0,
      "grad_norm": 2.4602624561305473,
      "language_loss": 0.84493101,
      "learning_rate": 3.957857524624068e-06,
      "loss": 0.86759925,
      "num_input_tokens_seen": 16623685,
      "step": 778,
      "time_per_iteration": 2.564415216445923
    },
    {
      "auxiliary_loss_clip": 0.01211148,
      "auxiliary_loss_mlp": 0.01043419,
      "balance_loss_clip": 1.07351625,
      "balance_loss_mlp": 1.02983546,
      "epoch": 0.09366921180785186,
      "flos": 24279563779200.0,
      "grad_norm": 1.651554443724088,
      "language_loss": 0.89796865,
      "learning_rate": 3.957698308516016e-06,
      "loss": 0.92051429,
      "num_input_tokens_seen": 16644985,
      "step": 779,
      "time_per_iteration": 2.5877115726470947
    },
    {
      "auxiliary_loss_clip": 0.01221197,
      "auxiliary_loss_mlp": 0.01044227,
      "balance_loss_clip": 1.0802443,
      "balance_loss_mlp": 1.03119159,
      "epoch": 0.09378945469849095,
      "flos": 18730036419840.0,
      "grad_norm": 2.058599426377717,
      "language_loss": 0.82236439,
      "learning_rate": 3.957538795426688e-06,
      "loss": 0.84501868,
      "num_input_tokens_seen": 16662410,
      "step": 780,
      "time_per_iteration": 2.5772976875305176
    },
    {
      "auxiliary_loss_clip": 0.01198027,
      "auxiliary_loss_mlp": 0.01047825,
      "balance_loss_clip": 1.07269514,
      "balance_loss_mlp": 1.03349042,
      "epoch": 0.09390969758913004,
      "flos": 23218834222080.0,
      "grad_norm": 2.4933187626932414,
      "language_loss": 0.77598286,
      "learning_rate": 3.9573789853802804e-06,
      "loss": 0.79844129,
      "num_input_tokens_seen": 16680885,
      "step": 781,
      "time_per_iteration": 2.6324877738952637
    },
    {
      "auxiliary_loss_clip": 0.01263675,
      "auxiliary_loss_mlp": 0.01041594,
      "balance_loss_clip": 1.08182943,
      "balance_loss_mlp": 1.02774179,
      "epoch": 0.09402994047976913,
      "flos": 19646728439040.0,
      "grad_norm": 1.8952456521409289,
      "language_loss": 0.74008119,
      "learning_rate": 3.957218878401037e-06,
      "loss": 0.76313388,
      "num_input_tokens_seen": 16699375,
      "step": 782,
      "time_per_iteration": 2.4792306423187256
    },
    {
      "auxiliary_loss_clip": 0.0120172,
      "auxiliary_loss_mlp": 0.01042989,
      "balance_loss_clip": 1.07462859,
      "balance_loss_mlp": 1.02835011,
      "epoch": 0.09415018337040823,
      "flos": 29420463041280.0,
      "grad_norm": 2.497192101623559,
      "language_loss": 0.89273876,
      "learning_rate": 3.957058474513246e-06,
      "loss": 0.91518581,
      "num_input_tokens_seen": 16719230,
      "step": 783,
      "time_per_iteration": 2.67545747756958
    },
    {
      "auxiliary_loss_clip": 0.01196361,
      "auxiliary_loss_mlp": 0.01043335,
      "balance_loss_clip": 1.06805229,
      "balance_loss_mlp": 1.02975106,
      "epoch": 0.09427042626104731,
      "flos": 24572092141440.0,
      "grad_norm": 1.921430406343138,
      "language_loss": 0.7840836,
      "learning_rate": 3.956897773741241e-06,
      "loss": 0.80648059,
      "num_input_tokens_seen": 16738220,
      "step": 784,
      "time_per_iteration": 2.640820026397705
    },
    {
      "auxiliary_loss_clip": 0.01190743,
      "auxiliary_loss_mlp": 0.01038733,
      "balance_loss_clip": 1.07362604,
      "balance_loss_mlp": 1.02504146,
      "epoch": 0.09439066915168641,
      "flos": 26359581576960.0,
      "grad_norm": 1.91413774091777,
      "language_loss": 0.7188561,
      "learning_rate": 3.956736776109398e-06,
      "loss": 0.74115086,
      "num_input_tokens_seen": 16759395,
      "step": 785,
      "time_per_iteration": 2.702176332473755
    },
    {
      "auxiliary_loss_clip": 0.0117085,
      "auxiliary_loss_mlp": 0.01039685,
      "balance_loss_clip": 1.06953275,
      "balance_loss_mlp": 1.02668512,
      "epoch": 0.09451091204232549,
      "flos": 19427278296960.0,
      "grad_norm": 1.9889835810550716,
      "language_loss": 0.8361026,
      "learning_rate": 3.956575481642143e-06,
      "loss": 0.85820794,
      "num_input_tokens_seen": 16778285,
      "step": 786,
      "time_per_iteration": 2.6887712478637695
    },
    {
      "auxiliary_loss_clip": 0.01244539,
      "auxiliary_loss_mlp": 0.01043849,
      "balance_loss_clip": 1.07517374,
      "balance_loss_mlp": 1.03058124,
      "epoch": 0.09463115493296459,
      "flos": 25368051571200.0,
      "grad_norm": 2.3260208946104193,
      "language_loss": 0.74681056,
      "learning_rate": 3.956413890363943e-06,
      "loss": 0.76969445,
      "num_input_tokens_seen": 16795265,
      "step": 787,
      "time_per_iteration": 2.608145236968994
    },
    {
      "auxiliary_loss_clip": 0.01235955,
      "auxiliary_loss_mlp": 0.01044714,
      "balance_loss_clip": 1.07573402,
      "balance_loss_mlp": 1.03008127,
      "epoch": 0.09475139782360369,
      "flos": 10123254869760.0,
      "grad_norm": 3.1084664358580802,
      "language_loss": 0.81690562,
      "learning_rate": 3.956252002299312e-06,
      "loss": 0.83971226,
      "num_input_tokens_seen": 16811165,
      "step": 788,
      "time_per_iteration": 2.5097529888153076
    },
    {
      "auxiliary_loss_clip": 0.01202565,
      "auxiliary_loss_mlp": 0.01039497,
      "balance_loss_clip": 1.0715971,
      "balance_loss_mlp": 1.02631271,
      "epoch": 0.09487164071424277,
      "flos": 17231088936960.0,
      "grad_norm": 1.7753958729127959,
      "language_loss": 0.90875709,
      "learning_rate": 3.956089817472807e-06,
      "loss": 0.93117774,
      "num_input_tokens_seen": 16828470,
      "step": 789,
      "time_per_iteration": 2.614766836166382
    },
    {
      "auxiliary_loss_clip": 0.01249564,
      "auxiliary_loss_mlp": 0.0105386,
      "balance_loss_clip": 1.07710171,
      "balance_loss_mlp": 1.03825545,
      "epoch": 0.09499188360488187,
      "flos": 30849564528000.0,
      "grad_norm": 1.993747375429304,
      "language_loss": 0.8548376,
      "learning_rate": 3.955927335909032e-06,
      "loss": 0.87787187,
      "num_input_tokens_seen": 16851680,
      "step": 790,
      "time_per_iteration": 2.615593194961548
    },
    {
      "auxiliary_loss_clip": 0.01150545,
      "auxiliary_loss_mlp": 0.01046799,
      "balance_loss_clip": 1.05918312,
      "balance_loss_mlp": 1.03373933,
      "epoch": 0.09511212649552095,
      "flos": 29351694453120.0,
      "grad_norm": 2.903765284257419,
      "language_loss": 0.7571162,
      "learning_rate": 3.955764557632634e-06,
      "loss": 0.77908963,
      "num_input_tokens_seen": 16871490,
      "step": 791,
      "time_per_iteration": 2.918743133544922
    },
    {
      "auxiliary_loss_clip": 0.01224313,
      "auxiliary_loss_mlp": 0.01037324,
      "balance_loss_clip": 1.07074904,
      "balance_loss_mlp": 1.02476525,
      "epoch": 0.09523236938616005,
      "flos": 10378687461120.0,
      "grad_norm": 2.605269934485072,
      "language_loss": 0.95022857,
      "learning_rate": 3.955601482668309e-06,
      "loss": 0.9728449,
      "num_input_tokens_seen": 16889350,
      "step": 792,
      "time_per_iteration": 2.7619688510894775
    },
    {
      "auxiliary_loss_clip": 0.01259611,
      "auxiliary_loss_mlp": 0.0103855,
      "balance_loss_clip": 1.07729888,
      "balance_loss_mlp": 1.02485871,
      "epoch": 0.09535261227679913,
      "flos": 19061815368960.0,
      "grad_norm": 1.9656632192877794,
      "language_loss": 0.89122081,
      "learning_rate": 3.955438111040794e-06,
      "loss": 0.91420239,
      "num_input_tokens_seen": 16907625,
      "step": 793,
      "time_per_iteration": 2.48758864402771
    },
    {
      "auxiliary_loss_clip": 0.01245604,
      "auxiliary_loss_mlp": 0.01036428,
      "balance_loss_clip": 1.07890391,
      "balance_loss_mlp": 1.02203321,
      "epoch": 0.09547285516743823,
      "flos": 20922993555840.0,
      "grad_norm": 2.178869288276746,
      "language_loss": 0.80176365,
      "learning_rate": 3.955274442774873e-06,
      "loss": 0.82458395,
      "num_input_tokens_seen": 16926205,
      "step": 794,
      "time_per_iteration": 2.514148473739624
    },
    {
      "auxiliary_loss_clip": 0.01181581,
      "auxiliary_loss_mlp": 0.0104018,
      "balance_loss_clip": 1.06509352,
      "balance_loss_mlp": 1.02531433,
      "epoch": 0.09559309805807732,
      "flos": 30154405639680.0,
      "grad_norm": 2.379188414494047,
      "language_loss": 0.70766586,
      "learning_rate": 3.9551104778953725e-06,
      "loss": 0.72988343,
      "num_input_tokens_seen": 16946500,
      "step": 795,
      "time_per_iteration": 2.699132204055786
    },
    {
      "auxiliary_loss_clip": 0.01154284,
      "auxiliary_loss_mlp": 0.01041832,
      "balance_loss_clip": 1.0631212,
      "balance_loss_mlp": 1.02784276,
      "epoch": 0.0957133409487164,
      "flos": 21066743784960.0,
      "grad_norm": 1.7535635046626077,
      "language_loss": 0.85352939,
      "learning_rate": 3.954946216427167e-06,
      "loss": 0.87549055,
      "num_input_tokens_seen": 16966960,
      "step": 796,
      "time_per_iteration": 4.609287977218628
    },
    {
      "auxiliary_loss_clip": 0.01046624,
      "auxiliary_loss_mlp": 0.0101155,
      "balance_loss_clip": 1.02013755,
      "balance_loss_mlp": 1.00856996,
      "epoch": 0.0958335838393555,
      "flos": 71297979315840.0,
      "grad_norm": 0.8827521008818159,
      "language_loss": 0.61586142,
      "learning_rate": 3.954781658395176e-06,
      "loss": 0.63644314,
      "num_input_tokens_seen": 17023215,
      "step": 797,
      "time_per_iteration": 3.414611577987671
    },
    {
      "auxiliary_loss_clip": 0.01215035,
      "auxiliary_loss_mlp": 0.01043902,
      "balance_loss_clip": 1.0751574,
      "balance_loss_mlp": 1.02967477,
      "epoch": 0.09595382672999458,
      "flos": 21872974504320.0,
      "grad_norm": 1.9221851020613556,
      "language_loss": 0.92346585,
      "learning_rate": 3.95461680382436e-06,
      "loss": 0.94605523,
      "num_input_tokens_seen": 17042140,
      "step": 798,
      "time_per_iteration": 2.974628448486328
    },
    {
      "auxiliary_loss_clip": 0.01227183,
      "auxiliary_loss_mlp": 0.00767191,
      "balance_loss_clip": 1.07372212,
      "balance_loss_mlp": 1.00069463,
      "epoch": 0.09607406962063368,
      "flos": 18695562341760.0,
      "grad_norm": 3.8558898268904427,
      "language_loss": 0.86101007,
      "learning_rate": 3.9544516527397295e-06,
      "loss": 0.88095379,
      "num_input_tokens_seen": 17058490,
      "step": 799,
      "time_per_iteration": 3.3352832794189453
    },
    {
      "auxiliary_loss_clip": 0.01219301,
      "auxiliary_loss_mlp": 0.01048436,
      "balance_loss_clip": 1.07521152,
      "balance_loss_mlp": 1.03409505,
      "epoch": 0.09619431251127276,
      "flos": 22568456615040.0,
      "grad_norm": 1.6770504987215533,
      "language_loss": 0.80753207,
      "learning_rate": 3.954286205166338e-06,
      "loss": 0.83020937,
      "num_input_tokens_seen": 17079655,
      "step": 800,
      "time_per_iteration": 2.622572660446167
    },
    {
      "auxiliary_loss_clip": 0.01199697,
      "auxiliary_loss_mlp": 0.01052491,
      "balance_loss_clip": 1.07259369,
      "balance_loss_mlp": 1.03666043,
      "epoch": 0.09631455540191186,
      "flos": 14246230608000.0,
      "grad_norm": 2.0413645357911774,
      "language_loss": 0.83958608,
      "learning_rate": 3.954120461129282e-06,
      "loss": 0.86210799,
      "num_input_tokens_seen": 17097065,
      "step": 801,
      "time_per_iteration": 3.384977340698242
    },
    {
      "auxiliary_loss_clip": 0.01250399,
      "auxiliary_loss_mlp": 0.010523,
      "balance_loss_clip": 1.08505166,
      "balance_loss_mlp": 1.03959274,
      "epoch": 0.09643479829255096,
      "flos": 20740387789440.0,
      "grad_norm": 1.8992135893507058,
      "language_loss": 0.83939946,
      "learning_rate": 3.953954420653706e-06,
      "loss": 0.86242646,
      "num_input_tokens_seen": 17114090,
      "step": 802,
      "time_per_iteration": 2.520634412765503
    },
    {
      "auxiliary_loss_clip": 0.01225564,
      "auxiliary_loss_mlp": 0.01056374,
      "balance_loss_clip": 1.07271147,
      "balance_loss_mlp": 1.04352295,
      "epoch": 0.09655504118319004,
      "flos": 24420476833920.0,
      "grad_norm": 2.1788410727468697,
      "language_loss": 0.8807323,
      "learning_rate": 3.953788083764798e-06,
      "loss": 0.9035517,
      "num_input_tokens_seen": 17133325,
      "step": 803,
      "time_per_iteration": 2.5921192169189453
    },
    {
      "auxiliary_loss_clip": 0.01210244,
      "auxiliary_loss_mlp": 0.01045317,
      "balance_loss_clip": 1.06996512,
      "balance_loss_mlp": 1.03189421,
      "epoch": 0.09667528407382914,
      "flos": 18441961344000.0,
      "grad_norm": 1.9040658943333064,
      "language_loss": 0.92417949,
      "learning_rate": 3.953621450487792e-06,
      "loss": 0.94673508,
      "num_input_tokens_seen": 17151945,
      "step": 804,
      "time_per_iteration": 2.5539534091949463
    },
    {
      "auxiliary_loss_clip": 0.01105762,
      "auxiliary_loss_mlp": 0.01009889,
      "balance_loss_clip": 1.03641057,
      "balance_loss_mlp": 1.00698018,
      "epoch": 0.09679552696446822,
      "flos": 70816455544320.0,
      "grad_norm": 0.8311436514418802,
      "language_loss": 0.61229169,
      "learning_rate": 3.953454520847964e-06,
      "loss": 0.63344824,
      "num_input_tokens_seen": 17216790,
      "step": 805,
      "time_per_iteration": 3.317516803741455
    },
    {
      "auxiliary_loss_clip": 0.01252302,
      "auxiliary_loss_mlp": 0.01051289,
      "balance_loss_clip": 1.0806452,
      "balance_loss_mlp": 1.03444457,
      "epoch": 0.09691576985510732,
      "flos": 21945514020480.0,
      "grad_norm": 2.115041816653555,
      "language_loss": 0.73167419,
      "learning_rate": 3.9532872948706395e-06,
      "loss": 0.75471008,
      "num_input_tokens_seen": 17236285,
      "step": 806,
      "time_per_iteration": 2.570624828338623
    },
    {
      "auxiliary_loss_clip": 0.01196353,
      "auxiliary_loss_mlp": 0.01050776,
      "balance_loss_clip": 1.06535554,
      "balance_loss_mlp": 1.03624439,
      "epoch": 0.09703601274574641,
      "flos": 17965211103360.0,
      "grad_norm": 2.2038163921646756,
      "language_loss": 0.82613003,
      "learning_rate": 3.9531197725811845e-06,
      "loss": 0.84860128,
      "num_input_tokens_seen": 17251670,
      "step": 807,
      "time_per_iteration": 2.549086093902588
    },
    {
      "auxiliary_loss_clip": 0.01110469,
      "auxiliary_loss_mlp": 0.01045463,
      "balance_loss_clip": 1.04841316,
      "balance_loss_mlp": 1.03227282,
      "epoch": 0.0971562556363855,
      "flos": 22162162901760.0,
      "grad_norm": 5.161026527414781,
      "language_loss": 0.88091117,
      "learning_rate": 3.952951954005013e-06,
      "loss": 0.90247047,
      "num_input_tokens_seen": 17271355,
      "step": 808,
      "time_per_iteration": 2.8812623023986816
    },
    {
      "auxiliary_loss_clip": 0.01228001,
      "auxiliary_loss_mlp": 0.01038303,
      "balance_loss_clip": 1.07141328,
      "balance_loss_mlp": 1.0257566,
      "epoch": 0.0972764985270246,
      "flos": 25848716394240.0,
      "grad_norm": 2.28431736870881,
      "language_loss": 0.8508088,
      "learning_rate": 3.952783839167584e-06,
      "loss": 0.87347186,
      "num_input_tokens_seen": 17291400,
      "step": 809,
      "time_per_iteration": 3.1631360054016113
    },
    {
      "auxiliary_loss_clip": 0.01165052,
      "auxiliary_loss_mlp": 0.01046,
      "balance_loss_clip": 1.06289876,
      "balance_loss_mlp": 1.03150451,
      "epoch": 0.09739674141766368,
      "flos": 20339373375360.0,
      "grad_norm": 3.6823842754283382,
      "language_loss": 0.74518037,
      "learning_rate": 3.952615428094398e-06,
      "loss": 0.76729089,
      "num_input_tokens_seen": 17310920,
      "step": 810,
      "time_per_iteration": 2.6887495517730713
    },
    {
      "auxiliary_loss_clip": 0.01198019,
      "auxiliary_loss_mlp": 0.01050899,
      "balance_loss_clip": 1.06814718,
      "balance_loss_mlp": 1.03832221,
      "epoch": 0.09751698430830277,
      "flos": 15743059188480.0,
      "grad_norm": 1.7654549590731348,
      "language_loss": 0.73305619,
      "learning_rate": 3.952446720811004e-06,
      "loss": 0.75554532,
      "num_input_tokens_seen": 17329245,
      "step": 811,
      "time_per_iteration": 2.573704957962036
    },
    {
      "auxiliary_loss_clip": 0.01112723,
      "auxiliary_loss_mlp": 0.01003458,
      "balance_loss_clip": 1.0413245,
      "balance_loss_mlp": 1.00050187,
      "epoch": 0.09763722719894186,
      "flos": 63716806800000.0,
      "grad_norm": 1.011191199414165,
      "language_loss": 0.63561058,
      "learning_rate": 3.952277717342995e-06,
      "loss": 0.65677238,
      "num_input_tokens_seen": 17395680,
      "step": 812,
      "time_per_iteration": 3.3107895851135254
    },
    {
      "auxiliary_loss_clip": 0.01197432,
      "auxiliary_loss_mlp": 0.01048573,
      "balance_loss_clip": 1.07182693,
      "balance_loss_mlp": 1.03293252,
      "epoch": 0.09775747008958095,
      "flos": 22090916275200.0,
      "grad_norm": 2.398846296610079,
      "language_loss": 0.85933089,
      "learning_rate": 3.952108417716009e-06,
      "loss": 0.88179094,
      "num_input_tokens_seen": 17415135,
      "step": 813,
      "time_per_iteration": 2.662886381149292
    },
    {
      "auxiliary_loss_clip": 0.01196728,
      "auxiliary_loss_mlp": 0.01054989,
      "balance_loss_clip": 1.06529868,
      "balance_loss_mlp": 1.03886604,
      "epoch": 0.09787771298022005,
      "flos": 21286050272640.0,
      "grad_norm": 1.8618504947881223,
      "language_loss": 0.85375905,
      "learning_rate": 3.951938821955727e-06,
      "loss": 0.87627625,
      "num_input_tokens_seen": 17434535,
      "step": 814,
      "time_per_iteration": 2.6127610206604004
    },
    {
      "auxiliary_loss_clip": 0.0123211,
      "auxiliary_loss_mlp": 0.01048544,
      "balance_loss_clip": 1.07883418,
      "balance_loss_mlp": 1.03358305,
      "epoch": 0.09799795587085913,
      "flos": 22054574689920.0,
      "grad_norm": 1.6420346020652892,
      "language_loss": 0.76368666,
      "learning_rate": 3.9517689300878786e-06,
      "loss": 0.78649318,
      "num_input_tokens_seen": 17454270,
      "step": 815,
      "time_per_iteration": 2.612335681915283
    },
    {
      "auxiliary_loss_clip": 0.01220212,
      "auxiliary_loss_mlp": 0.01043562,
      "balance_loss_clip": 1.07074046,
      "balance_loss_mlp": 1.02938771,
      "epoch": 0.09811819876149823,
      "flos": 22163743100160.0,
      "grad_norm": 1.910110151959232,
      "language_loss": 0.78669798,
      "learning_rate": 3.951598742138236e-06,
      "loss": 0.80933571,
      "num_input_tokens_seen": 17472995,
      "step": 816,
      "time_per_iteration": 2.573904275894165
    },
    {
      "auxiliary_loss_clip": 0.01172202,
      "auxiliary_loss_mlp": 0.01052825,
      "balance_loss_clip": 1.06724548,
      "balance_loss_mlp": 1.03821003,
      "epoch": 0.09823844165213731,
      "flos": 22231111057920.0,
      "grad_norm": 3.3801622376646665,
      "language_loss": 0.79841149,
      "learning_rate": 3.951428258132615e-06,
      "loss": 0.82066166,
      "num_input_tokens_seen": 17491115,
      "step": 817,
      "time_per_iteration": 2.676811456680298
    },
    {
      "auxiliary_loss_clip": 0.01263849,
      "auxiliary_loss_mlp": 0.0076623,
      "balance_loss_clip": 1.08086085,
      "balance_loss_mlp": 1.00058818,
      "epoch": 0.09835868454277641,
      "flos": 22487728798080.0,
      "grad_norm": 2.277215126869,
      "language_loss": 0.84478199,
      "learning_rate": 3.951257478096879e-06,
      "loss": 0.86508274,
      "num_input_tokens_seen": 17509480,
      "step": 818,
      "time_per_iteration": 2.5241403579711914
    },
    {
      "auxiliary_loss_clip": 0.01195855,
      "auxiliary_loss_mlp": 0.01042567,
      "balance_loss_clip": 1.06441736,
      "balance_loss_mlp": 1.02810681,
      "epoch": 0.0984789274334155,
      "flos": 16362554077440.0,
      "grad_norm": 4.24229775723784,
      "language_loss": 0.68511677,
      "learning_rate": 3.951086402056936e-06,
      "loss": 0.70750099,
      "num_input_tokens_seen": 17524080,
      "step": 819,
      "time_per_iteration": 2.5918140411376953
    },
    {
      "auxiliary_loss_clip": 0.01212015,
      "auxiliary_loss_mlp": 0.01049884,
      "balance_loss_clip": 1.07494068,
      "balance_loss_mlp": 1.03644323,
      "epoch": 0.09859917032405459,
      "flos": 24243545416320.0,
      "grad_norm": 1.646730482772727,
      "language_loss": 0.83875191,
      "learning_rate": 3.950915030038735e-06,
      "loss": 0.86137092,
      "num_input_tokens_seen": 17543875,
      "step": 820,
      "time_per_iteration": 2.618865966796875
    },
    {
      "auxiliary_loss_clip": 0.01203932,
      "auxiliary_loss_mlp": 0.01040654,
      "balance_loss_clip": 1.07139015,
      "balance_loss_mlp": 1.02773762,
      "epoch": 0.09871941321469369,
      "flos": 17420195064960.0,
      "grad_norm": 2.206224917392231,
      "language_loss": 0.83808041,
      "learning_rate": 3.9507433620682765e-06,
      "loss": 0.86052626,
      "num_input_tokens_seen": 17560810,
      "step": 821,
      "time_per_iteration": 2.6492209434509277
    },
    {
      "auxiliary_loss_clip": 0.01230923,
      "auxiliary_loss_mlp": 0.00766137,
      "balance_loss_clip": 1.071383,
      "balance_loss_mlp": 1.0007478,
      "epoch": 0.09883965610533277,
      "flos": 28477341590400.0,
      "grad_norm": 2.081814893924499,
      "language_loss": 0.88000453,
      "learning_rate": 3.9505713981716e-06,
      "loss": 0.89997518,
      "num_input_tokens_seen": 17583640,
      "step": 822,
      "time_per_iteration": 3.3198883533477783
    },
    {
      "auxiliary_loss_clip": 0.01238154,
      "auxiliary_loss_mlp": 0.01043915,
      "balance_loss_clip": 1.07631683,
      "balance_loss_mlp": 1.03171372,
      "epoch": 0.09895989899597187,
      "flos": 23693932437120.0,
      "grad_norm": 2.087062340823377,
      "language_loss": 0.8135699,
      "learning_rate": 3.950399138374795e-06,
      "loss": 0.83639055,
      "num_input_tokens_seen": 17602720,
      "step": 823,
      "time_per_iteration": 3.3089611530303955
    },
    {
      "auxiliary_loss_clip": 0.01221212,
      "auxiliary_loss_mlp": 0.01044296,
      "balance_loss_clip": 1.06691933,
      "balance_loss_mlp": 1.03058696,
      "epoch": 0.09908014188661095,
      "flos": 24679608526080.0,
      "grad_norm": 1.7480145181429554,
      "language_loss": 0.74187076,
      "learning_rate": 3.95022658270399e-06,
      "loss": 0.76452589,
      "num_input_tokens_seen": 17623085,
      "step": 824,
      "time_per_iteration": 2.60074782371521
    },
    {
      "auxiliary_loss_clip": 0.01174084,
      "auxiliary_loss_mlp": 0.01055034,
      "balance_loss_clip": 1.06320333,
      "balance_loss_mlp": 1.04058564,
      "epoch": 0.09920038477725004,
      "flos": 14064307200000.0,
      "grad_norm": 1.774988419626696,
      "language_loss": 0.78091991,
      "learning_rate": 3.9500537311853635e-06,
      "loss": 0.80321115,
      "num_input_tokens_seen": 17641040,
      "step": 825,
      "time_per_iteration": 3.3376951217651367
    },
    {
      "auxiliary_loss_clip": 0.01177709,
      "auxiliary_loss_mlp": 0.01048393,
      "balance_loss_clip": 1.06013966,
      "balance_loss_mlp": 1.03343225,
      "epoch": 0.09932062766788914,
      "flos": 13407070095360.0,
      "grad_norm": 2.7720586033944183,
      "language_loss": 0.8392601,
      "learning_rate": 3.949880583845136e-06,
      "loss": 0.86152112,
      "num_input_tokens_seen": 17659115,
      "step": 826,
      "time_per_iteration": 2.640408992767334
    },
    {
      "auxiliary_loss_clip": 0.01211489,
      "auxiliary_loss_mlp": 0.00766618,
      "balance_loss_clip": 1.07628453,
      "balance_loss_mlp": 1.00073171,
      "epoch": 0.09944087055852822,
      "flos": 19500751566720.0,
      "grad_norm": 1.9878340262123462,
      "language_loss": 0.81235236,
      "learning_rate": 3.949707140709575e-06,
      "loss": 0.83213341,
      "num_input_tokens_seen": 17678845,
      "step": 827,
      "time_per_iteration": 3.453151226043701
    },
    {
      "auxiliary_loss_clip": 0.01202787,
      "auxiliary_loss_mlp": 0.00766537,
      "balance_loss_clip": 1.07082367,
      "balance_loss_mlp": 1.0006175,
      "epoch": 0.09956111344916732,
      "flos": 17749100926080.0,
      "grad_norm": 2.2910292808460384,
      "language_loss": 0.83520222,
      "learning_rate": 3.949533401804991e-06,
      "loss": 0.85489547,
      "num_input_tokens_seen": 17695750,
      "step": 828,
      "time_per_iteration": 2.6348588466644287
    },
    {
      "auxiliary_loss_clip": 0.01224926,
      "auxiliary_loss_mlp": 0.01048968,
      "balance_loss_clip": 1.07377803,
      "balance_loss_mlp": 1.03549731,
      "epoch": 0.0996813563398064,
      "flos": 17967581400960.0,
      "grad_norm": 1.952328578362024,
      "language_loss": 0.9079442,
      "learning_rate": 3.949359367157739e-06,
      "loss": 0.9306832,
      "num_input_tokens_seen": 17714445,
      "step": 829,
      "time_per_iteration": 2.55387544631958
    },
    {
      "auxiliary_loss_clip": 0.0125156,
      "auxiliary_loss_mlp": 0.01041612,
      "balance_loss_clip": 1.07801867,
      "balance_loss_mlp": 1.02749789,
      "epoch": 0.0998015992304455,
      "flos": 17457039440640.0,
      "grad_norm": 2.0762809389388566,
      "language_loss": 0.75863278,
      "learning_rate": 3.949185036794222e-06,
      "loss": 0.78156447,
      "num_input_tokens_seen": 17732455,
      "step": 830,
      "time_per_iteration": 2.4995908737182617
    },
    {
      "auxiliary_loss_clip": 0.01144284,
      "auxiliary_loss_mlp": 0.0104038,
      "balance_loss_clip": 1.05751705,
      "balance_loss_mlp": 1.02620625,
      "epoch": 0.0999218421210846,
      "flos": 25888757080320.0,
      "grad_norm": 1.6201265152400437,
      "language_loss": 0.7883023,
      "learning_rate": 3.949010410740884e-06,
      "loss": 0.81014895,
      "num_input_tokens_seen": 17755280,
      "step": 831,
      "time_per_iteration": 2.8712410926818848
    },
    {
      "auxiliary_loss_clip": 0.01215818,
      "auxiliary_loss_mlp": 0.01046081,
      "balance_loss_clip": 1.06283665,
      "balance_loss_mlp": 1.03126335,
      "epoch": 0.10004208501172368,
      "flos": 21215916967680.0,
      "grad_norm": 1.9940510506555011,
      "language_loss": 0.86244321,
      "learning_rate": 3.948835489024216e-06,
      "loss": 0.88506222,
      "num_input_tokens_seen": 17775015,
      "step": 832,
      "time_per_iteration": 2.992931604385376
    },
    {
      "auxiliary_loss_clip": 0.01126114,
      "auxiliary_loss_mlp": 0.01046147,
      "balance_loss_clip": 1.0482688,
      "balance_loss_mlp": 1.03197289,
      "epoch": 0.10016232790236278,
      "flos": 17348409734400.0,
      "grad_norm": 1.9871879166289972,
      "language_loss": 0.90229934,
      "learning_rate": 3.948660271670755e-06,
      "loss": 0.92402196,
      "num_input_tokens_seen": 17792165,
      "step": 833,
      "time_per_iteration": 2.767047882080078
    },
    {
      "auxiliary_loss_clip": 0.01241393,
      "auxiliary_loss_mlp": 0.01041437,
      "balance_loss_clip": 1.07574987,
      "balance_loss_mlp": 1.02789521,
      "epoch": 0.10028257079300186,
      "flos": 25666541591040.0,
      "grad_norm": 2.0608690138138583,
      "language_loss": 0.84008908,
      "learning_rate": 3.948484758707079e-06,
      "loss": 0.86291742,
      "num_input_tokens_seen": 17811765,
      "step": 834,
      "time_per_iteration": 2.7839086055755615
    },
    {
      "auxiliary_loss_clip": 0.01211208,
      "auxiliary_loss_mlp": 0.01047678,
      "balance_loss_clip": 1.07058406,
      "balance_loss_mlp": 1.03356349,
      "epoch": 0.10040281368364096,
      "flos": 25156035544320.0,
      "grad_norm": 2.065266934868026,
      "language_loss": 0.8350004,
      "learning_rate": 3.948308950159815e-06,
      "loss": 0.85758924,
      "num_input_tokens_seen": 17830445,
      "step": 835,
      "time_per_iteration": 2.577228546142578
    },
    {
      "auxiliary_loss_clip": 0.01161309,
      "auxiliary_loss_mlp": 0.01042457,
      "balance_loss_clip": 1.06260228,
      "balance_loss_mlp": 1.02750838,
      "epoch": 0.10052305657428004,
      "flos": 17603303621760.0,
      "grad_norm": 2.5560651475708633,
      "language_loss": 0.7589972,
      "learning_rate": 3.9481328460556326e-06,
      "loss": 0.78103483,
      "num_input_tokens_seen": 17847665,
      "step": 836,
      "time_per_iteration": 2.6527047157287598
    },
    {
      "auxiliary_loss_clip": 0.01161863,
      "auxiliary_loss_mlp": 0.01046726,
      "balance_loss_clip": 1.05471706,
      "balance_loss_mlp": 1.03250444,
      "epoch": 0.10064329946491914,
      "flos": 18660154510080.0,
      "grad_norm": 2.3339767275859016,
      "language_loss": 0.89757961,
      "learning_rate": 3.9479564464212455e-06,
      "loss": 0.91966552,
      "num_input_tokens_seen": 17866825,
      "step": 837,
      "time_per_iteration": 2.645338773727417
    },
    {
      "auxiliary_loss_clip": 0.0119622,
      "auxiliary_loss_mlp": 0.01043331,
      "balance_loss_clip": 1.06596804,
      "balance_loss_mlp": 1.0296514,
      "epoch": 0.10076354235555823,
      "flos": 17199056983680.0,
      "grad_norm": 2.1817872389433974,
      "language_loss": 0.76502991,
      "learning_rate": 3.947779751283414e-06,
      "loss": 0.7874254,
      "num_input_tokens_seen": 17883995,
      "step": 838,
      "time_per_iteration": 2.581969976425171
    },
    {
      "auxiliary_loss_clip": 0.01210735,
      "auxiliary_loss_mlp": 0.01045398,
      "balance_loss_clip": 1.07236063,
      "balance_loss_mlp": 1.03049731,
      "epoch": 0.10088378524619732,
      "flos": 22962252395520.0,
      "grad_norm": 3.0278634325104816,
      "language_loss": 0.75932658,
      "learning_rate": 3.947602760668944e-06,
      "loss": 0.78188795,
      "num_input_tokens_seen": 17903785,
      "step": 839,
      "time_per_iteration": 2.606968641281128
    },
    {
      "auxiliary_loss_clip": 0.01192371,
      "auxiliary_loss_mlp": 0.01059502,
      "balance_loss_clip": 1.06406546,
      "balance_loss_mlp": 1.04346251,
      "epoch": 0.10100402813683641,
      "flos": 37885828746240.0,
      "grad_norm": 1.9697447471009712,
      "language_loss": 0.71759319,
      "learning_rate": 3.947425474604684e-06,
      "loss": 0.74011189,
      "num_input_tokens_seen": 17927720,
      "step": 840,
      "time_per_iteration": 2.725980281829834
    },
    {
      "auxiliary_loss_clip": 0.01206266,
      "auxiliary_loss_mlp": 0.01051961,
      "balance_loss_clip": 1.06945026,
      "balance_loss_mlp": 1.03858519,
      "epoch": 0.1011242710274755,
      "flos": 21543458112000.0,
      "grad_norm": 2.5330155038397137,
      "language_loss": 0.92196941,
      "learning_rate": 3.947247893117528e-06,
      "loss": 0.94455165,
      "num_input_tokens_seen": 17946225,
      "step": 841,
      "time_per_iteration": 2.582261323928833
    },
    {
      "auxiliary_loss_clip": 0.01212913,
      "auxiliary_loss_mlp": 0.01049507,
      "balance_loss_clip": 1.07373166,
      "balance_loss_mlp": 1.033795,
      "epoch": 0.10124451391811459,
      "flos": 13621456419840.0,
      "grad_norm": 3.305871430750937,
      "language_loss": 0.69589943,
      "learning_rate": 3.947070016234413e-06,
      "loss": 0.71852362,
      "num_input_tokens_seen": 17962015,
      "step": 842,
      "time_per_iteration": 2.562213897705078
    },
    {
      "auxiliary_loss_clip": 0.01193071,
      "auxiliary_loss_mlp": 0.01043422,
      "balance_loss_clip": 1.07260895,
      "balance_loss_mlp": 1.02950442,
      "epoch": 0.10136475680875369,
      "flos": 16649228522880.0,
      "grad_norm": 3.6155698059209675,
      "language_loss": 0.74858648,
      "learning_rate": 3.946891843982326e-06,
      "loss": 0.77095145,
      "num_input_tokens_seen": 17979680,
      "step": 843,
      "time_per_iteration": 2.583146572113037
    },
    {
      "auxiliary_loss_clip": 0.01125318,
      "auxiliary_loss_mlp": 0.01043759,
      "balance_loss_clip": 1.0510906,
      "balance_loss_mlp": 1.02810693,
      "epoch": 0.10148499969939277,
      "flos": 19461034103040.0,
      "grad_norm": 3.684678331689521,
      "language_loss": 0.74576443,
      "learning_rate": 3.9467133763882935e-06,
      "loss": 0.76745522,
      "num_input_tokens_seen": 17998145,
      "step": 844,
      "time_per_iteration": 3.0082902908325195
    },
    {
      "auxiliary_loss_clip": 0.01207088,
      "auxiliary_loss_mlp": 0.01044045,
      "balance_loss_clip": 1.06604218,
      "balance_loss_mlp": 1.02938271,
      "epoch": 0.10160524259003187,
      "flos": 21104988791040.0,
      "grad_norm": 1.8872537056346657,
      "language_loss": 0.86355877,
      "learning_rate": 3.9465346134793905e-06,
      "loss": 0.88607013,
      "num_input_tokens_seen": 18017955,
      "step": 845,
      "time_per_iteration": 2.850191831588745
    },
    {
      "auxiliary_loss_clip": 0.01134399,
      "auxiliary_loss_mlp": 0.01040801,
      "balance_loss_clip": 1.05199623,
      "balance_loss_mlp": 1.02683008,
      "epoch": 0.10172548548067095,
      "flos": 17712687513600.0,
      "grad_norm": 2.25634488558993,
      "language_loss": 0.79880774,
      "learning_rate": 3.9463555552827335e-06,
      "loss": 0.82055974,
      "num_input_tokens_seen": 18035125,
      "step": 846,
      "time_per_iteration": 2.7756619453430176
    },
    {
      "auxiliary_loss_clip": 0.01219895,
      "auxiliary_loss_mlp": 0.01043723,
      "balance_loss_clip": 1.0697999,
      "balance_loss_mlp": 1.02907252,
      "epoch": 0.10184572837131005,
      "flos": 21104845136640.0,
      "grad_norm": 3.049652993902711,
      "language_loss": 0.86080176,
      "learning_rate": 3.946176201825487e-06,
      "loss": 0.88343799,
      "num_input_tokens_seen": 18053160,
      "step": 847,
      "time_per_iteration": 3.9099183082580566
    },
    {
      "auxiliary_loss_clip": 0.01216452,
      "auxiliary_loss_mlp": 0.01042489,
      "balance_loss_clip": 1.06768584,
      "balance_loss_mlp": 1.02806497,
      "epoch": 0.10196597126194913,
      "flos": 26067591918720.0,
      "grad_norm": 2.172911733060853,
      "language_loss": 0.83580518,
      "learning_rate": 3.9459965531348575e-06,
      "loss": 0.85839462,
      "num_input_tokens_seen": 18072815,
      "step": 848,
      "time_per_iteration": 3.6096181869506836
    },
    {
      "auxiliary_loss_clip": 0.01190872,
      "auxiliary_loss_mlp": 0.01046491,
      "balance_loss_clip": 1.06111324,
      "balance_loss_mlp": 1.03203106,
      "epoch": 0.10208621415258823,
      "flos": 29314634595840.0,
      "grad_norm": 2.213691964465037,
      "language_loss": 0.85669178,
      "learning_rate": 3.945816609238098e-06,
      "loss": 0.87906539,
      "num_input_tokens_seen": 18092225,
      "step": 849,
      "time_per_iteration": 2.7153854370117188
    },
    {
      "auxiliary_loss_clip": 0.01211455,
      "auxiliary_loss_mlp": 0.01046874,
      "balance_loss_clip": 1.06959081,
      "balance_loss_mlp": 1.03233671,
      "epoch": 0.10220645704322733,
      "flos": 23805794367360.0,
      "grad_norm": 2.0103398959925802,
      "language_loss": 0.85123289,
      "learning_rate": 3.945636370162507e-06,
      "loss": 0.87381619,
      "num_input_tokens_seen": 18112335,
      "step": 850,
      "time_per_iteration": 3.3869006633758545
    },
    {
      "auxiliary_loss_clip": 0.01237898,
      "auxiliary_loss_mlp": 0.01041739,
      "balance_loss_clip": 1.07602632,
      "balance_loss_mlp": 1.02943683,
      "epoch": 0.10232669993386641,
      "flos": 23218546913280.0,
      "grad_norm": 1.8492016446611736,
      "language_loss": 0.79049587,
      "learning_rate": 3.945455835935425e-06,
      "loss": 0.81329221,
      "num_input_tokens_seen": 18131520,
      "step": 851,
      "time_per_iteration": 2.532423734664917
    },
    {
      "auxiliary_loss_clip": 0.01158261,
      "auxiliary_loss_mlp": 0.01051445,
      "balance_loss_clip": 1.05706823,
      "balance_loss_mlp": 1.03629398,
      "epoch": 0.1024469428245055,
      "flos": 22922929981440.0,
      "grad_norm": 2.5051165176179504,
      "language_loss": 0.75070375,
      "learning_rate": 3.94527500658424e-06,
      "loss": 0.7728008,
      "num_input_tokens_seen": 18149185,
      "step": 852,
      "time_per_iteration": 2.652567148208618
    },
    {
      "auxiliary_loss_clip": 0.01242937,
      "auxiliary_loss_mlp": 0.0103984,
      "balance_loss_clip": 1.07803977,
      "balance_loss_mlp": 1.02539778,
      "epoch": 0.10256718571514459,
      "flos": 31359495957120.0,
      "grad_norm": 1.9083268170422893,
      "language_loss": 0.81132305,
      "learning_rate": 3.945093882136382e-06,
      "loss": 0.83415091,
      "num_input_tokens_seen": 18172960,
      "step": 853,
      "time_per_iteration": 3.373048782348633
    },
    {
      "auxiliary_loss_clip": 0.01188746,
      "auxiliary_loss_mlp": 0.0104037,
      "balance_loss_clip": 1.0634644,
      "balance_loss_mlp": 1.02671456,
      "epoch": 0.10268742860578368,
      "flos": 23474877344640.0,
      "grad_norm": 2.0774016133961104,
      "language_loss": 0.84695256,
      "learning_rate": 3.944912462619329e-06,
      "loss": 0.86924374,
      "num_input_tokens_seen": 18191925,
      "step": 854,
      "time_per_iteration": 2.7142140865325928
    },
    {
      "auxiliary_loss_clip": 0.01220987,
      "auxiliary_loss_mlp": 0.01044465,
      "balance_loss_clip": 1.07024646,
      "balance_loss_mlp": 1.02864563,
      "epoch": 0.10280767149642277,
      "flos": 25520313323520.0,
      "grad_norm": 2.2734950267612666,
      "language_loss": 0.80374777,
      "learning_rate": 3.9447307480606025e-06,
      "loss": 0.82640231,
      "num_input_tokens_seen": 18212010,
      "step": 855,
      "time_per_iteration": 2.565293312072754
    },
    {
      "auxiliary_loss_clip": 0.01121201,
      "auxiliary_loss_mlp": 0.01043051,
      "balance_loss_clip": 1.05438805,
      "balance_loss_mlp": 1.02901363,
      "epoch": 0.10292791438706186,
      "flos": 17347691462400.0,
      "grad_norm": 3.573685893793923,
      "language_loss": 0.90098161,
      "learning_rate": 3.944548738487767e-06,
      "loss": 0.92262417,
      "num_input_tokens_seen": 18229525,
      "step": 856,
      "time_per_iteration": 2.7497429847717285
    },
    {
      "auxiliary_loss_clip": 0.01227719,
      "auxiliary_loss_mlp": 0.01045812,
      "balance_loss_clip": 1.07204735,
      "balance_loss_mlp": 1.03128004,
      "epoch": 0.10304815727770096,
      "flos": 27052693390080.0,
      "grad_norm": 2.385227426961414,
      "language_loss": 0.90730345,
      "learning_rate": 3.944366433928434e-06,
      "loss": 0.93003869,
      "num_input_tokens_seen": 18249505,
      "step": 857,
      "time_per_iteration": 2.750467538833618
    },
    {
      "auxiliary_loss_clip": 0.01201453,
      "auxiliary_loss_mlp": 0.01041566,
      "balance_loss_clip": 1.06712425,
      "balance_loss_mlp": 1.02912033,
      "epoch": 0.10316840016834004,
      "flos": 22782591544320.0,
      "grad_norm": 1.6128379921871308,
      "language_loss": 0.83658475,
      "learning_rate": 3.9441838344102594e-06,
      "loss": 0.85901493,
      "num_input_tokens_seen": 18269230,
      "step": 858,
      "time_per_iteration": 2.6212432384490967
    },
    {
      "auxiliary_loss_clip": 0.01193344,
      "auxiliary_loss_mlp": 0.01043263,
      "balance_loss_clip": 1.06644344,
      "balance_loss_mlp": 1.02940512,
      "epoch": 0.10328864305897914,
      "flos": 20704584908160.0,
      "grad_norm": 2.386253646351055,
      "language_loss": 0.67645985,
      "learning_rate": 3.944000939960943e-06,
      "loss": 0.6988259,
      "num_input_tokens_seen": 18287955,
      "step": 859,
      "time_per_iteration": 2.5577807426452637
    },
    {
      "auxiliary_loss_clip": 0.01255578,
      "auxiliary_loss_mlp": 0.01041015,
      "balance_loss_clip": 1.0747776,
      "balance_loss_mlp": 1.02848029,
      "epoch": 0.10340888594961822,
      "flos": 28478814048000.0,
      "grad_norm": 1.515526199042971,
      "language_loss": 0.80022901,
      "learning_rate": 3.943817750608229e-06,
      "loss": 0.82319492,
      "num_input_tokens_seen": 18310505,
      "step": 860,
      "time_per_iteration": 2.589761257171631
    },
    {
      "auxiliary_loss_clip": 0.01209509,
      "auxiliary_loss_mlp": 0.01041876,
      "balance_loss_clip": 1.06455362,
      "balance_loss_mlp": 1.02771401,
      "epoch": 0.10352912884025732,
      "flos": 13370333460480.0,
      "grad_norm": 2.3994047691714435,
      "language_loss": 0.82426083,
      "learning_rate": 3.943634266379908e-06,
      "loss": 0.8467747,
      "num_input_tokens_seen": 18327400,
      "step": 861,
      "time_per_iteration": 2.5170817375183105
    },
    {
      "auxiliary_loss_clip": 0.01204622,
      "auxiliary_loss_mlp": 0.01048632,
      "balance_loss_clip": 1.06937659,
      "balance_loss_mlp": 1.03427315,
      "epoch": 0.10364937173089642,
      "flos": 25558558329600.0,
      "grad_norm": 1.9002821751486765,
      "language_loss": 0.84937608,
      "learning_rate": 3.943450487303815e-06,
      "loss": 0.87190861,
      "num_input_tokens_seen": 18347895,
      "step": 862,
      "time_per_iteration": 2.6753017902374268
    },
    {
      "auxiliary_loss_clip": 0.01208494,
      "auxiliary_loss_mlp": 0.01045602,
      "balance_loss_clip": 1.06501269,
      "balance_loss_mlp": 1.0306592,
      "epoch": 0.1037696146215355,
      "flos": 21215486004480.0,
      "grad_norm": 1.7984880709431788,
      "language_loss": 0.85407698,
      "learning_rate": 3.943266413407827e-06,
      "loss": 0.87661803,
      "num_input_tokens_seen": 18367170,
      "step": 863,
      "time_per_iteration": 2.6124658584594727
    },
    {
      "auxiliary_loss_clip": 0.0121082,
      "auxiliary_loss_mlp": 0.01042511,
      "balance_loss_clip": 1.06633568,
      "balance_loss_mlp": 1.02977347,
      "epoch": 0.1038898575121746,
      "flos": 25807382818560.0,
      "grad_norm": 1.9411807888157466,
      "language_loss": 0.84839058,
      "learning_rate": 3.94308204471987e-06,
      "loss": 0.870924,
      "num_input_tokens_seen": 18386185,
      "step": 864,
      "time_per_iteration": 2.623241424560547
    },
    {
      "auxiliary_loss_clip": 0.01216369,
      "auxiliary_loss_mlp": 0.01047083,
      "balance_loss_clip": 1.07225657,
      "balance_loss_mlp": 1.03391063,
      "epoch": 0.10401010040281368,
      "flos": 19062425900160.0,
      "grad_norm": 2.2763750699524823,
      "language_loss": 0.74796653,
      "learning_rate": 3.942897381267912e-06,
      "loss": 0.77060103,
      "num_input_tokens_seen": 18402550,
      "step": 865,
      "time_per_iteration": 2.5738019943237305
    },
    {
      "auxiliary_loss_clip": 0.01188506,
      "auxiliary_loss_mlp": 0.01043826,
      "balance_loss_clip": 1.06215632,
      "balance_loss_mlp": 1.02931786,
      "epoch": 0.10413034329345278,
      "flos": 16355119962240.0,
      "grad_norm": 2.7043823988437445,
      "language_loss": 0.66326421,
      "learning_rate": 3.942712423079965e-06,
      "loss": 0.68558753,
      "num_input_tokens_seen": 18418940,
      "step": 866,
      "time_per_iteration": 2.591092348098755
    },
    {
      "auxiliary_loss_clip": 0.01197098,
      "auxiliary_loss_mlp": 0.01036366,
      "balance_loss_clip": 1.06496894,
      "balance_loss_mlp": 1.0232414,
      "epoch": 0.10425058618409186,
      "flos": 17236511890560.0,
      "grad_norm": 2.271892509172447,
      "language_loss": 0.90049791,
      "learning_rate": 3.942527170184088e-06,
      "loss": 0.92283249,
      "num_input_tokens_seen": 18435560,
      "step": 867,
      "time_per_iteration": 2.5413784980773926
    },
    {
      "auxiliary_loss_clip": 0.0119638,
      "auxiliary_loss_mlp": 0.01050634,
      "balance_loss_clip": 1.06484604,
      "balance_loss_mlp": 1.03607893,
      "epoch": 0.10437082907473096,
      "flos": 17967365919360.0,
      "grad_norm": 2.330742563763357,
      "language_loss": 0.77337605,
      "learning_rate": 3.942341622608385e-06,
      "loss": 0.79584616,
      "num_input_tokens_seen": 18452590,
      "step": 868,
      "time_per_iteration": 2.6620025634765625
    },
    {
      "auxiliary_loss_clip": 0.01197132,
      "auxiliary_loss_mlp": 0.01044965,
      "balance_loss_clip": 1.06975603,
      "balance_loss_mlp": 1.03110147,
      "epoch": 0.10449107196537005,
      "flos": 36283315374720.0,
      "grad_norm": 1.75741537752318,
      "language_loss": 0.77997088,
      "learning_rate": 3.942155780381001e-06,
      "loss": 0.80239177,
      "num_input_tokens_seen": 18476325,
      "step": 869,
      "time_per_iteration": 2.741560697555542
    },
    {
      "auxiliary_loss_clip": 0.01175631,
      "auxiliary_loss_mlp": 0.01040131,
      "balance_loss_clip": 1.05968916,
      "balance_loss_mlp": 1.02545619,
      "epoch": 0.10461131485600914,
      "flos": 23802095266560.0,
      "grad_norm": 1.884428086179851,
      "language_loss": 0.76234663,
      "learning_rate": 3.94196964353013e-06,
      "loss": 0.78450429,
      "num_input_tokens_seen": 18495775,
      "step": 870,
      "time_per_iteration": 2.6622979640960693
    },
    {
      "auxiliary_loss_clip": 0.01220132,
      "auxiliary_loss_mlp": 0.00765334,
      "balance_loss_clip": 1.07309818,
      "balance_loss_mlp": 1.00075829,
      "epoch": 0.10473155774664823,
      "flos": 18405476104320.0,
      "grad_norm": 1.965654932186182,
      "language_loss": 0.81019652,
      "learning_rate": 3.941783212084008e-06,
      "loss": 0.83005118,
      "num_input_tokens_seen": 18513530,
      "step": 871,
      "time_per_iteration": 2.548255681991577
    },
    {
      "auxiliary_loss_clip": 0.0118975,
      "auxiliary_loss_mlp": 0.01043766,
      "balance_loss_clip": 1.0698278,
      "balance_loss_mlp": 1.03030181,
      "epoch": 0.10485180063728732,
      "flos": 25592637358080.0,
      "grad_norm": 2.7402323974941343,
      "language_loss": 0.78594887,
      "learning_rate": 3.941596486070916e-06,
      "loss": 0.80828404,
      "num_input_tokens_seen": 18531575,
      "step": 872,
      "time_per_iteration": 2.6290245056152344
    },
    {
      "auxiliary_loss_clip": 0.01181822,
      "auxiliary_loss_mlp": 0.01047187,
      "balance_loss_clip": 1.0597105,
      "balance_loss_mlp": 1.03247714,
      "epoch": 0.10497204352792641,
      "flos": 27088747666560.0,
      "grad_norm": 3.4246121773412,
      "language_loss": 0.58668268,
      "learning_rate": 3.941409465519182e-06,
      "loss": 0.60897273,
      "num_input_tokens_seen": 18552100,
      "step": 873,
      "time_per_iteration": 4.111186742782593
    },
    {
      "auxiliary_loss_clip": 0.01233194,
      "auxiliary_loss_mlp": 0.01048759,
      "balance_loss_clip": 1.07133698,
      "balance_loss_mlp": 1.03435302,
      "epoch": 0.10509228641856551,
      "flos": 32858479353600.0,
      "grad_norm": 3.2115179935011926,
      "language_loss": 0.85377187,
      "learning_rate": 3.941222150457176e-06,
      "loss": 0.87659144,
      "num_input_tokens_seen": 18575355,
      "step": 874,
      "time_per_iteration": 2.6669631004333496
    },
    {
      "auxiliary_loss_clip": 0.01224481,
      "auxiliary_loss_mlp": 0.0103896,
      "balance_loss_clip": 1.07200491,
      "balance_loss_mlp": 1.02393329,
      "epoch": 0.10521252930920459,
      "flos": 14319165173760.0,
      "grad_norm": 7.377447387245225,
      "language_loss": 0.71728057,
      "learning_rate": 3.941034540913311e-06,
      "loss": 0.73991495,
      "num_input_tokens_seen": 18592885,
      "step": 875,
      "time_per_iteration": 2.511070966720581
    },
    {
      "auxiliary_loss_clip": 0.01197141,
      "auxiliary_loss_mlp": 0.01051897,
      "balance_loss_clip": 1.06766176,
      "balance_loss_mlp": 1.03700149,
      "epoch": 0.10533277219984369,
      "flos": 21687028773120.0,
      "grad_norm": 2.057787854941557,
      "language_loss": 0.82494807,
      "learning_rate": 3.940846636916051e-06,
      "loss": 0.84743845,
      "num_input_tokens_seen": 18612920,
      "step": 876,
      "time_per_iteration": 2.594775438308716
    },
    {
      "auxiliary_loss_clip": 0.01196555,
      "auxiliary_loss_mlp": 0.01043269,
      "balance_loss_clip": 1.06613278,
      "balance_loss_mlp": 1.03071642,
      "epoch": 0.10545301509048277,
      "flos": 22269787027200.0,
      "grad_norm": 2.046127893368992,
      "language_loss": 0.86572671,
      "learning_rate": 3.940658438493899e-06,
      "loss": 0.88812494,
      "num_input_tokens_seen": 18630765,
      "step": 877,
      "time_per_iteration": 3.3123316764831543
    },
    {
      "auxiliary_loss_clip": 0.01183623,
      "auxiliary_loss_mlp": 0.00766774,
      "balance_loss_clip": 1.06410289,
      "balance_loss_mlp": 1.00076556,
      "epoch": 0.10557325798112187,
      "flos": 22199725549440.0,
      "grad_norm": 2.1012974603734245,
      "language_loss": 0.75613421,
      "learning_rate": 3.940469945675405e-06,
      "loss": 0.77563822,
      "num_input_tokens_seen": 18649150,
      "step": 878,
      "time_per_iteration": 3.464261293411255
    },
    {
      "auxiliary_loss_clip": 0.01214876,
      "auxiliary_loss_mlp": 0.01037106,
      "balance_loss_clip": 1.07161617,
      "balance_loss_mlp": 1.02516699,
      "epoch": 0.10569350087176095,
      "flos": 25775889569280.0,
      "grad_norm": 1.9397513658467318,
      "language_loss": 0.91293311,
      "learning_rate": 3.940281158489163e-06,
      "loss": 0.93545294,
      "num_input_tokens_seen": 18668380,
      "step": 879,
      "time_per_iteration": 2.590463399887085
    },
    {
      "auxiliary_loss_clip": 0.01203817,
      "auxiliary_loss_mlp": 0.01043596,
      "balance_loss_clip": 1.06745076,
      "balance_loss_mlp": 1.02952361,
      "epoch": 0.10581374376240005,
      "flos": 17311385790720.0,
      "grad_norm": 1.8151465410661773,
      "language_loss": 0.8291586,
      "learning_rate": 3.940092076963812e-06,
      "loss": 0.85163271,
      "num_input_tokens_seen": 18685875,
      "step": 880,
      "time_per_iteration": 2.533750057220459
    },
    {
      "auxiliary_loss_clip": 0.01204741,
      "auxiliary_loss_mlp": 0.0104839,
      "balance_loss_clip": 1.06324172,
      "balance_loss_mlp": 1.03282762,
      "epoch": 0.10593398665303914,
      "flos": 34349454017280.0,
      "grad_norm": 2.061149073509295,
      "language_loss": 0.7883867,
      "learning_rate": 3.9399027011280355e-06,
      "loss": 0.81091797,
      "num_input_tokens_seen": 18707970,
      "step": 881,
      "time_per_iteration": 2.672267198562622
    },
    {
      "auxiliary_loss_clip": 0.01187291,
      "auxiliary_loss_mlp": 0.01040068,
      "balance_loss_clip": 1.06646121,
      "balance_loss_mlp": 1.02510095,
      "epoch": 0.10605422954367823,
      "flos": 23257977068160.0,
      "grad_norm": 2.021208637605614,
      "language_loss": 0.77320427,
      "learning_rate": 3.939713031010561e-06,
      "loss": 0.79547787,
      "num_input_tokens_seen": 18726335,
      "step": 882,
      "time_per_iteration": 2.6062216758728027
    },
    {
      "auxiliary_loss_clip": 0.01184745,
      "auxiliary_loss_mlp": 0.00766018,
      "balance_loss_clip": 1.06888723,
      "balance_loss_mlp": 1.00065184,
      "epoch": 0.10617447243431732,
      "flos": 22820118278400.0,
      "grad_norm": 3.3212040238126566,
      "language_loss": 0.77791297,
      "learning_rate": 3.939523066640163e-06,
      "loss": 0.79742062,
      "num_input_tokens_seen": 18745230,
      "step": 883,
      "time_per_iteration": 2.682677745819092
    },
    {
      "auxiliary_loss_clip": 0.01253067,
      "auxiliary_loss_mlp": 0.0104434,
      "balance_loss_clip": 1.07587206,
      "balance_loss_mlp": 1.03098226,
      "epoch": 0.10629471532495641,
      "flos": 24386577373440.0,
      "grad_norm": 1.7696738516851398,
      "language_loss": 0.81173611,
      "learning_rate": 3.939332808045657e-06,
      "loss": 0.83471012,
      "num_input_tokens_seen": 18764880,
      "step": 884,
      "time_per_iteration": 2.5243101119995117
    },
    {
      "auxiliary_loss_clip": 0.01212781,
      "auxiliary_loss_mlp": 0.0104217,
      "balance_loss_clip": 1.06626058,
      "balance_loss_mlp": 1.02933741,
      "epoch": 0.1064149582155955,
      "flos": 21105491581440.0,
      "grad_norm": 1.770545712757226,
      "language_loss": 0.84746087,
      "learning_rate": 3.939142255255906e-06,
      "loss": 0.87001038,
      "num_input_tokens_seen": 18785765,
      "step": 885,
      "time_per_iteration": 2.572498321533203
    },
    {
      "auxiliary_loss_clip": 0.01248707,
      "auxiliary_loss_mlp": 0.01040461,
      "balance_loss_clip": 1.07233024,
      "balance_loss_mlp": 1.02741313,
      "epoch": 0.1065352011062346,
      "flos": 20702035042560.0,
      "grad_norm": 1.9370988594423155,
      "language_loss": 0.86894405,
      "learning_rate": 3.938951408299817e-06,
      "loss": 0.89183581,
      "num_input_tokens_seen": 18804605,
      "step": 886,
      "time_per_iteration": 2.47084641456604
    },
    {
      "auxiliary_loss_clip": 0.01095683,
      "auxiliary_loss_mlp": 0.0075967,
      "balance_loss_clip": 1.04608774,
      "balance_loss_mlp": 1.00021577,
      "epoch": 0.10665544399687368,
      "flos": 62659632689280.0,
      "grad_norm": 0.7930617263440134,
      "language_loss": 0.54424465,
      "learning_rate": 3.938760267206342e-06,
      "loss": 0.56279826,
      "num_input_tokens_seen": 18866425,
      "step": 887,
      "time_per_iteration": 3.1559910774230957
    },
    {
      "auxiliary_loss_clip": 0.01186217,
      "auxiliary_loss_mlp": 0.01043188,
      "balance_loss_clip": 1.06918418,
      "balance_loss_mlp": 1.02829885,
      "epoch": 0.10677568688751278,
      "flos": 26140382830080.0,
      "grad_norm": 2.2005996258939926,
      "language_loss": 0.78618485,
      "learning_rate": 3.938568832004475e-06,
      "loss": 0.80847889,
      "num_input_tokens_seen": 18885130,
      "step": 888,
      "time_per_iteration": 2.637180805206299
    },
    {
      "auxiliary_loss_clip": 0.01203869,
      "auxiliary_loss_mlp": 0.01043718,
      "balance_loss_clip": 1.06724548,
      "balance_loss_mlp": 1.02977097,
      "epoch": 0.10689592977815186,
      "flos": 12786533712000.0,
      "grad_norm": 1.9899028292709258,
      "language_loss": 0.7523098,
      "learning_rate": 3.938377102723257e-06,
      "loss": 0.77478564,
      "num_input_tokens_seen": 18902265,
      "step": 889,
      "time_per_iteration": 2.575805187225342
    },
    {
      "auxiliary_loss_clip": 0.01216791,
      "auxiliary_loss_mlp": 0.01041937,
      "balance_loss_clip": 1.07026982,
      "balance_loss_mlp": 1.02702391,
      "epoch": 0.10701617266879096,
      "flos": 22126683242880.0,
      "grad_norm": 2.417839226307633,
      "language_loss": 0.8330915,
      "learning_rate": 3.938185079391774e-06,
      "loss": 0.8556788,
      "num_input_tokens_seen": 18919310,
      "step": 890,
      "time_per_iteration": 2.574413776397705
    },
    {
      "auxiliary_loss_clip": 0.01201291,
      "auxiliary_loss_mlp": 0.00766013,
      "balance_loss_clip": 1.06625187,
      "balance_loss_mlp": 1.00076127,
      "epoch": 0.10713641555943004,
      "flos": 19745625559680.0,
      "grad_norm": 3.7039439180833873,
      "language_loss": 1.0584929,
      "learning_rate": 3.937992762039157e-06,
      "loss": 1.07816601,
      "num_input_tokens_seen": 18932635,
      "step": 891,
      "time_per_iteration": 2.595458984375
    },
    {
      "auxiliary_loss_clip": 0.01203093,
      "auxiliary_loss_mlp": 0.01044311,
      "balance_loss_clip": 1.06362939,
      "balance_loss_mlp": 1.03261685,
      "epoch": 0.10725665845006914,
      "flos": 23952992302080.0,
      "grad_norm": 1.7240318979572535,
      "language_loss": 0.80298245,
      "learning_rate": 3.937800150694577e-06,
      "loss": 0.8254565,
      "num_input_tokens_seen": 18953810,
      "step": 892,
      "time_per_iteration": 2.5926191806793213
    },
    {
      "auxiliary_loss_clip": 0.01225354,
      "auxiliary_loss_mlp": 0.01042379,
      "balance_loss_clip": 1.07089138,
      "balance_loss_mlp": 1.02871132,
      "epoch": 0.10737690134070824,
      "flos": 18551704371840.0,
      "grad_norm": 2.3072631609985943,
      "language_loss": 0.7640062,
      "learning_rate": 3.937607245387255e-06,
      "loss": 0.78668356,
      "num_input_tokens_seen": 18973175,
      "step": 893,
      "time_per_iteration": 2.560758590698242
    },
    {
      "auxiliary_loss_clip": 0.0119794,
      "auxiliary_loss_mlp": 0.01039688,
      "balance_loss_clip": 1.06329226,
      "balance_loss_mlp": 1.02768373,
      "epoch": 0.10749714423134732,
      "flos": 22707609903360.0,
      "grad_norm": 1.8428034638489703,
      "language_loss": 0.71896613,
      "learning_rate": 3.937414046146455e-06,
      "loss": 0.74134243,
      "num_input_tokens_seen": 18991130,
      "step": 894,
      "time_per_iteration": 2.566373109817505
    },
    {
      "auxiliary_loss_clip": 0.01212923,
      "auxiliary_loss_mlp": 0.01042509,
      "balance_loss_clip": 1.07047653,
      "balance_loss_mlp": 1.02823913,
      "epoch": 0.10761738712198642,
      "flos": 21106066199040.0,
      "grad_norm": 2.8810970251079873,
      "language_loss": 0.75962281,
      "learning_rate": 3.9372205530014845e-06,
      "loss": 0.78217709,
      "num_input_tokens_seen": 19009610,
      "step": 895,
      "time_per_iteration": 2.579453229904175
    },
    {
      "auxiliary_loss_clip": 0.01166676,
      "auxiliary_loss_mlp": 0.01047751,
      "balance_loss_clip": 1.05855298,
      "balance_loss_mlp": 1.03356552,
      "epoch": 0.1077376300126255,
      "flos": 23766723348480.0,
      "grad_norm": 3.2884393057718926,
      "language_loss": 0.71570909,
      "learning_rate": 3.937026765981696e-06,
      "loss": 0.73785335,
      "num_input_tokens_seen": 19029680,
      "step": 896,
      "time_per_iteration": 2.6462671756744385
    },
    {
      "auxiliary_loss_clip": 0.01182511,
      "auxiliary_loss_mlp": 0.01050707,
      "balance_loss_clip": 1.06377912,
      "balance_loss_mlp": 1.03650296,
      "epoch": 0.1078578729032646,
      "flos": 20919581763840.0,
      "grad_norm": 1.8269610129355556,
      "language_loss": 0.79527903,
      "learning_rate": 3.936832685116488e-06,
      "loss": 0.81761116,
      "num_input_tokens_seen": 19047775,
      "step": 897,
      "time_per_iteration": 2.6102874279022217
    },
    {
      "auxiliary_loss_clip": 0.01235062,
      "auxiliary_loss_mlp": 0.01036238,
      "balance_loss_clip": 1.06985736,
      "balance_loss_mlp": 1.02212405,
      "epoch": 0.10797811579390369,
      "flos": 14829886702080.0,
      "grad_norm": 9.027780430225858,
      "language_loss": 0.90311027,
      "learning_rate": 3.936638310435301e-06,
      "loss": 0.92582327,
      "num_input_tokens_seen": 19065640,
      "step": 898,
      "time_per_iteration": 2.4740183353424072
    },
    {
      "auxiliary_loss_clip": 0.01213913,
      "auxiliary_loss_mlp": 0.01049445,
      "balance_loss_clip": 1.06781054,
      "balance_loss_mlp": 1.03528905,
      "epoch": 0.10809835868454278,
      "flos": 19536985411200.0,
      "grad_norm": 1.9621274393795667,
      "language_loss": 0.81945938,
      "learning_rate": 3.936443641967623e-06,
      "loss": 0.84209287,
      "num_input_tokens_seen": 19084470,
      "step": 899,
      "time_per_iteration": 4.070073366165161
    },
    {
      "auxiliary_loss_clip": 0.01216458,
      "auxiliary_loss_mlp": 0.00766193,
      "balance_loss_clip": 1.06411624,
      "balance_loss_mlp": 1.00061059,
      "epoch": 0.10821860157518187,
      "flos": 18442320480000.0,
      "grad_norm": 2.0880274215355032,
      "language_loss": 0.83069742,
      "learning_rate": 3.936248679742983e-06,
      "loss": 0.85052389,
      "num_input_tokens_seen": 19102965,
      "step": 900,
      "time_per_iteration": 2.5632741451263428
    },
    {
      "auxiliary_loss_clip": 0.01031428,
      "auxiliary_loss_mlp": 0.01005311,
      "balance_loss_clip": 1.02901816,
      "balance_loss_mlp": 1.00242662,
      "epoch": 0.10833884446582095,
      "flos": 49359468447360.0,
      "grad_norm": 1.0535660561450408,
      "language_loss": 0.70188135,
      "learning_rate": 3.936053423790959e-06,
      "loss": 0.72224873,
      "num_input_tokens_seen": 19151285,
      "step": 901,
      "time_per_iteration": 3.2843010425567627
    },
    {
      "auxiliary_loss_clip": 0.01185475,
      "auxiliary_loss_mlp": 0.01039669,
      "balance_loss_clip": 1.06456208,
      "balance_loss_mlp": 1.02585816,
      "epoch": 0.10845908735646005,
      "flos": 20411912891520.0,
      "grad_norm": 2.047487289802419,
      "language_loss": 0.77247477,
      "learning_rate": 3.935857874141168e-06,
      "loss": 0.79472613,
      "num_input_tokens_seen": 19170120,
      "step": 902,
      "time_per_iteration": 4.0000410079956055
    },
    {
      "auxiliary_loss_clip": 0.01227341,
      "auxiliary_loss_mlp": 0.01039854,
      "balance_loss_clip": 1.06674945,
      "balance_loss_mlp": 1.02649093,
      "epoch": 0.10857933024709913,
      "flos": 14027750133120.0,
      "grad_norm": 2.073580122456351,
      "language_loss": 0.84034705,
      "learning_rate": 3.935662030823279e-06,
      "loss": 0.86301899,
      "num_input_tokens_seen": 19186305,
      "step": 903,
      "time_per_iteration": 3.3097851276397705
    },
    {
      "auxiliary_loss_clip": 0.01229085,
      "auxiliary_loss_mlp": 0.01046281,
      "balance_loss_clip": 1.06786764,
      "balance_loss_mlp": 1.03220224,
      "epoch": 0.10869957313773823,
      "flos": 13369004657280.0,
      "grad_norm": 2.124764084580887,
      "language_loss": 0.72498751,
      "learning_rate": 3.935465893866998e-06,
      "loss": 0.7477411,
      "num_input_tokens_seen": 19204530,
      "step": 904,
      "time_per_iteration": 3.2366373538970947
    },
    {
      "auxiliary_loss_clip": 0.01138092,
      "auxiliary_loss_mlp": 0.01045074,
      "balance_loss_clip": 1.05598569,
      "balance_loss_mlp": 1.03202689,
      "epoch": 0.10881981602837733,
      "flos": 25807095509760.0,
      "grad_norm": 1.860425892440872,
      "language_loss": 0.80162394,
      "learning_rate": 3.935269463302079e-06,
      "loss": 0.82345557,
      "num_input_tokens_seen": 19222735,
      "step": 905,
      "time_per_iteration": 2.7384297847747803
    },
    {
      "auxiliary_loss_clip": 0.01174418,
      "auxiliary_loss_mlp": 0.0103646,
      "balance_loss_clip": 1.0642786,
      "balance_loss_mlp": 1.02268553,
      "epoch": 0.10894005891901641,
      "flos": 20777555387520.0,
      "grad_norm": 1.7203673377637874,
      "language_loss": 0.76885927,
      "learning_rate": 3.935072739158322e-06,
      "loss": 0.79096806,
      "num_input_tokens_seen": 19242445,
      "step": 906,
      "time_per_iteration": 2.6317200660705566
    },
    {
      "auxiliary_loss_clip": 0.01198343,
      "auxiliary_loss_mlp": 0.01043915,
      "balance_loss_clip": 1.06725824,
      "balance_loss_mlp": 1.03034329,
      "epoch": 0.10906030180965551,
      "flos": 26649883296000.0,
      "grad_norm": 1.5135806443818385,
      "language_loss": 0.79666531,
      "learning_rate": 3.934875721465569e-06,
      "loss": 0.81908792,
      "num_input_tokens_seen": 19262865,
      "step": 907,
      "time_per_iteration": 2.6111178398132324
    },
    {
      "auxiliary_loss_clip": 0.01207621,
      "auxiliary_loss_mlp": 0.01051096,
      "balance_loss_clip": 1.07218838,
      "balance_loss_mlp": 1.03491974,
      "epoch": 0.10918054470029459,
      "flos": 36534402420480.0,
      "grad_norm": 2.6853264277897186,
      "language_loss": 0.71476108,
      "learning_rate": 3.9346784102537076e-06,
      "loss": 0.7373482,
      "num_input_tokens_seen": 19285000,
      "step": 908,
      "time_per_iteration": 2.7180309295654297
    },
    {
      "auxiliary_loss_clip": 0.01166727,
      "auxiliary_loss_mlp": 0.00766579,
      "balance_loss_clip": 1.05902338,
      "balance_loss_mlp": 1.00052571,
      "epoch": 0.10930078759093369,
      "flos": 21762549118080.0,
      "grad_norm": 2.3933581599377507,
      "language_loss": 0.78287417,
      "learning_rate": 3.934480805552669e-06,
      "loss": 0.80220723,
      "num_input_tokens_seen": 19306010,
      "step": 909,
      "time_per_iteration": 2.6372597217559814
    },
    {
      "auxiliary_loss_clip": 0.01236257,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.06782877,
      "balance_loss_mlp": 1.02399373,
      "epoch": 0.10942103048157277,
      "flos": 22601781457920.0,
      "grad_norm": 2.0766177620537816,
      "language_loss": 0.87920487,
      "learning_rate": 3.93428290739243e-06,
      "loss": 0.90195173,
      "num_input_tokens_seen": 19325380,
      "step": 910,
      "time_per_iteration": 2.5318233966827393
    },
    {
      "auxiliary_loss_clip": 0.01168445,
      "auxiliary_loss_mlp": 0.01037601,
      "balance_loss_clip": 1.05413079,
      "balance_loss_mlp": 1.02533484,
      "epoch": 0.10954127337221187,
      "flos": 15045781397760.0,
      "grad_norm": 4.033478373283635,
      "language_loss": 0.79770648,
      "learning_rate": 3.9340847158030125e-06,
      "loss": 0.819767,
      "num_input_tokens_seen": 19338960,
      "step": 911,
      "time_per_iteration": 2.5872583389282227
    },
    {
      "auxiliary_loss_clip": 0.01178495,
      "auxiliary_loss_mlp": 0.01036344,
      "balance_loss_clip": 1.06112039,
      "balance_loss_mlp": 1.0225935,
      "epoch": 0.10966151626285096,
      "flos": 21650974496640.0,
      "grad_norm": 1.718193232635442,
      "language_loss": 0.75457501,
      "learning_rate": 3.9338862308144814e-06,
      "loss": 0.77672338,
      "num_input_tokens_seen": 19357780,
      "step": 912,
      "time_per_iteration": 2.612847089767456
    },
    {
      "auxiliary_loss_clip": 0.012109,
      "auxiliary_loss_mlp": 0.01041888,
      "balance_loss_clip": 1.06650662,
      "balance_loss_mlp": 1.02771413,
      "epoch": 0.10978175915349005,
      "flos": 20121359777280.0,
      "grad_norm": 1.6094010681820003,
      "language_loss": 0.84799105,
      "learning_rate": 3.933687452456946e-06,
      "loss": 0.87051892,
      "num_input_tokens_seen": 19377680,
      "step": 913,
      "time_per_iteration": 2.517390727996826
    },
    {
      "auxiliary_loss_clip": 0.01193262,
      "auxiliary_loss_mlp": 0.01041072,
      "balance_loss_clip": 1.06005526,
      "balance_loss_mlp": 1.0254612,
      "epoch": 0.10990200204412914,
      "flos": 20412667077120.0,
      "grad_norm": 2.741414068931162,
      "language_loss": 0.86487222,
      "learning_rate": 3.933488380760562e-06,
      "loss": 0.8872155,
      "num_input_tokens_seen": 19397040,
      "step": 914,
      "time_per_iteration": 2.6241419315338135
    },
    {
      "auxiliary_loss_clip": 0.01154288,
      "auxiliary_loss_mlp": 0.01050384,
      "balance_loss_clip": 1.05395699,
      "balance_loss_mlp": 1.03672242,
      "epoch": 0.11002224493476823,
      "flos": 17530117660800.0,
      "grad_norm": 2.1353964706019903,
      "language_loss": 0.87052655,
      "learning_rate": 3.9332890157555286e-06,
      "loss": 0.8925733,
      "num_input_tokens_seen": 19413975,
      "step": 915,
      "time_per_iteration": 2.6464390754699707
    },
    {
      "auxiliary_loss_clip": 0.01228829,
      "auxiliary_loss_mlp": 0.00767387,
      "balance_loss_clip": 1.07247424,
      "balance_loss_mlp": 1.0007062,
      "epoch": 0.11014248782540732,
      "flos": 12203093099520.0,
      "grad_norm": 2.16372481284702,
      "language_loss": 0.76095378,
      "learning_rate": 3.933089357472088e-06,
      "loss": 0.78091592,
      "num_input_tokens_seen": 19432005,
      "step": 916,
      "time_per_iteration": 2.5488717555999756
    },
    {
      "auxiliary_loss_clip": 0.01141128,
      "auxiliary_loss_mlp": 0.01047775,
      "balance_loss_clip": 1.05826688,
      "balance_loss_mlp": 1.03327346,
      "epoch": 0.11026273071604642,
      "flos": 22382977760640.0,
      "grad_norm": 1.8012945732808043,
      "language_loss": 0.86138082,
      "learning_rate": 3.932889405940529e-06,
      "loss": 0.88326985,
      "num_input_tokens_seen": 19450100,
      "step": 917,
      "time_per_iteration": 2.7054522037506104
    },
    {
      "auxiliary_loss_clip": 0.01193412,
      "auxiliary_loss_mlp": 0.01039686,
      "balance_loss_clip": 1.06676233,
      "balance_loss_mlp": 1.02669239,
      "epoch": 0.1103829736066855,
      "flos": 19829046896640.0,
      "grad_norm": 2.3137562813143786,
      "language_loss": 0.79984611,
      "learning_rate": 3.932689161191184e-06,
      "loss": 0.82217705,
      "num_input_tokens_seen": 19467805,
      "step": 918,
      "time_per_iteration": 2.5855553150177
    },
    {
      "auxiliary_loss_clip": 0.01219208,
      "auxiliary_loss_mlp": 0.01041953,
      "balance_loss_clip": 1.07117462,
      "balance_loss_mlp": 1.02840447,
      "epoch": 0.1105032164973246,
      "flos": 22669616292480.0,
      "grad_norm": 2.1140388686926475,
      "language_loss": 0.8808136,
      "learning_rate": 3.93248862325443e-06,
      "loss": 0.9034251,
      "num_input_tokens_seen": 19486710,
      "step": 919,
      "time_per_iteration": 2.545273542404175
    },
    {
      "auxiliary_loss_clip": 0.01110676,
      "auxiliary_loss_mlp": 0.01007835,
      "balance_loss_clip": 1.05079234,
      "balance_loss_mlp": 1.0051291,
      "epoch": 0.11062345938796368,
      "flos": 66483507876480.0,
      "grad_norm": 0.933753167498213,
      "language_loss": 0.64444351,
      "learning_rate": 3.932287792160688e-06,
      "loss": 0.66562867,
      "num_input_tokens_seen": 19545170,
      "step": 920,
      "time_per_iteration": 3.0841758251190186
    },
    {
      "auxiliary_loss_clip": 0.01179119,
      "auxiliary_loss_mlp": 0.01049521,
      "balance_loss_clip": 1.06347489,
      "balance_loss_mlp": 1.03471518,
      "epoch": 0.11074370227860278,
      "flos": 21907771804800.0,
      "grad_norm": 2.3662876830298183,
      "language_loss": 0.81279767,
      "learning_rate": 3.932086667940424e-06,
      "loss": 0.83508408,
      "num_input_tokens_seen": 19561875,
      "step": 921,
      "time_per_iteration": 2.662923574447632
    },
    {
      "auxiliary_loss_clip": 0.01181906,
      "auxiliary_loss_mlp": 0.01040709,
      "balance_loss_clip": 1.05942357,
      "balance_loss_mlp": 1.02708364,
      "epoch": 0.11086394516924186,
      "flos": 28658115763200.0,
      "grad_norm": 2.4774136773021467,
      "language_loss": 0.81570452,
      "learning_rate": 3.93188525062415e-06,
      "loss": 0.83793068,
      "num_input_tokens_seen": 19582340,
      "step": 922,
      "time_per_iteration": 2.663127899169922
    },
    {
      "auxiliary_loss_clip": 0.01237166,
      "auxiliary_loss_mlp": 0.01046591,
      "balance_loss_clip": 1.0722065,
      "balance_loss_mlp": 1.03128994,
      "epoch": 0.11098418805988096,
      "flos": 24535247765760.0,
      "grad_norm": 2.1607818240408263,
      "language_loss": 0.8651517,
      "learning_rate": 3.931683540242418e-06,
      "loss": 0.88798928,
      "num_input_tokens_seen": 19603405,
      "step": 923,
      "time_per_iteration": 2.585080623626709
    },
    {
      "auxiliary_loss_clip": 0.01192684,
      "auxiliary_loss_mlp": 0.00766426,
      "balance_loss_clip": 1.06273615,
      "balance_loss_mlp": 1.00070989,
      "epoch": 0.11110443095052006,
      "flos": 22960384888320.0,
      "grad_norm": 2.7465512254982434,
      "language_loss": 0.91151285,
      "learning_rate": 3.9314815368258295e-06,
      "loss": 0.93110394,
      "num_input_tokens_seen": 19619885,
      "step": 924,
      "time_per_iteration": 3.3413596153259277
    },
    {
      "auxiliary_loss_clip": 0.01203269,
      "auxiliary_loss_mlp": 0.0076553,
      "balance_loss_clip": 1.06797576,
      "balance_loss_mlp": 1.00076044,
      "epoch": 0.11122467384115914,
      "flos": 18950025265920.0,
      "grad_norm": 1.7267102381252148,
      "language_loss": 0.7910943,
      "learning_rate": 3.9312792404050275e-06,
      "loss": 0.81078231,
      "num_input_tokens_seen": 19637940,
      "step": 925,
      "time_per_iteration": 3.4750888347625732
    },
    {
      "auxiliary_loss_clip": 0.01190014,
      "auxiliary_loss_mlp": 0.01039571,
      "balance_loss_clip": 1.05819499,
      "balance_loss_mlp": 1.02771521,
      "epoch": 0.11134491673179824,
      "flos": 25082957324160.0,
      "grad_norm": 1.9950393981545949,
      "language_loss": 0.77179933,
      "learning_rate": 3.9310766510107e-06,
      "loss": 0.79409516,
      "num_input_tokens_seen": 19657115,
      "step": 926,
      "time_per_iteration": 2.6011717319488525
    },
    {
      "auxiliary_loss_clip": 0.01189647,
      "auxiliary_loss_mlp": 0.01049586,
      "balance_loss_clip": 1.06557512,
      "balance_loss_mlp": 1.03423226,
      "epoch": 0.11146515962243732,
      "flos": 24499121662080.0,
      "grad_norm": 1.815780286070079,
      "language_loss": 0.92009586,
      "learning_rate": 3.9308737686735806e-06,
      "loss": 0.94248819,
      "num_input_tokens_seen": 19677075,
      "step": 927,
      "time_per_iteration": 2.635396957397461
    },
    {
      "auxiliary_loss_clip": 0.01215764,
      "auxiliary_loss_mlp": 0.01045632,
      "balance_loss_clip": 1.06795931,
      "balance_loss_mlp": 1.03193498,
      "epoch": 0.11158540251307641,
      "flos": 22343763087360.0,
      "grad_norm": 2.065189640730255,
      "language_loss": 0.82637626,
      "learning_rate": 3.9306705934244455e-06,
      "loss": 0.8489902,
      "num_input_tokens_seen": 19697155,
      "step": 928,
      "time_per_iteration": 3.321164608001709
    },
    {
      "auxiliary_loss_clip": 0.01216812,
      "auxiliary_loss_mlp": 0.01041288,
      "balance_loss_clip": 1.06591141,
      "balance_loss_mlp": 1.02870548,
      "epoch": 0.11170564540371551,
      "flos": 19902304684800.0,
      "grad_norm": 1.6165103422228437,
      "language_loss": 0.88420558,
      "learning_rate": 3.930467125294116e-06,
      "loss": 0.90678656,
      "num_input_tokens_seen": 19716705,
      "step": 929,
      "time_per_iteration": 3.232994318008423
    },
    {
      "auxiliary_loss_clip": 0.01138656,
      "auxiliary_loss_mlp": 0.01005608,
      "balance_loss_clip": 1.04017305,
      "balance_loss_mlp": 1.00246143,
      "epoch": 0.1118258882943546,
      "flos": 64586239499520.0,
      "grad_norm": 0.9281921486835243,
      "language_loss": 0.60467029,
      "learning_rate": 3.930263364313458e-06,
      "loss": 0.62611294,
      "num_input_tokens_seen": 19767275,
      "step": 930,
      "time_per_iteration": 2.9030425548553467
    },
    {
      "auxiliary_loss_clip": 0.01206655,
      "auxiliary_loss_mlp": 0.01051612,
      "balance_loss_clip": 1.06924117,
      "balance_loss_mlp": 1.03763509,
      "epoch": 0.11194613118499369,
      "flos": 17201965985280.0,
      "grad_norm": 2.111863472703249,
      "language_loss": 0.83066154,
      "learning_rate": 3.930059310513384e-06,
      "loss": 0.85324419,
      "num_input_tokens_seen": 19786315,
      "step": 931,
      "time_per_iteration": 2.542827844619751
    },
    {
      "auxiliary_loss_clip": 0.012335,
      "auxiliary_loss_mlp": 0.00766481,
      "balance_loss_clip": 1.07182813,
      "balance_loss_mlp": 1.00066543,
      "epoch": 0.11206637407563277,
      "flos": 31863465728640.0,
      "grad_norm": 1.7499761799537434,
      "language_loss": 0.83898258,
      "learning_rate": 3.929854963924846e-06,
      "loss": 0.85898244,
      "num_input_tokens_seen": 19806580,
      "step": 932,
      "time_per_iteration": 2.605806827545166
    },
    {
      "auxiliary_loss_clip": 0.01188053,
      "auxiliary_loss_mlp": 0.00766232,
      "balance_loss_clip": 1.05985725,
      "balance_loss_mlp": 1.00064075,
      "epoch": 0.11218661696627187,
      "flos": 21945621761280.0,
      "grad_norm": 1.913201345774471,
      "language_loss": 0.77212948,
      "learning_rate": 3.929650324578845e-06,
      "loss": 0.79167235,
      "num_input_tokens_seen": 19826045,
      "step": 933,
      "time_per_iteration": 2.568127155303955
    },
    {
      "auxiliary_loss_clip": 0.01235844,
      "auxiliary_loss_mlp": 0.01046842,
      "balance_loss_clip": 1.07070017,
      "balance_loss_mlp": 1.03158355,
      "epoch": 0.11230685985691095,
      "flos": 25878198481920.0,
      "grad_norm": 2.576914378044361,
      "language_loss": 0.8197611,
      "learning_rate": 3.929445392506423e-06,
      "loss": 0.84258795,
      "num_input_tokens_seen": 19843985,
      "step": 934,
      "time_per_iteration": 2.5736160278320312
    },
    {
      "auxiliary_loss_clip": 0.01205909,
      "auxiliary_loss_mlp": 0.01040222,
      "balance_loss_clip": 1.06638145,
      "balance_loss_mlp": 1.02730012,
      "epoch": 0.11242710274755005,
      "flos": 22231506107520.0,
      "grad_norm": 1.9318197269838955,
      "language_loss": 0.76138365,
      "learning_rate": 3.92924016773867e-06,
      "loss": 0.78384495,
      "num_input_tokens_seen": 19860480,
      "step": 935,
      "time_per_iteration": 2.576019287109375
    },
    {
      "auxiliary_loss_clip": 0.01173845,
      "auxiliary_loss_mlp": 0.01040975,
      "balance_loss_clip": 1.05867958,
      "balance_loss_mlp": 1.02841067,
      "epoch": 0.11254734563818915,
      "flos": 17712184723200.0,
      "grad_norm": 2.2180632434278915,
      "language_loss": 0.73228741,
      "learning_rate": 3.9290346503067175e-06,
      "loss": 0.7544356,
      "num_input_tokens_seen": 19877145,
      "step": 936,
      "time_per_iteration": 2.647620916366577
    },
    {
      "auxiliary_loss_clip": 0.01182094,
      "auxiliary_loss_mlp": 0.01035408,
      "balance_loss_clip": 1.06453991,
      "balance_loss_mlp": 1.02161551,
      "epoch": 0.11266758852882823,
      "flos": 54930397334400.0,
      "grad_norm": 1.7073256130212096,
      "language_loss": 0.78858912,
      "learning_rate": 3.9288288402417415e-06,
      "loss": 0.81076413,
      "num_input_tokens_seen": 19903405,
      "step": 937,
      "time_per_iteration": 2.9585273265838623
    },
    {
      "auxiliary_loss_clip": 0.012174,
      "auxiliary_loss_mlp": 0.01043338,
      "balance_loss_clip": 1.06901503,
      "balance_loss_mlp": 1.02907515,
      "epoch": 0.11278783141946733,
      "flos": 18878132194560.0,
      "grad_norm": 2.6686784881343866,
      "language_loss": 0.70566726,
      "learning_rate": 3.928622737574964e-06,
      "loss": 0.72827458,
      "num_input_tokens_seen": 19918740,
      "step": 938,
      "time_per_iteration": 2.545724868774414
    },
    {
      "auxiliary_loss_clip": 0.01178213,
      "auxiliary_loss_mlp": 0.01042875,
      "balance_loss_clip": 1.06295848,
      "balance_loss_mlp": 1.0291779,
      "epoch": 0.11290807431010641,
      "flos": 26469252777600.0,
      "grad_norm": 2.3655354315722135,
      "language_loss": 0.90999693,
      "learning_rate": 3.928416342337652e-06,
      "loss": 0.93220782,
      "num_input_tokens_seen": 19938475,
      "step": 939,
      "time_per_iteration": 2.6883881092071533
    },
    {
      "auxiliary_loss_clip": 0.01178999,
      "auxiliary_loss_mlp": 0.01046783,
      "balance_loss_clip": 1.06266999,
      "balance_loss_mlp": 1.03372383,
      "epoch": 0.1130283172007455,
      "flos": 22710590732160.0,
      "grad_norm": 1.762718899553483,
      "language_loss": 0.82446063,
      "learning_rate": 3.928209654561113e-06,
      "loss": 0.84671843,
      "num_input_tokens_seen": 19959310,
      "step": 940,
      "time_per_iteration": 2.604806423187256
    },
    {
      "auxiliary_loss_clip": 0.01229112,
      "auxiliary_loss_mlp": 0.01044784,
      "balance_loss_clip": 1.06806612,
      "balance_loss_mlp": 1.03092599,
      "epoch": 0.1131485600913846,
      "flos": 23219911630080.0,
      "grad_norm": 2.3548668874619185,
      "language_loss": 0.81587833,
      "learning_rate": 3.928002674276703e-06,
      "loss": 0.83861732,
      "num_input_tokens_seen": 19978700,
      "step": 941,
      "time_per_iteration": 2.5430824756622314
    },
    {
      "auxiliary_loss_clip": 0.01220404,
      "auxiliary_loss_mlp": 0.01039458,
      "balance_loss_clip": 1.06130219,
      "balance_loss_mlp": 1.02518225,
      "epoch": 0.11326880298202369,
      "flos": 14064271286400.0,
      "grad_norm": 2.345709025085225,
      "language_loss": 0.75578511,
      "learning_rate": 3.92779540151582e-06,
      "loss": 0.77838379,
      "num_input_tokens_seen": 19995785,
      "step": 942,
      "time_per_iteration": 2.4904446601867676
    },
    {
      "auxiliary_loss_clip": 0.01185447,
      "auxiliary_loss_mlp": 0.01039862,
      "balance_loss_clip": 1.06238174,
      "balance_loss_mlp": 1.0258671,
      "epoch": 0.11338904587266278,
      "flos": 16325386479360.0,
      "grad_norm": 1.8429005444193853,
      "language_loss": 0.8551417,
      "learning_rate": 3.927587836309907e-06,
      "loss": 0.8773948,
      "num_input_tokens_seen": 20013615,
      "step": 943,
      "time_per_iteration": 2.5953094959259033
    },
    {
      "auxiliary_loss_clip": 0.01209566,
      "auxiliary_loss_mlp": 0.00767589,
      "balance_loss_clip": 1.06378841,
      "balance_loss_mlp": 1.00080705,
      "epoch": 0.11350928876330187,
      "flos": 24426258923520.0,
      "grad_norm": 1.793190542169677,
      "language_loss": 0.78145695,
      "learning_rate": 3.927379978690452e-06,
      "loss": 0.80122852,
      "num_input_tokens_seen": 20032880,
      "step": 944,
      "time_per_iteration": 2.5603854656219482
    },
    {
      "auxiliary_loss_clip": 0.01163657,
      "auxiliary_loss_mlp": 0.01040463,
      "balance_loss_clip": 1.05593574,
      "balance_loss_mlp": 1.02718925,
      "epoch": 0.11362953165394096,
      "flos": 24497074586880.0,
      "grad_norm": 2.6215042585293684,
      "language_loss": 0.87422204,
      "learning_rate": 3.927171828688987e-06,
      "loss": 0.89626324,
      "num_input_tokens_seen": 20052405,
      "step": 945,
      "time_per_iteration": 2.645218849182129
    },
    {
      "auxiliary_loss_clip": 0.01247765,
      "auxiliary_loss_mlp": 0.00767125,
      "balance_loss_clip": 1.07112956,
      "balance_loss_mlp": 1.00078702,
      "epoch": 0.11374977454458005,
      "flos": 24060831909120.0,
      "grad_norm": 2.029549003788791,
      "language_loss": 0.82190233,
      "learning_rate": 3.926963386337088e-06,
      "loss": 0.84205115,
      "num_input_tokens_seen": 20070635,
      "step": 946,
      "time_per_iteration": 2.502505302429199
    },
    {
      "auxiliary_loss_clip": 0.01211098,
      "auxiliary_loss_mlp": 0.01037102,
      "balance_loss_clip": 1.06467593,
      "balance_loss_mlp": 1.02246881,
      "epoch": 0.11387001743521914,
      "flos": 39457638967680.0,
      "grad_norm": 2.484314636960739,
      "language_loss": 0.70077884,
      "learning_rate": 3.926754651666375e-06,
      "loss": 0.72326082,
      "num_input_tokens_seen": 20091195,
      "step": 947,
      "time_per_iteration": 2.698945999145508
    },
    {
      "auxiliary_loss_clip": 0.01171202,
      "auxiliary_loss_mlp": 0.01042884,
      "balance_loss_clip": 1.06041646,
      "balance_loss_mlp": 1.02898455,
      "epoch": 0.11399026032585824,
      "flos": 25082454533760.0,
      "grad_norm": 2.8868488902129346,
      "language_loss": 0.78596413,
      "learning_rate": 3.926545624708513e-06,
      "loss": 0.80810499,
      "num_input_tokens_seen": 20110435,
      "step": 948,
      "time_per_iteration": 2.623741865158081
    },
    {
      "auxiliary_loss_clip": 0.01173084,
      "auxiliary_loss_mlp": 0.01041014,
      "balance_loss_clip": 1.05837059,
      "balance_loss_mlp": 1.02766848,
      "epoch": 0.11411050321649732,
      "flos": 17961835224960.0,
      "grad_norm": 1.8024236228996258,
      "language_loss": 0.86162901,
      "learning_rate": 3.926336305495213e-06,
      "loss": 0.88376999,
      "num_input_tokens_seen": 20128995,
      "step": 949,
      "time_per_iteration": 2.5955612659454346
    },
    {
      "auxiliary_loss_clip": 0.01191973,
      "auxiliary_loss_mlp": 0.01037471,
      "balance_loss_clip": 1.06423473,
      "balance_loss_mlp": 1.02465606,
      "epoch": 0.11423074610713642,
      "flos": 22455409536000.0,
      "grad_norm": 2.0275062535149644,
      "language_loss": 0.89020491,
      "learning_rate": 3.926126694058226e-06,
      "loss": 0.91249931,
      "num_input_tokens_seen": 20148145,
      "step": 950,
      "time_per_iteration": 3.3800806999206543
    },
    {
      "auxiliary_loss_clip": 0.01244957,
      "auxiliary_loss_mlp": 0.0103651,
      "balance_loss_clip": 1.07044387,
      "balance_loss_mlp": 1.02420139,
      "epoch": 0.1143509889977755,
      "flos": 19717687756800.0,
      "grad_norm": 1.8146665712783248,
      "language_loss": 0.82135713,
      "learning_rate": 3.92591679042935e-06,
      "loss": 0.84417188,
      "num_input_tokens_seen": 20168035,
      "step": 951,
      "time_per_iteration": 2.511060953140259
    },
    {
      "auxiliary_loss_clip": 0.01199084,
      "auxiliary_loss_mlp": 0.01047342,
      "balance_loss_clip": 1.06334472,
      "balance_loss_mlp": 1.03224432,
      "epoch": 0.1144712318884146,
      "flos": 19822869757440.0,
      "grad_norm": 1.665773875549534,
      "language_loss": 0.82590175,
      "learning_rate": 3.92570659464043e-06,
      "loss": 0.84836602,
      "num_input_tokens_seen": 20186095,
      "step": 952,
      "time_per_iteration": 2.5645394325256348
    },
    {
      "auxiliary_loss_clip": 0.01178045,
      "auxiliary_loss_mlp": 0.0104729,
      "balance_loss_clip": 1.05802417,
      "balance_loss_mlp": 1.0348804,
      "epoch": 0.1145914747790537,
      "flos": 14939198766720.0,
      "grad_norm": 2.0019405959962766,
      "language_loss": 0.79524505,
      "learning_rate": 3.925496106723349e-06,
      "loss": 0.81749845,
      "num_input_tokens_seen": 20203535,
      "step": 953,
      "time_per_iteration": 2.572575330734253
    },
    {
      "auxiliary_loss_clip": 0.01190221,
      "auxiliary_loss_mlp": 0.0104611,
      "balance_loss_clip": 1.06007755,
      "balance_loss_mlp": 1.03414178,
      "epoch": 0.11471171766969278,
      "flos": 19865029345920.0,
      "grad_norm": 1.9967530296826448,
      "language_loss": 0.84163308,
      "learning_rate": 3.9252853267100405e-06,
      "loss": 0.86399639,
      "num_input_tokens_seen": 20222780,
      "step": 954,
      "time_per_iteration": 3.3291542530059814
    },
    {
      "auxiliary_loss_clip": 0.01237073,
      "auxiliary_loss_mlp": 0.01053443,
      "balance_loss_clip": 1.06929934,
      "balance_loss_mlp": 1.03947783,
      "epoch": 0.11483196056033187,
      "flos": 22526476594560.0,
      "grad_norm": 1.8325093403662367,
      "language_loss": 0.84003055,
      "learning_rate": 3.9250742546324786e-06,
      "loss": 0.86293566,
      "num_input_tokens_seen": 20243015,
      "step": 955,
      "time_per_iteration": 3.3009915351867676
    },
    {
      "auxiliary_loss_clip": 0.01232197,
      "auxiliary_loss_mlp": 0.01042974,
      "balance_loss_clip": 1.06506419,
      "balance_loss_mlp": 1.02981913,
      "epoch": 0.11495220345097096,
      "flos": 28220292887040.0,
      "grad_norm": 1.7723813997533315,
      "language_loss": 0.86931247,
      "learning_rate": 3.924862890522683e-06,
      "loss": 0.89206415,
      "num_input_tokens_seen": 20263025,
      "step": 956,
      "time_per_iteration": 2.5561745166778564
    },
    {
      "auxiliary_loss_clip": 0.0115907,
      "auxiliary_loss_mlp": 0.01044237,
      "balance_loss_clip": 1.05602622,
      "balance_loss_mlp": 1.02934754,
      "epoch": 0.11507244634161005,
      "flos": 17492267704320.0,
      "grad_norm": 2.1058256672440643,
      "language_loss": 0.86500013,
      "learning_rate": 3.9246512344127174e-06,
      "loss": 0.88703322,
      "num_input_tokens_seen": 20280685,
      "step": 957,
      "time_per_iteration": 2.6035900115966797
    },
    {
      "auxiliary_loss_clip": 0.01180896,
      "auxiliary_loss_mlp": 0.01039674,
      "balance_loss_clip": 1.05878758,
      "balance_loss_mlp": 1.02683496,
      "epoch": 0.11519268923224914,
      "flos": 22564937082240.0,
      "grad_norm": 1.7696548931604672,
      "language_loss": 0.82256544,
      "learning_rate": 3.9244392863346895e-06,
      "loss": 0.84477115,
      "num_input_tokens_seen": 20300090,
      "step": 958,
      "time_per_iteration": 2.611914873123169
    },
    {
      "auxiliary_loss_clip": 0.01201069,
      "auxiliary_loss_mlp": 0.01042728,
      "balance_loss_clip": 1.06513739,
      "balance_loss_mlp": 1.02861381,
      "epoch": 0.11531293212288823,
      "flos": 16982839065600.0,
      "grad_norm": 1.8608183138284695,
      "language_loss": 0.92013133,
      "learning_rate": 3.9242270463207524e-06,
      "loss": 0.94256938,
      "num_input_tokens_seen": 20318480,
      "step": 959,
      "time_per_iteration": 2.543086528778076
    },
    {
      "auxiliary_loss_clip": 0.01143668,
      "auxiliary_loss_mlp": 0.01040107,
      "balance_loss_clip": 1.05183482,
      "balance_loss_mlp": 1.0270834,
      "epoch": 0.11543317501352733,
      "flos": 12422004537600.0,
      "grad_norm": 2.7731018815416775,
      "language_loss": 0.84499401,
      "learning_rate": 3.924014514403102e-06,
      "loss": 0.86683178,
      "num_input_tokens_seen": 20334635,
      "step": 960,
      "time_per_iteration": 2.6375176906585693
    },
    {
      "auxiliary_loss_clip": 0.01185608,
      "auxiliary_loss_mlp": 0.01044021,
      "balance_loss_clip": 1.05948842,
      "balance_loss_mlp": 1.0291853,
      "epoch": 0.11555341790416641,
      "flos": 19821648695040.0,
      "grad_norm": 2.4579028934092846,
      "language_loss": 0.91053987,
      "learning_rate": 3.92380169061398e-06,
      "loss": 0.93283617,
      "num_input_tokens_seen": 20352415,
      "step": 961,
      "time_per_iteration": 2.601290464401245
    },
    {
      "auxiliary_loss_clip": 0.01176132,
      "auxiliary_loss_mlp": 0.01042129,
      "balance_loss_clip": 1.05879223,
      "balance_loss_mlp": 1.02899837,
      "epoch": 0.11567366079480551,
      "flos": 25738865625600.0,
      "grad_norm": 2.0698649576899437,
      "language_loss": 0.84046119,
      "learning_rate": 3.9235885749856705e-06,
      "loss": 0.86264384,
      "num_input_tokens_seen": 20371095,
      "step": 962,
      "time_per_iteration": 2.679368734359741
    },
    {
      "auxiliary_loss_clip": 0.01221963,
      "auxiliary_loss_mlp": 0.007662,
      "balance_loss_clip": 1.0671103,
      "balance_loss_mlp": 1.00087428,
      "epoch": 0.1157939036854446,
      "flos": 18223301301120.0,
      "grad_norm": 1.875288890566907,
      "language_loss": 0.82543957,
      "learning_rate": 3.9233751675505035e-06,
      "loss": 0.84532118,
      "num_input_tokens_seen": 20389805,
      "step": 963,
      "time_per_iteration": 2.551074266433716
    },
    {
      "auxiliary_loss_clip": 0.01205185,
      "auxiliary_loss_mlp": 0.0076598,
      "balance_loss_clip": 1.07042015,
      "balance_loss_mlp": 1.00092781,
      "epoch": 0.11591414657608369,
      "flos": 23073755189760.0,
      "grad_norm": 2.049434625909889,
      "language_loss": 0.85073507,
      "learning_rate": 3.923161468340853e-06,
      "loss": 0.87044668,
      "num_input_tokens_seen": 20409640,
      "step": 964,
      "time_per_iteration": 2.5852391719818115
    },
    {
      "auxiliary_loss_clip": 0.011961,
      "auxiliary_loss_mlp": 0.01043719,
      "balance_loss_clip": 1.05903387,
      "balance_loss_mlp": 1.02999234,
      "epoch": 0.11603438946672277,
      "flos": 19461716461440.0,
      "grad_norm": 1.679923851936627,
      "language_loss": 0.81618524,
      "learning_rate": 3.9229474773891374e-06,
      "loss": 0.83858341,
      "num_input_tokens_seen": 20428180,
      "step": 965,
      "time_per_iteration": 2.5803239345550537
    },
    {
      "auxiliary_loss_clip": 0.0117679,
      "auxiliary_loss_mlp": 0.01042856,
      "balance_loss_clip": 1.05865729,
      "balance_loss_mlp": 1.02734101,
      "epoch": 0.11615463235736187,
      "flos": 26831986272000.0,
      "grad_norm": 1.9496756768854056,
      "language_loss": 0.83660436,
      "learning_rate": 3.922733194727818e-06,
      "loss": 0.85880083,
      "num_input_tokens_seen": 20447975,
      "step": 966,
      "time_per_iteration": 2.6636054515838623
    },
    {
      "auxiliary_loss_clip": 0.01179234,
      "auxiliary_loss_mlp": 0.01043529,
      "balance_loss_clip": 1.06299078,
      "balance_loss_mlp": 1.02782297,
      "epoch": 0.11627487524800097,
      "flos": 18580324533120.0,
      "grad_norm": 2.9736752131959485,
      "language_loss": 0.87607688,
      "learning_rate": 3.922518620389402e-06,
      "loss": 0.89830446,
      "num_input_tokens_seen": 20464840,
      "step": 967,
      "time_per_iteration": 2.5793185234069824
    },
    {
      "auxiliary_loss_clip": 0.01238438,
      "auxiliary_loss_mlp": 0.01046158,
      "balance_loss_clip": 1.075225,
      "balance_loss_mlp": 1.03395069,
      "epoch": 0.11639511813864005,
      "flos": 18150474476160.0,
      "grad_norm": 1.7833075189632903,
      "language_loss": 0.89628649,
      "learning_rate": 3.922303754406439e-06,
      "loss": 0.91913247,
      "num_input_tokens_seen": 20482680,
      "step": 968,
      "time_per_iteration": 2.4695136547088623
    },
    {
      "auxiliary_loss_clip": 0.01186168,
      "auxiliary_loss_mlp": 0.00765792,
      "balance_loss_clip": 1.06397247,
      "balance_loss_mlp": 1.00082588,
      "epoch": 0.11651536102927915,
      "flos": 20922023888640.0,
      "grad_norm": 1.8064490093295422,
      "language_loss": 0.78954262,
      "learning_rate": 3.922088596811526e-06,
      "loss": 0.80906218,
      "num_input_tokens_seen": 20501810,
      "step": 969,
      "time_per_iteration": 2.6166789531707764
    },
    {
      "auxiliary_loss_clip": 0.01218254,
      "auxiliary_loss_mlp": 0.01041102,
      "balance_loss_clip": 1.06605244,
      "balance_loss_mlp": 1.02827489,
      "epoch": 0.11663560391991823,
      "flos": 16508602776960.0,
      "grad_norm": 2.1170427119905377,
      "language_loss": 0.86707604,
      "learning_rate": 3.9218731476373e-06,
      "loss": 0.88966966,
      "num_input_tokens_seen": 20517995,
      "step": 970,
      "time_per_iteration": 2.5106582641601562
    },
    {
      "auxiliary_loss_clip": 0.01200918,
      "auxiliary_loss_mlp": 0.01040541,
      "balance_loss_clip": 1.06525695,
      "balance_loss_mlp": 1.02463841,
      "epoch": 0.11675584681055733,
      "flos": 19865029345920.0,
      "grad_norm": 2.1693582119813373,
      "language_loss": 0.84817225,
      "learning_rate": 3.9216574069164455e-06,
      "loss": 0.87058687,
      "num_input_tokens_seen": 20536970,
      "step": 971,
      "time_per_iteration": 2.5749921798706055
    },
    {
      "auxiliary_loss_clip": 0.01214722,
      "auxiliary_loss_mlp": 0.01040553,
      "balance_loss_clip": 1.0645299,
      "balance_loss_mlp": 1.02807808,
      "epoch": 0.11687608970119642,
      "flos": 21944364785280.0,
      "grad_norm": 1.542621550661787,
      "language_loss": 0.8027575,
      "learning_rate": 3.921441374681691e-06,
      "loss": 0.82531023,
      "num_input_tokens_seen": 20557030,
      "step": 972,
      "time_per_iteration": 2.5419719219207764
    },
    {
      "auxiliary_loss_clip": 0.01156129,
      "auxiliary_loss_mlp": 0.01033088,
      "balance_loss_clip": 1.05417168,
      "balance_loss_mlp": 1.02019536,
      "epoch": 0.1169963325918355,
      "flos": 24061155131520.0,
      "grad_norm": 1.8138945728435152,
      "language_loss": 0.65369475,
      "learning_rate": 3.921225050965808e-06,
      "loss": 0.67558694,
      "num_input_tokens_seen": 20576915,
      "step": 973,
      "time_per_iteration": 2.6651973724365234
    },
    {
      "auxiliary_loss_clip": 0.01139382,
      "auxiliary_loss_mlp": 0.01041847,
      "balance_loss_clip": 1.05404282,
      "balance_loss_mlp": 1.02865648,
      "epoch": 0.1171165754824746,
      "flos": 23368151059200.0,
      "grad_norm": 3.4620084931929074,
      "language_loss": 0.74781364,
      "learning_rate": 3.921008435801612e-06,
      "loss": 0.76962596,
      "num_input_tokens_seen": 20596000,
      "step": 974,
      "time_per_iteration": 2.646962881088257
    },
    {
      "auxiliary_loss_clip": 0.01191456,
      "auxiliary_loss_mlp": 0.01040473,
      "balance_loss_clip": 1.06341696,
      "balance_loss_mlp": 1.02519608,
      "epoch": 0.11723681837311369,
      "flos": 18552243075840.0,
      "grad_norm": 2.2862956114680744,
      "language_loss": 0.76361823,
      "learning_rate": 3.920791529221963e-06,
      "loss": 0.78593755,
      "num_input_tokens_seen": 20614675,
      "step": 975,
      "time_per_iteration": 2.5627570152282715
    },
    {
      "auxiliary_loss_clip": 0.01159362,
      "auxiliary_loss_mlp": 0.01038133,
      "balance_loss_clip": 1.05946064,
      "balance_loss_mlp": 1.0251627,
      "epoch": 0.11735706126375278,
      "flos": 23550541344000.0,
      "grad_norm": 1.991008762949035,
      "language_loss": 0.76327062,
      "learning_rate": 3.920574331259768e-06,
      "loss": 0.78524554,
      "num_input_tokens_seen": 20635875,
      "step": 976,
      "time_per_iteration": 4.140315294265747
    },
    {
      "auxiliary_loss_clip": 0.01179394,
      "auxiliary_loss_mlp": 0.01048908,
      "balance_loss_clip": 1.06521916,
      "balance_loss_mlp": 1.03504992,
      "epoch": 0.11747730415439187,
      "flos": 22381541216640.0,
      "grad_norm": 2.4590418492408386,
      "language_loss": 0.79753846,
      "learning_rate": 3.9203568419479716e-06,
      "loss": 0.81982154,
      "num_input_tokens_seen": 20656430,
      "step": 977,
      "time_per_iteration": 2.6045749187469482
    },
    {
      "auxiliary_loss_clip": 0.0121152,
      "auxiliary_loss_mlp": 0.01045653,
      "balance_loss_clip": 1.06242776,
      "balance_loss_mlp": 1.03209305,
      "epoch": 0.11759754704503096,
      "flos": 22200731130240.0,
      "grad_norm": 1.980793668214709,
      "language_loss": 0.75440991,
      "learning_rate": 3.92013906131957e-06,
      "loss": 0.77698159,
      "num_input_tokens_seen": 20675360,
      "step": 978,
      "time_per_iteration": 2.564990997314453
    },
    {
      "auxiliary_loss_clip": 0.01166923,
      "auxiliary_loss_mlp": 0.01035234,
      "balance_loss_clip": 1.05913854,
      "balance_loss_mlp": 1.02337217,
      "epoch": 0.11771778993567006,
      "flos": 22309755886080.0,
      "grad_norm": 3.329014713209344,
      "language_loss": 0.82658076,
      "learning_rate": 3.9199209894076e-06,
      "loss": 0.84860229,
      "num_input_tokens_seen": 20695675,
      "step": 979,
      "time_per_iteration": 3.3532755374908447
    },
    {
      "auxiliary_loss_clip": 0.01220613,
      "auxiliary_loss_mlp": 0.01037925,
      "balance_loss_clip": 1.06303167,
      "balance_loss_mlp": 1.02299416,
      "epoch": 0.11783803282630914,
      "flos": 21288169175040.0,
      "grad_norm": 2.1128891042727553,
      "language_loss": 0.8997817,
      "learning_rate": 3.919702626245142e-06,
      "loss": 0.92236716,
      "num_input_tokens_seen": 20715330,
      "step": 980,
      "time_per_iteration": 2.545938014984131
    },
    {
      "auxiliary_loss_clip": 0.01111594,
      "auxiliary_loss_mlp": 0.01042437,
      "balance_loss_clip": 1.04634988,
      "balance_loss_mlp": 1.02915668,
      "epoch": 0.11795827571694824,
      "flos": 25371535190400.0,
      "grad_norm": 1.9103576704790204,
      "language_loss": 0.65950811,
      "learning_rate": 3.919483971865322e-06,
      "loss": 0.68104845,
      "num_input_tokens_seen": 20735325,
      "step": 981,
      "time_per_iteration": 3.7562129497528076
    },
    {
      "auxiliary_loss_clip": 0.01211039,
      "auxiliary_loss_mlp": 0.01042234,
      "balance_loss_clip": 1.0664711,
      "balance_loss_mlp": 1.02831078,
      "epoch": 0.11807851860758732,
      "flos": 23622218933760.0,
      "grad_norm": 2.0405352897698807,
      "language_loss": 0.88031673,
      "learning_rate": 3.91926502630131e-06,
      "loss": 0.9028495,
      "num_input_tokens_seen": 20755940,
      "step": 982,
      "time_per_iteration": 2.8308756351470947
    },
    {
      "auxiliary_loss_clip": 0.01217431,
      "auxiliary_loss_mlp": 0.01042002,
      "balance_loss_clip": 1.06690848,
      "balance_loss_mlp": 1.02907968,
      "epoch": 0.11819876149822642,
      "flos": 24972496024320.0,
      "grad_norm": 2.7212615910912263,
      "language_loss": 0.72214782,
      "learning_rate": 3.91904578958632e-06,
      "loss": 0.74474204,
      "num_input_tokens_seen": 20775355,
      "step": 983,
      "time_per_iteration": 2.5972788333892822
    },
    {
      "auxiliary_loss_clip": 0.01231957,
      "auxiliary_loss_mlp": 0.00765864,
      "balance_loss_clip": 1.06694484,
      "balance_loss_mlp": 1.00049448,
      "epoch": 0.11831900438886551,
      "flos": 23003226835200.0,
      "grad_norm": 2.6907255235014036,
      "language_loss": 0.84444171,
      "learning_rate": 3.918826261753608e-06,
      "loss": 0.86441994,
      "num_input_tokens_seen": 20794935,
      "step": 984,
      "time_per_iteration": 2.51131534576416
    },
    {
      "auxiliary_loss_clip": 0.01142498,
      "auxiliary_loss_mlp": 0.01042914,
      "balance_loss_clip": 1.05132651,
      "balance_loss_mlp": 1.03169608,
      "epoch": 0.1184392472795046,
      "flos": 27965147604480.0,
      "grad_norm": 3.149263550503857,
      "language_loss": 0.7112186,
      "learning_rate": 3.918606442836478e-06,
      "loss": 0.73307276,
      "num_input_tokens_seen": 20817155,
      "step": 985,
      "time_per_iteration": 2.7172093391418457
    },
    {
      "auxiliary_loss_clip": 0.0110536,
      "auxiliary_loss_mlp": 0.01042428,
      "balance_loss_clip": 1.05055475,
      "balance_loss_mlp": 1.03034663,
      "epoch": 0.1185594901701437,
      "flos": 19898497843200.0,
      "grad_norm": 3.391291525634438,
      "language_loss": 0.77548897,
      "learning_rate": 3.918386332868277e-06,
      "loss": 0.79696691,
      "num_input_tokens_seen": 20835125,
      "step": 986,
      "time_per_iteration": 2.713308811187744
    },
    {
      "auxiliary_loss_clip": 0.01201242,
      "auxiliary_loss_mlp": 0.01040328,
      "balance_loss_clip": 1.06549382,
      "balance_loss_mlp": 1.02649999,
      "epoch": 0.11867973306078278,
      "flos": 18912354877440.0,
      "grad_norm": 2.76619216570231,
      "language_loss": 0.9446829,
      "learning_rate": 3.918165931882394e-06,
      "loss": 0.96709859,
      "num_input_tokens_seen": 20853525,
      "step": 987,
      "time_per_iteration": 2.566558599472046
    },
    {
      "auxiliary_loss_clip": 0.01162148,
      "auxiliary_loss_mlp": 0.01038465,
      "balance_loss_clip": 1.06031847,
      "balance_loss_mlp": 1.0263114,
      "epoch": 0.11879997595142187,
      "flos": 16982803152000.0,
      "grad_norm": 3.9308381979106377,
      "language_loss": 0.7575599,
      "learning_rate": 3.917945239912264e-06,
      "loss": 0.77956605,
      "num_input_tokens_seen": 20871000,
      "step": 988,
      "time_per_iteration": 2.595061779022217
    },
    {
      "auxiliary_loss_clip": 0.01206694,
      "auxiliary_loss_mlp": 0.01046157,
      "balance_loss_clip": 1.06313324,
      "balance_loss_mlp": 1.03364027,
      "epoch": 0.11892021884206096,
      "flos": 17530369056000.0,
      "grad_norm": 2.1801815863547773,
      "language_loss": 0.75732249,
      "learning_rate": 3.917724256991367e-06,
      "loss": 0.77985102,
      "num_input_tokens_seen": 20889745,
      "step": 989,
      "time_per_iteration": 2.503769874572754
    },
    {
      "auxiliary_loss_clip": 0.01153373,
      "auxiliary_loss_mlp": 0.01042102,
      "balance_loss_clip": 1.05328524,
      "balance_loss_mlp": 1.02862585,
      "epoch": 0.11904046173270005,
      "flos": 30955895763840.0,
      "grad_norm": 2.3040023357161266,
      "language_loss": 0.81422848,
      "learning_rate": 3.9175029831532245e-06,
      "loss": 0.83618325,
      "num_input_tokens_seen": 20909260,
      "step": 990,
      "time_per_iteration": 2.6841368675231934
    },
    {
      "auxiliary_loss_clip": 0.01196777,
      "auxiliary_loss_mlp": 0.0103597,
      "balance_loss_clip": 1.06210041,
      "balance_loss_mlp": 1.02340555,
      "epoch": 0.11916070462333915,
      "flos": 20157234485760.0,
      "grad_norm": 2.1321293534470773,
      "language_loss": 0.88868678,
      "learning_rate": 3.917281418431404e-06,
      "loss": 0.91101432,
      "num_input_tokens_seen": 20928305,
      "step": 991,
      "time_per_iteration": 2.536902666091919
    },
    {
      "auxiliary_loss_clip": 0.011907,
      "auxiliary_loss_mlp": 0.01041177,
      "balance_loss_clip": 1.06166804,
      "balance_loss_mlp": 1.02747369,
      "epoch": 0.11928094751397823,
      "flos": 23551115961600.0,
      "grad_norm": 1.9352303031214648,
      "language_loss": 0.77112675,
      "learning_rate": 3.917059562859516e-06,
      "loss": 0.79344559,
      "num_input_tokens_seen": 20947630,
      "step": 992,
      "time_per_iteration": 2.5804169178009033
    },
    {
      "auxiliary_loss_clip": 0.01214539,
      "auxiliary_loss_mlp": 0.00764719,
      "balance_loss_clip": 1.06436729,
      "balance_loss_mlp": 1.0004884,
      "epoch": 0.11940119040461733,
      "flos": 23908426502400.0,
      "grad_norm": 2.407140685440624,
      "language_loss": 0.88776076,
      "learning_rate": 3.916837416471218e-06,
      "loss": 0.90755332,
      "num_input_tokens_seen": 20964250,
      "step": 993,
      "time_per_iteration": 2.590132474899292
    },
    {
      "auxiliary_loss_clip": 0.011747,
      "auxiliary_loss_mlp": 0.00765141,
      "balance_loss_clip": 1.05810118,
      "balance_loss_mlp": 1.00046182,
      "epoch": 0.11952143329525641,
      "flos": 13844533835520.0,
      "grad_norm": 3.2703796640943743,
      "language_loss": 0.72445655,
      "learning_rate": 3.916614979300207e-06,
      "loss": 0.743855,
      "num_input_tokens_seen": 20979095,
      "step": 994,
      "time_per_iteration": 2.583158254623413
    },
    {
      "auxiliary_loss_clip": 0.01184237,
      "auxiliary_loss_mlp": 0.01046018,
      "balance_loss_clip": 1.0622685,
      "balance_loss_mlp": 1.03488433,
      "epoch": 0.11964167618589551,
      "flos": 27015525792000.0,
      "grad_norm": 1.583416276796931,
      "language_loss": 0.78805143,
      "learning_rate": 3.9163922513802274e-06,
      "loss": 0.81035399,
      "num_input_tokens_seen": 21001430,
      "step": 995,
      "time_per_iteration": 2.652757406234741
    },
    {
      "auxiliary_loss_clip": 0.01162532,
      "auxiliary_loss_mlp": 0.01046365,
      "balance_loss_clip": 1.05143273,
      "balance_loss_mlp": 1.03309059,
      "epoch": 0.1197619190765346,
      "flos": 12567622273920.0,
      "grad_norm": 3.083845017732706,
      "language_loss": 0.82381743,
      "learning_rate": 3.916169232745067e-06,
      "loss": 0.84590638,
      "num_input_tokens_seen": 21019105,
      "step": 996,
      "time_per_iteration": 2.591881513595581
    },
    {
      "auxiliary_loss_clip": 0.01154186,
      "auxiliary_loss_mlp": 0.01039624,
      "balance_loss_clip": 1.0577755,
      "balance_loss_mlp": 1.02477062,
      "epoch": 0.11988216196717369,
      "flos": 16909437623040.0,
      "grad_norm": 2.640461433819908,
      "language_loss": 0.9212293,
      "learning_rate": 3.915945923428559e-06,
      "loss": 0.94316733,
      "num_input_tokens_seen": 21035630,
      "step": 997,
      "time_per_iteration": 2.655946731567383
    },
    {
      "auxiliary_loss_clip": 0.01165312,
      "auxiliary_loss_mlp": 0.0103741,
      "balance_loss_clip": 1.05767155,
      "balance_loss_mlp": 1.02472067,
      "epoch": 0.12000240485781279,
      "flos": 16216577205120.0,
      "grad_norm": 6.067814618181352,
      "language_loss": 0.8313511,
      "learning_rate": 3.915722323464577e-06,
      "loss": 0.8533783,
      "num_input_tokens_seen": 21054235,
      "step": 998,
      "time_per_iteration": 2.5845346450805664
    },
    {
      "auxiliary_loss_clip": 0.01196947,
      "auxiliary_loss_mlp": 0.01037717,
      "balance_loss_clip": 1.06393814,
      "balance_loss_mlp": 1.02353144,
      "epoch": 0.12012264774845187,
      "flos": 49344887525760.0,
      "grad_norm": 2.733118567902373,
      "language_loss": 0.70638406,
      "learning_rate": 3.91549843288704e-06,
      "loss": 0.72873074,
      "num_input_tokens_seen": 21077915,
      "step": 999,
      "time_per_iteration": 2.8278205394744873
    },
    {
      "auxiliary_loss_clip": 0.01152977,
      "auxiliary_loss_mlp": 0.01040898,
      "balance_loss_clip": 1.05712914,
      "balance_loss_mlp": 1.02810085,
      "epoch": 0.12024289063909097,
      "flos": 26979435601920.0,
      "grad_norm": 2.0530316506432675,
      "language_loss": 0.79202044,
      "learning_rate": 3.915274251729916e-06,
      "loss": 0.81395918,
      "num_input_tokens_seen": 21099205,
      "step": 1000,
      "time_per_iteration": 2.762308359146118
    },
    {
      "auxiliary_loss_clip": 0.0118332,
      "auxiliary_loss_mlp": 0.01037551,
      "balance_loss_clip": 1.05764747,
      "balance_loss_mlp": 1.02456295,
      "epoch": 0.12036313352973005,
      "flos": 19537308633600.0,
      "grad_norm": 2.0665558885989426,
      "language_loss": 0.90138161,
      "learning_rate": 3.91504978002721e-06,
      "loss": 0.92359024,
      "num_input_tokens_seen": 21118260,
      "step": 1001,
      "time_per_iteration": 2.5852644443511963
    },
    {
      "auxiliary_loss_clip": 0.01165077,
      "auxiliary_loss_mlp": 0.01039141,
      "balance_loss_clip": 1.05751097,
      "balance_loss_mlp": 1.02723253,
      "epoch": 0.12048337642036915,
      "flos": 17268256535040.0,
      "grad_norm": 1.8768777511466503,
      "language_loss": 0.76074916,
      "learning_rate": 3.914825017812974e-06,
      "loss": 0.78279126,
      "num_input_tokens_seen": 21134910,
      "step": 1002,
      "time_per_iteration": 4.609687328338623
    },
    {
      "auxiliary_loss_clip": 0.01117541,
      "auxiliary_loss_mlp": 0.01042594,
      "balance_loss_clip": 1.05089784,
      "balance_loss_mlp": 1.02887273,
      "epoch": 0.12060361931100824,
      "flos": 22856962654080.0,
      "grad_norm": 2.6637570659027547,
      "language_loss": 0.72739697,
      "learning_rate": 3.9145999651213065e-06,
      "loss": 0.7489984,
      "num_input_tokens_seen": 21154150,
      "step": 1003,
      "time_per_iteration": 2.941099166870117
    },
    {
      "auxiliary_loss_clip": 0.01190546,
      "auxiliary_loss_mlp": 0.01039176,
      "balance_loss_clip": 1.06223869,
      "balance_loss_mlp": 1.02590847,
      "epoch": 0.12072386220164733,
      "flos": 16726795943040.0,
      "grad_norm": 2.372477624204327,
      "language_loss": 0.88503659,
      "learning_rate": 3.9143746219863465e-06,
      "loss": 0.90733385,
      "num_input_tokens_seen": 21171255,
      "step": 1004,
      "time_per_iteration": 2.786655902862549
    },
    {
      "auxiliary_loss_clip": 0.0114767,
      "auxiliary_loss_mlp": 0.01007315,
      "balance_loss_clip": 1.05660129,
      "balance_loss_mlp": 1.00448954,
      "epoch": 0.12084410509228642,
      "flos": 55144176105600.0,
      "grad_norm": 0.9572563729384361,
      "language_loss": 0.64783895,
      "learning_rate": 3.914148988442278e-06,
      "loss": 0.66938877,
      "num_input_tokens_seen": 21227045,
      "step": 1005,
      "time_per_iteration": 3.7805685997009277
    },
    {
      "auxiliary_loss_clip": 0.01214892,
      "auxiliary_loss_mlp": 0.01037548,
      "balance_loss_clip": 1.06590736,
      "balance_loss_mlp": 1.02367759,
      "epoch": 0.1209643479829255,
      "flos": 26760236855040.0,
      "grad_norm": 2.6878085860304317,
      "language_loss": 0.95347297,
      "learning_rate": 3.91392306452333e-06,
      "loss": 0.97599733,
      "num_input_tokens_seen": 21244120,
      "step": 1006,
      "time_per_iteration": 3.460954427719116
    },
    {
      "auxiliary_loss_clip": 0.01228215,
      "auxiliary_loss_mlp": 0.01037316,
      "balance_loss_clip": 1.06742907,
      "balance_loss_mlp": 1.02363634,
      "epoch": 0.1210845908735646,
      "flos": 11035026725760.0,
      "grad_norm": 3.8088694232394076,
      "language_loss": 0.66256034,
      "learning_rate": 3.913696850263774e-06,
      "loss": 0.68521571,
      "num_input_tokens_seen": 21258485,
      "step": 1007,
      "time_per_iteration": 2.501830816268921
    },
    {
      "auxiliary_loss_clip": 0.01227877,
      "auxiliary_loss_mlp": 0.01047233,
      "balance_loss_clip": 1.06832981,
      "balance_loss_mlp": 1.0340966,
      "epoch": 0.1212048337642037,
      "flos": 20484631975680.0,
      "grad_norm": 2.2490922408170384,
      "language_loss": 0.78930467,
      "learning_rate": 3.913470345697929e-06,
      "loss": 0.81205577,
      "num_input_tokens_seen": 21277115,
      "step": 1008,
      "time_per_iteration": 2.5565948486328125
    },
    {
      "auxiliary_loss_clip": 0.01241211,
      "auxiliary_loss_mlp": 0.0103834,
      "balance_loss_clip": 1.06767011,
      "balance_loss_mlp": 1.02606106,
      "epoch": 0.12132507665484278,
      "flos": 22346061557760.0,
      "grad_norm": 2.045645581495251,
      "language_loss": 0.85797834,
      "learning_rate": 3.913243550860153e-06,
      "loss": 0.88077384,
      "num_input_tokens_seen": 21294880,
      "step": 1009,
      "time_per_iteration": 2.493187189102173
    },
    {
      "auxiliary_loss_clip": 0.01140247,
      "auxiliary_loss_mlp": 0.00765309,
      "balance_loss_clip": 1.05329466,
      "balance_loss_mlp": 1.00055742,
      "epoch": 0.12144531954548188,
      "flos": 29314957818240.0,
      "grad_norm": 1.8347552775411453,
      "language_loss": 0.76213127,
      "learning_rate": 3.913016465784852e-06,
      "loss": 0.78118682,
      "num_input_tokens_seen": 21315555,
      "step": 1010,
      "time_per_iteration": 2.7933332920074463
    },
    {
      "auxiliary_loss_clip": 0.01184896,
      "auxiliary_loss_mlp": 0.01039575,
      "balance_loss_clip": 1.06487679,
      "balance_loss_mlp": 1.02659869,
      "epoch": 0.12156556243612096,
      "flos": 20485242506880.0,
      "grad_norm": 5.652755895105785,
      "language_loss": 0.71929514,
      "learning_rate": 3.912789090506474e-06,
      "loss": 0.74153984,
      "num_input_tokens_seen": 21334815,
      "step": 1011,
      "time_per_iteration": 2.6193461418151855
    },
    {
      "auxiliary_loss_clip": 0.01113608,
      "auxiliary_loss_mlp": 0.00766021,
      "balance_loss_clip": 1.04719496,
      "balance_loss_mlp": 1.00044179,
      "epoch": 0.12168580532676006,
      "flos": 16472009796480.0,
      "grad_norm": 2.6206457570685435,
      "language_loss": 0.72180057,
      "learning_rate": 3.9125614250595114e-06,
      "loss": 0.74059683,
      "num_input_tokens_seen": 21351025,
      "step": 1012,
      "time_per_iteration": 2.9471726417541504
    },
    {
      "auxiliary_loss_clip": 0.01157755,
      "auxiliary_loss_mlp": 0.01037205,
      "balance_loss_clip": 1.05813444,
      "balance_loss_mlp": 1.02394915,
      "epoch": 0.12180604821739914,
      "flos": 15341290588800.0,
      "grad_norm": 2.5588384344068786,
      "language_loss": 0.88869154,
      "learning_rate": 3.912333469478502e-06,
      "loss": 0.91064119,
      "num_input_tokens_seen": 21368990,
      "step": 1013,
      "time_per_iteration": 3.507758617401123
    },
    {
      "auxiliary_loss_clip": 0.01209174,
      "auxiliary_loss_mlp": 0.01038208,
      "balance_loss_clip": 1.0662601,
      "balance_loss_mlp": 1.0254705,
      "epoch": 0.12192629110803824,
      "flos": 19318038059520.0,
      "grad_norm": 1.962443795549677,
      "language_loss": 0.7812748,
      "learning_rate": 3.912105223798025e-06,
      "loss": 0.80374861,
      "num_input_tokens_seen": 21388410,
      "step": 1014,
      "time_per_iteration": 2.574082136154175
    },
    {
      "auxiliary_loss_clip": 0.01136349,
      "auxiliary_loss_mlp": 0.01003416,
      "balance_loss_clip": 1.05115175,
      "balance_loss_mlp": 1.00014973,
      "epoch": 0.12204653399867733,
      "flos": 47725354085760.0,
      "grad_norm": 0.9941614328015591,
      "language_loss": 0.67676455,
      "learning_rate": 3.9118766880527065e-06,
      "loss": 0.6981622,
      "num_input_tokens_seen": 21442845,
      "step": 1015,
      "time_per_iteration": 3.0381064414978027
    },
    {
      "auxiliary_loss_clip": 0.01211936,
      "auxiliary_loss_mlp": 0.01037008,
      "balance_loss_clip": 1.06442964,
      "balance_loss_mlp": 1.0240562,
      "epoch": 0.12216677688931642,
      "flos": 18221936584320.0,
      "grad_norm": 1.8253630863535035,
      "language_loss": 0.73654604,
      "learning_rate": 3.9116478622772145e-06,
      "loss": 0.75903547,
      "num_input_tokens_seen": 21461420,
      "step": 1016,
      "time_per_iteration": 2.5442891120910645
    },
    {
      "auxiliary_loss_clip": 0.01190559,
      "auxiliary_loss_mlp": 0.0103874,
      "balance_loss_clip": 1.06236768,
      "balance_loss_mlp": 1.02600825,
      "epoch": 0.12228701977995551,
      "flos": 27525636789120.0,
      "grad_norm": 2.999269831985931,
      "language_loss": 0.88142848,
      "learning_rate": 3.911418746506261e-06,
      "loss": 0.90372145,
      "num_input_tokens_seen": 21481550,
      "step": 1017,
      "time_per_iteration": 2.6411659717559814
    },
    {
      "auxiliary_loss_clip": 0.01142748,
      "auxiliary_loss_mlp": 0.01043701,
      "balance_loss_clip": 1.05981779,
      "balance_loss_mlp": 1.03050435,
      "epoch": 0.1224072626705946,
      "flos": 21798136517760.0,
      "grad_norm": 2.416894158581703,
      "language_loss": 0.78371406,
      "learning_rate": 3.911189340774604e-06,
      "loss": 0.80557859,
      "num_input_tokens_seen": 21501680,
      "step": 1018,
      "time_per_iteration": 2.698493003845215
    },
    {
      "auxiliary_loss_clip": 0.01142343,
      "auxiliary_loss_mlp": 0.01039736,
      "balance_loss_clip": 1.05775106,
      "balance_loss_mlp": 1.02679539,
      "epoch": 0.1225275055612337,
      "flos": 20703758895360.0,
      "grad_norm": 1.9396957894728926,
      "language_loss": 0.79471308,
      "learning_rate": 3.910959645117043e-06,
      "loss": 0.81653386,
      "num_input_tokens_seen": 21521015,
      "step": 1019,
      "time_per_iteration": 2.674903154373169
    },
    {
      "auxiliary_loss_clip": 0.01100551,
      "auxiliary_loss_mlp": 0.01002431,
      "balance_loss_clip": 1.03511369,
      "balance_loss_mlp": 0.99945039,
      "epoch": 0.12264774845187278,
      "flos": 57745294462080.0,
      "grad_norm": 0.8155230554896737,
      "language_loss": 0.56734842,
      "learning_rate": 3.910729659568423e-06,
      "loss": 0.58837825,
      "num_input_tokens_seen": 21578200,
      "step": 1020,
      "time_per_iteration": 3.1456363201141357
    },
    {
      "auxiliary_loss_clip": 0.01239408,
      "auxiliary_loss_mlp": 0.01041004,
      "balance_loss_clip": 1.0690167,
      "balance_loss_mlp": 1.02966738,
      "epoch": 0.12276799134251187,
      "flos": 26396282298240.0,
      "grad_norm": 2.3740079140510875,
      "language_loss": 0.82528198,
      "learning_rate": 3.9104993841636344e-06,
      "loss": 0.84808612,
      "num_input_tokens_seen": 21598770,
      "step": 1021,
      "time_per_iteration": 2.540498971939087
    },
    {
      "auxiliary_loss_clip": 0.01193188,
      "auxiliary_loss_mlp": 0.01034188,
      "balance_loss_clip": 1.06433392,
      "balance_loss_mlp": 1.02238023,
      "epoch": 0.12288823423315097,
      "flos": 21064193919360.0,
      "grad_norm": 1.7148544899659721,
      "language_loss": 0.80912006,
      "learning_rate": 3.910268818937608e-06,
      "loss": 0.83139384,
      "num_input_tokens_seen": 21616925,
      "step": 1022,
      "time_per_iteration": 2.596661329269409
    },
    {
      "auxiliary_loss_clip": 0.01191093,
      "auxiliary_loss_mlp": 0.0104271,
      "balance_loss_clip": 1.06232285,
      "balance_loss_mlp": 1.02986503,
      "epoch": 0.12300847712379005,
      "flos": 12312441077760.0,
      "grad_norm": 3.476297045571986,
      "language_loss": 0.86976802,
      "learning_rate": 3.9100379639253196e-06,
      "loss": 0.89210606,
      "num_input_tokens_seen": 21633645,
      "step": 1023,
      "time_per_iteration": 2.5571413040161133
    },
    {
      "auxiliary_loss_clip": 0.01210951,
      "auxiliary_loss_mlp": 0.0104542,
      "balance_loss_clip": 1.06418157,
      "balance_loss_mlp": 1.03242052,
      "epoch": 0.12312872001442915,
      "flos": 16762239688320.0,
      "grad_norm": 2.557630521220721,
      "language_loss": 0.86391002,
      "learning_rate": 3.909806819161791e-06,
      "loss": 0.88647377,
      "num_input_tokens_seen": 21649120,
      "step": 1024,
      "time_per_iteration": 2.5208370685577393
    },
    {
      "auxiliary_loss_clip": 0.0119488,
      "auxiliary_loss_mlp": 0.01040477,
      "balance_loss_clip": 1.06199825,
      "balance_loss_mlp": 1.02632689,
      "epoch": 0.12324896290506823,
      "flos": 18404937400320.0,
      "grad_norm": 1.9409691926105588,
      "language_loss": 0.86721194,
      "learning_rate": 3.909575384682086e-06,
      "loss": 0.88956553,
      "num_input_tokens_seen": 21668000,
      "step": 1025,
      "time_per_iteration": 2.554084300994873
    },
    {
      "auxiliary_loss_clip": 0.01157536,
      "auxiliary_loss_mlp": 0.01035637,
      "balance_loss_clip": 1.0590353,
      "balance_loss_mlp": 1.0236032,
      "epoch": 0.12336920579570733,
      "flos": 18915407533440.0,
      "grad_norm": 2.0047645205518694,
      "language_loss": 0.69647014,
      "learning_rate": 3.9093436605213144e-06,
      "loss": 0.71840179,
      "num_input_tokens_seen": 21688500,
      "step": 1026,
      "time_per_iteration": 2.6350014209747314
    },
    {
      "auxiliary_loss_clip": 0.0115901,
      "auxiliary_loss_mlp": 0.00765699,
      "balance_loss_clip": 1.05333447,
      "balance_loss_mlp": 1.00035214,
      "epoch": 0.12348944868634643,
      "flos": 23878369797120.0,
      "grad_norm": 1.8833231579321557,
      "language_loss": 0.79707944,
      "learning_rate": 3.909111646714627e-06,
      "loss": 0.81632656,
      "num_input_tokens_seen": 21709345,
      "step": 1027,
      "time_per_iteration": 3.699223756790161
    },
    {
      "auxiliary_loss_clip": 0.01178541,
      "auxiliary_loss_mlp": 0.01035862,
      "balance_loss_clip": 1.05438495,
      "balance_loss_mlp": 1.0231607,
      "epoch": 0.12360969157698551,
      "flos": 19026084314880.0,
      "grad_norm": 1.9649782470134745,
      "language_loss": 0.72488672,
      "learning_rate": 3.9088793432972206e-06,
      "loss": 0.74703074,
      "num_input_tokens_seen": 21728165,
      "step": 1028,
      "time_per_iteration": 2.5571069717407227
    },
    {
      "auxiliary_loss_clip": 0.0121038,
      "auxiliary_loss_mlp": 0.01042142,
      "balance_loss_clip": 1.06320572,
      "balance_loss_mlp": 1.03133011,
      "epoch": 0.1237299344676246,
      "flos": 13224607983360.0,
      "grad_norm": 2.0480157976746396,
      "language_loss": 0.82124019,
      "learning_rate": 3.908646750304336e-06,
      "loss": 0.8437655,
      "num_input_tokens_seen": 21745850,
      "step": 1029,
      "time_per_iteration": 2.5417799949645996
    },
    {
      "auxiliary_loss_clip": 0.01165947,
      "auxiliary_loss_mlp": 0.01046028,
      "balance_loss_clip": 1.06006551,
      "balance_loss_mlp": 1.03196132,
      "epoch": 0.12385017735826369,
      "flos": 20485673470080.0,
      "grad_norm": 1.6267299497909231,
      "language_loss": 0.87226021,
      "learning_rate": 3.908413867771257e-06,
      "loss": 0.89437997,
      "num_input_tokens_seen": 21764760,
      "step": 1030,
      "time_per_iteration": 3.647871971130371
    },
    {
      "auxiliary_loss_clip": 0.01145405,
      "auxiliary_loss_mlp": 0.01037256,
      "balance_loss_clip": 1.05254364,
      "balance_loss_mlp": 1.02463758,
      "epoch": 0.12397042024890279,
      "flos": 17347835116800.0,
      "grad_norm": 1.524408929272715,
      "language_loss": 0.80319017,
      "learning_rate": 3.908180695733311e-06,
      "loss": 0.82501686,
      "num_input_tokens_seen": 21784250,
      "step": 1031,
      "time_per_iteration": 2.648484468460083
    },
    {
      "auxiliary_loss_clip": 0.01182151,
      "auxiliary_loss_mlp": 0.01034503,
      "balance_loss_clip": 1.05799556,
      "balance_loss_mlp": 1.02243257,
      "epoch": 0.12409066313954187,
      "flos": 20412343854720.0,
      "grad_norm": 2.6852034483522518,
      "language_loss": 0.82516372,
      "learning_rate": 3.907947234225871e-06,
      "loss": 0.84733021,
      "num_input_tokens_seen": 21803260,
      "step": 1032,
      "time_per_iteration": 3.3604347705841064
    },
    {
      "auxiliary_loss_clip": 0.01169045,
      "auxiliary_loss_mlp": 0.00765093,
      "balance_loss_clip": 1.06125534,
      "balance_loss_mlp": 1.00022864,
      "epoch": 0.12421090603018096,
      "flos": 20736688688640.0,
      "grad_norm": 2.889544294026695,
      "language_loss": 0.87320733,
      "learning_rate": 3.907713483284352e-06,
      "loss": 0.89254868,
      "num_input_tokens_seen": 21822735,
      "step": 1033,
      "time_per_iteration": 2.6481363773345947
    },
    {
      "auxiliary_loss_clip": 0.01141915,
      "auxiliary_loss_mlp": 0.01041338,
      "balance_loss_clip": 1.04723144,
      "balance_loss_mlp": 1.02814138,
      "epoch": 0.12433114892082006,
      "flos": 24498834353280.0,
      "grad_norm": 2.2427117077793435,
      "language_loss": 0.97605789,
      "learning_rate": 3.907479442944216e-06,
      "loss": 0.99789035,
      "num_input_tokens_seen": 21841140,
      "step": 1034,
      "time_per_iteration": 2.6853604316711426
    },
    {
      "auxiliary_loss_clip": 0.01190188,
      "auxiliary_loss_mlp": 0.01036987,
      "balance_loss_clip": 1.06003511,
      "balance_loss_mlp": 1.02381492,
      "epoch": 0.12445139181145914,
      "flos": 19682315838720.0,
      "grad_norm": 2.3571210547877843,
      "language_loss": 0.92086506,
      "learning_rate": 3.907245113240963e-06,
      "loss": 0.94313681,
      "num_input_tokens_seen": 21859260,
      "step": 1035,
      "time_per_iteration": 2.5853703022003174
    },
    {
      "auxiliary_loss_clip": 0.01150754,
      "auxiliary_loss_mlp": 0.00765641,
      "balance_loss_clip": 1.05666637,
      "balance_loss_mlp": 1.00030804,
      "epoch": 0.12457163470209824,
      "flos": 46423087522560.0,
      "grad_norm": 2.3267401225289994,
      "language_loss": 0.73581517,
      "learning_rate": 3.907010494210144e-06,
      "loss": 0.75497913,
      "num_input_tokens_seen": 21881920,
      "step": 1036,
      "time_per_iteration": 2.9937655925750732
    },
    {
      "auxiliary_loss_clip": 0.01228263,
      "auxiliary_loss_mlp": 0.01040778,
      "balance_loss_clip": 1.0698427,
      "balance_loss_mlp": 1.02694368,
      "epoch": 0.12469187759273732,
      "flos": 20376289578240.0,
      "grad_norm": 2.002585741217859,
      "language_loss": 0.91801763,
      "learning_rate": 3.9067755858873495e-06,
      "loss": 0.94070816,
      "num_input_tokens_seen": 21898720,
      "step": 1037,
      "time_per_iteration": 2.7478253841400146
    },
    {
      "auxiliary_loss_clip": 0.01066286,
      "auxiliary_loss_mlp": 0.01008171,
      "balance_loss_clip": 1.02872181,
      "balance_loss_mlp": 1.00513077,
      "epoch": 0.12481212048337642,
      "flos": 69224641447680.0,
      "grad_norm": 0.8638611707032909,
      "language_loss": 0.62851834,
      "learning_rate": 3.906540388308214e-06,
      "loss": 0.64926291,
      "num_input_tokens_seen": 21958305,
      "step": 1038,
      "time_per_iteration": 3.291968822479248
    },
    {
      "auxiliary_loss_clip": 0.01179571,
      "auxiliary_loss_mlp": 0.010485,
      "balance_loss_clip": 1.05291259,
      "balance_loss_mlp": 1.03550625,
      "epoch": 0.12493236337401552,
      "flos": 18223696350720.0,
      "grad_norm": 1.845429960145007,
      "language_loss": 0.81219929,
      "learning_rate": 3.906304901508417e-06,
      "loss": 0.83447999,
      "num_input_tokens_seen": 21977205,
      "step": 1039,
      "time_per_iteration": 2.924229621887207
    },
    {
      "auxiliary_loss_clip": 0.01195296,
      "auxiliary_loss_mlp": 0.01049225,
      "balance_loss_clip": 1.06273901,
      "balance_loss_mlp": 1.03644001,
      "epoch": 0.12505260626465461,
      "flos": 30044375303040.0,
      "grad_norm": 3.5672044439007182,
      "language_loss": 0.75723374,
      "learning_rate": 3.9060691255236835e-06,
      "loss": 0.77967894,
      "num_input_tokens_seen": 21997770,
      "step": 1040,
      "time_per_iteration": 2.673452615737915
    },
    {
      "auxiliary_loss_clip": 0.01209333,
      "auxiliary_loss_mlp": 0.01042061,
      "balance_loss_clip": 1.06679845,
      "balance_loss_mlp": 1.029055,
      "epoch": 0.1251728491552937,
      "flos": 24433980347520.0,
      "grad_norm": 3.183680925716664,
      "language_loss": 0.80551809,
      "learning_rate": 3.905833060389778e-06,
      "loss": 0.82803208,
      "num_input_tokens_seen": 22021890,
      "step": 1041,
      "time_per_iteration": 2.638031482696533
    },
    {
      "auxiliary_loss_clip": 0.01181931,
      "auxiliary_loss_mlp": 0.01042124,
      "balance_loss_clip": 1.06664228,
      "balance_loss_mlp": 1.02908242,
      "epoch": 0.12529309204593278,
      "flos": 27119809952640.0,
      "grad_norm": 2.166609554138021,
      "language_loss": 0.78352815,
      "learning_rate": 3.905596706142513e-06,
      "loss": 0.80576873,
      "num_input_tokens_seen": 22043300,
      "step": 1042,
      "time_per_iteration": 2.651240348815918
    },
    {
      "auxiliary_loss_clip": 0.01126569,
      "auxiliary_loss_mlp": 0.01039594,
      "balance_loss_clip": 1.04941106,
      "balance_loss_mlp": 1.02643299,
      "epoch": 0.12541333493657186,
      "flos": 30774151923840.0,
      "grad_norm": 1.9822977057792772,
      "language_loss": 0.86120522,
      "learning_rate": 3.9053600628177435e-06,
      "loss": 0.88286686,
      "num_input_tokens_seen": 22062910,
      "step": 1043,
      "time_per_iteration": 2.7367255687713623
    },
    {
      "auxiliary_loss_clip": 0.01182061,
      "auxiliary_loss_mlp": 0.0104238,
      "balance_loss_clip": 1.05613101,
      "balance_loss_mlp": 1.02940965,
      "epoch": 0.12553357782721097,
      "flos": 23659566099840.0,
      "grad_norm": 1.988413109010975,
      "language_loss": 0.84559882,
      "learning_rate": 3.905123130451367e-06,
      "loss": 0.86784327,
      "num_input_tokens_seen": 22084010,
      "step": 1044,
      "time_per_iteration": 2.5980658531188965
    },
    {
      "auxiliary_loss_clip": 0.01154497,
      "auxiliary_loss_mlp": 0.00765721,
      "balance_loss_clip": 1.05280972,
      "balance_loss_mlp": 1.00039577,
      "epoch": 0.12565382071785006,
      "flos": 24863758577280.0,
      "grad_norm": 1.732340000842594,
      "language_loss": 0.79327172,
      "learning_rate": 3.904885909079326e-06,
      "loss": 0.81247389,
      "num_input_tokens_seen": 22102795,
      "step": 1045,
      "time_per_iteration": 2.695735216140747
    },
    {
      "auxiliary_loss_clip": 0.01226613,
      "auxiliary_loss_mlp": 0.00766383,
      "balance_loss_clip": 1.06585622,
      "balance_loss_mlp": 1.00028372,
      "epoch": 0.12577406360848914,
      "flos": 21360780518400.0,
      "grad_norm": 3.031987767461805,
      "language_loss": 0.78658402,
      "learning_rate": 3.904648398737607e-06,
      "loss": 0.80651397,
      "num_input_tokens_seen": 22121360,
      "step": 1046,
      "time_per_iteration": 2.5151336193084717
    },
    {
      "auxiliary_loss_clip": 0.01196983,
      "auxiliary_loss_mlp": 0.01045301,
      "balance_loss_clip": 1.05645752,
      "balance_loss_mlp": 1.03265905,
      "epoch": 0.12589430649912825,
      "flos": 36138056774400.0,
      "grad_norm": 1.9243053911304993,
      "language_loss": 0.77732939,
      "learning_rate": 3.9044105994622406e-06,
      "loss": 0.79975224,
      "num_input_tokens_seen": 22142505,
      "step": 1047,
      "time_per_iteration": 2.643507957458496
    },
    {
      "auxiliary_loss_clip": 0.01115667,
      "auxiliary_loss_mlp": 0.01044704,
      "balance_loss_clip": 1.05168605,
      "balance_loss_mlp": 1.02947521,
      "epoch": 0.12601454938976733,
      "flos": 25337671643520.0,
      "grad_norm": 1.8177091169299477,
      "language_loss": 0.81957865,
      "learning_rate": 3.9041725112893005e-06,
      "loss": 0.84118235,
      "num_input_tokens_seen": 22163730,
      "step": 1048,
      "time_per_iteration": 2.840036153793335
    },
    {
      "auxiliary_loss_clip": 0.01208839,
      "auxiliary_loss_mlp": 0.0104505,
      "balance_loss_clip": 1.06308544,
      "balance_loss_mlp": 1.03173399,
      "epoch": 0.12613479228040642,
      "flos": 15560094286080.0,
      "grad_norm": 1.8726913388736974,
      "language_loss": 0.75175798,
      "learning_rate": 3.903934134254904e-06,
      "loss": 0.77429688,
      "num_input_tokens_seen": 22181520,
      "step": 1049,
      "time_per_iteration": 2.8832597732543945
    },
    {
      "auxiliary_loss_clip": 0.01232851,
      "auxiliary_loss_mlp": 0.01044252,
      "balance_loss_clip": 1.06631458,
      "balance_loss_mlp": 1.03110361,
      "epoch": 0.1262550351710455,
      "flos": 21470595373440.0,
      "grad_norm": 2.032530154929103,
      "language_loss": 0.84973693,
      "learning_rate": 3.903695468395213e-06,
      "loss": 0.87250793,
      "num_input_tokens_seen": 22199390,
      "step": 1050,
      "time_per_iteration": 2.5235023498535156
    },
    {
      "auxiliary_loss_clip": 0.01244935,
      "auxiliary_loss_mlp": 0.01039928,
      "balance_loss_clip": 1.07108617,
      "balance_loss_mlp": 1.02725017,
      "epoch": 0.1263752780616846,
      "flos": 31576719456000.0,
      "grad_norm": 2.0904660427015993,
      "language_loss": 0.55293679,
      "learning_rate": 3.903456513746434e-06,
      "loss": 0.5757854,
      "num_input_tokens_seen": 22220365,
      "step": 1051,
      "time_per_iteration": 2.5350747108459473
    },
    {
      "auxiliary_loss_clip": 0.01173714,
      "auxiliary_loss_mlp": 0.01048457,
      "balance_loss_clip": 1.05905962,
      "balance_loss_mlp": 1.03486156,
      "epoch": 0.1264955209523237,
      "flos": 28768217927040.0,
      "grad_norm": 1.7431992245214825,
      "language_loss": 0.87510121,
      "learning_rate": 3.903217270344815e-06,
      "loss": 0.89732295,
      "num_input_tokens_seen": 22240615,
      "step": 1052,
      "time_per_iteration": 2.7249910831451416
    },
    {
      "auxiliary_loss_clip": 0.01153563,
      "auxiliary_loss_mlp": 0.01043072,
      "balance_loss_clip": 1.05133677,
      "balance_loss_mlp": 1.03012013,
      "epoch": 0.12661576384296278,
      "flos": 29241125412480.0,
      "grad_norm": 1.7883833187314064,
      "language_loss": 0.82457697,
      "learning_rate": 3.902977738226648e-06,
      "loss": 0.84654331,
      "num_input_tokens_seen": 22261350,
      "step": 1053,
      "time_per_iteration": 3.4210753440856934
    },
    {
      "auxiliary_loss_clip": 0.01125178,
      "auxiliary_loss_mlp": 0.01047352,
      "balance_loss_clip": 1.04793358,
      "balance_loss_mlp": 1.03329098,
      "epoch": 0.12673600673360189,
      "flos": 20850346298880.0,
      "grad_norm": 2.1106844061412184,
      "language_loss": 0.91275167,
      "learning_rate": 3.902737917428273e-06,
      "loss": 0.93447691,
      "num_input_tokens_seen": 22279515,
      "step": 1054,
      "time_per_iteration": 2.8802103996276855
    },
    {
      "auxiliary_loss_clip": 0.01190223,
      "auxiliary_loss_mlp": 0.01041592,
      "balance_loss_clip": 1.05720091,
      "balance_loss_mlp": 1.02877676,
      "epoch": 0.12685624962424097,
      "flos": 25263695583360.0,
      "grad_norm": 1.6353217256104104,
      "language_loss": 0.83843362,
      "learning_rate": 3.902497807986068e-06,
      "loss": 0.86075175,
      "num_input_tokens_seen": 22299535,
      "step": 1055,
      "time_per_iteration": 2.744133472442627
    },
    {
      "auxiliary_loss_clip": 0.01204508,
      "auxiliary_loss_mlp": 0.01038709,
      "balance_loss_clip": 1.0628624,
      "balance_loss_mlp": 1.02523184,
      "epoch": 0.12697649251488005,
      "flos": 27527109246720.0,
      "grad_norm": 1.6697096433503427,
      "language_loss": 0.83932161,
      "learning_rate": 3.902257409936458e-06,
      "loss": 0.86175382,
      "num_input_tokens_seen": 22320300,
      "step": 1056,
      "time_per_iteration": 3.7992489337921143
    },
    {
      "auxiliary_loss_clip": 0.01166188,
      "auxiliary_loss_mlp": 0.01038848,
      "balance_loss_clip": 1.0559684,
      "balance_loss_mlp": 1.02598524,
      "epoch": 0.12709673540551916,
      "flos": 21251863503360.0,
      "grad_norm": 3.5976382798489297,
      "language_loss": 0.84046268,
      "learning_rate": 3.902016723315912e-06,
      "loss": 0.86251301,
      "num_input_tokens_seen": 22338240,
      "step": 1057,
      "time_per_iteration": 2.6122212409973145
    },
    {
      "auxiliary_loss_clip": 0.0121539,
      "auxiliary_loss_mlp": 0.01049213,
      "balance_loss_clip": 1.06669581,
      "balance_loss_mlp": 1.03598642,
      "epoch": 0.12721697829615825,
      "flos": 25337707557120.0,
      "grad_norm": 2.916333657570754,
      "language_loss": 0.69284177,
      "learning_rate": 3.901775748160941e-06,
      "loss": 0.71548784,
      "num_input_tokens_seen": 22357420,
      "step": 1058,
      "time_per_iteration": 3.3130557537078857
    },
    {
      "auxiliary_loss_clip": 0.01122843,
      "auxiliary_loss_mlp": 0.01004883,
      "balance_loss_clip": 1.04509032,
      "balance_loss_mlp": 1.00190246,
      "epoch": 0.12733722118679733,
      "flos": 61943287754880.0,
      "grad_norm": 1.017029294332185,
      "language_loss": 0.6089347,
      "learning_rate": 3.901534484508101e-06,
      "loss": 0.63021195,
      "num_input_tokens_seen": 22420095,
      "step": 1059,
      "time_per_iteration": 3.137895107269287
    },
    {
      "auxiliary_loss_clip": 0.01195608,
      "auxiliary_loss_mlp": 0.01042976,
      "balance_loss_clip": 1.06241,
      "balance_loss_mlp": 1.02978539,
      "epoch": 0.1274574640774364,
      "flos": 26976742081920.0,
      "grad_norm": 4.583301007622591,
      "language_loss": 0.74837589,
      "learning_rate": 3.901292932393991e-06,
      "loss": 0.77076173,
      "num_input_tokens_seen": 22438975,
      "step": 1060,
      "time_per_iteration": 2.61873459815979
    },
    {
      "auxiliary_loss_clip": 0.01176823,
      "auxiliary_loss_mlp": 0.01048329,
      "balance_loss_clip": 1.06397665,
      "balance_loss_mlp": 1.03451812,
      "epoch": 0.12757770696807552,
      "flos": 22236318529920.0,
      "grad_norm": 2.0915225056238613,
      "language_loss": 0.8496995,
      "learning_rate": 3.9010510918552555e-06,
      "loss": 0.87195098,
      "num_input_tokens_seen": 22458050,
      "step": 1061,
      "time_per_iteration": 2.6358890533447266
    },
    {
      "auxiliary_loss_clip": 0.01209483,
      "auxiliary_loss_mlp": 0.00766391,
      "balance_loss_clip": 1.06300414,
      "balance_loss_mlp": 1.00030279,
      "epoch": 0.1276979498587146,
      "flos": 28547905858560.0,
      "grad_norm": 2.9494335622740477,
      "language_loss": 0.74421835,
      "learning_rate": 3.900808962928581e-06,
      "loss": 0.76397711,
      "num_input_tokens_seen": 22475665,
      "step": 1062,
      "time_per_iteration": 2.5951764583587646
    },
    {
      "auxiliary_loss_clip": 0.01205604,
      "auxiliary_loss_mlp": 0.01049962,
      "balance_loss_clip": 1.0624845,
      "balance_loss_mlp": 1.03801119,
      "epoch": 0.1278181927493537,
      "flos": 17420338719360.0,
      "grad_norm": 2.2456819310863847,
      "language_loss": 0.89357227,
      "learning_rate": 3.900566545650698e-06,
      "loss": 0.91612792,
      "num_input_tokens_seen": 22493335,
      "step": 1063,
      "time_per_iteration": 2.4975411891937256
    },
    {
      "auxiliary_loss_clip": 0.01180581,
      "auxiliary_loss_mlp": 0.01039532,
      "balance_loss_clip": 1.06138432,
      "balance_loss_mlp": 1.02548957,
      "epoch": 0.1279384356399928,
      "flos": 21138636856320.0,
      "grad_norm": 2.984691691710047,
      "language_loss": 0.82046503,
      "learning_rate": 3.900323840058381e-06,
      "loss": 0.84266615,
      "num_input_tokens_seen": 22511045,
      "step": 1064,
      "time_per_iteration": 2.590270757675171
    },
    {
      "auxiliary_loss_clip": 0.0119265,
      "auxiliary_loss_mlp": 0.01036064,
      "balance_loss_clip": 1.06112337,
      "balance_loss_mlp": 1.02327871,
      "epoch": 0.12805867853063188,
      "flos": 26576733248640.0,
      "grad_norm": 3.0539537336487186,
      "language_loss": 0.82220745,
      "learning_rate": 3.900080846188449e-06,
      "loss": 0.84449458,
      "num_input_tokens_seen": 22529635,
      "step": 1065,
      "time_per_iteration": 2.6053686141967773
    },
    {
      "auxiliary_loss_clip": 0.01181249,
      "auxiliary_loss_mlp": 0.01039872,
      "balance_loss_clip": 1.05989099,
      "balance_loss_mlp": 1.02596045,
      "epoch": 0.12817892142127096,
      "flos": 16436206915200.0,
      "grad_norm": 1.746860347123221,
      "language_loss": 0.81771636,
      "learning_rate": 3.8998375640777625e-06,
      "loss": 0.83992755,
      "num_input_tokens_seen": 22547505,
      "step": 1066,
      "time_per_iteration": 2.5773913860321045
    },
    {
      "auxiliary_loss_clip": 0.01059984,
      "auxiliary_loss_mlp": 0.01005942,
      "balance_loss_clip": 1.03952181,
      "balance_loss_mlp": 1.00336671,
      "epoch": 0.12829916431191005,
      "flos": 60757049099520.0,
      "grad_norm": 0.705743362611502,
      "language_loss": 0.52595162,
      "learning_rate": 3.899593993763229e-06,
      "loss": 0.54661095,
      "num_input_tokens_seen": 22608465,
      "step": 1067,
      "time_per_iteration": 3.2863776683807373
    },
    {
      "auxiliary_loss_clip": 0.01216035,
      "auxiliary_loss_mlp": 0.01044161,
      "balance_loss_clip": 1.06307304,
      "balance_loss_mlp": 1.03008819,
      "epoch": 0.12841940720254916,
      "flos": 29786895636480.0,
      "grad_norm": 3.578596515015446,
      "language_loss": 0.8135401,
      "learning_rate": 3.899350135281796e-06,
      "loss": 0.83614206,
      "num_input_tokens_seen": 22629465,
      "step": 1068,
      "time_per_iteration": 2.777129888534546
    },
    {
      "auxiliary_loss_clip": 0.01193028,
      "auxiliary_loss_mlp": 0.01042222,
      "balance_loss_clip": 1.06532359,
      "balance_loss_mlp": 1.02990174,
      "epoch": 0.12853965009318824,
      "flos": 25951851319680.0,
      "grad_norm": 2.2009137417247455,
      "language_loss": 0.79943752,
      "learning_rate": 3.8991059886704585e-06,
      "loss": 0.82179004,
      "num_input_tokens_seen": 22648970,
      "step": 1069,
      "time_per_iteration": 2.6172280311584473
    },
    {
      "auxiliary_loss_clip": 0.01229517,
      "auxiliary_loss_mlp": 0.01040282,
      "balance_loss_clip": 1.06649673,
      "balance_loss_mlp": 1.02564359,
      "epoch": 0.12865989298382732,
      "flos": 30846871008000.0,
      "grad_norm": 2.2772241762977425,
      "language_loss": 0.83015907,
      "learning_rate": 3.898861553966252e-06,
      "loss": 0.85285711,
      "num_input_tokens_seen": 22668620,
      "step": 1070,
      "time_per_iteration": 2.568157911300659
    },
    {
      "auxiliary_loss_clip": 0.01200772,
      "auxiliary_loss_mlp": 0.01049528,
      "balance_loss_clip": 1.06115711,
      "balance_loss_mlp": 1.03673077,
      "epoch": 0.12878013587446643,
      "flos": 25885776251520.0,
      "grad_norm": 2.012611131982674,
      "language_loss": 0.88284552,
      "learning_rate": 3.898616831206257e-06,
      "loss": 0.90534854,
      "num_input_tokens_seen": 22689045,
      "step": 1071,
      "time_per_iteration": 2.5904295444488525
    },
    {
      "auxiliary_loss_clip": 0.01218022,
      "auxiliary_loss_mlp": 0.01042874,
      "balance_loss_clip": 1.06303239,
      "balance_loss_mlp": 1.02892685,
      "epoch": 0.12890037876510552,
      "flos": 23333138277120.0,
      "grad_norm": 2.5371860217804176,
      "language_loss": 0.7705158,
      "learning_rate": 3.8983718204276e-06,
      "loss": 0.79312479,
      "num_input_tokens_seen": 22711265,
      "step": 1072,
      "time_per_iteration": 2.556265115737915
    },
    {
      "auxiliary_loss_clip": 0.01163159,
      "auxiliary_loss_mlp": 0.01049885,
      "balance_loss_clip": 1.05835342,
      "balance_loss_mlp": 1.03602111,
      "epoch": 0.1290206216557446,
      "flos": 23587242065280.0,
      "grad_norm": 1.6652895752883516,
      "language_loss": 0.82642382,
      "learning_rate": 3.898126521667446e-06,
      "loss": 0.84855425,
      "num_input_tokens_seen": 22731420,
      "step": 1073,
      "time_per_iteration": 2.6518611907958984
    },
    {
      "auxiliary_loss_clip": 0.01199174,
      "auxiliary_loss_mlp": 0.01042514,
      "balance_loss_clip": 1.06496334,
      "balance_loss_mlp": 1.02962148,
      "epoch": 0.12914086454638368,
      "flos": 24170610850560.0,
      "grad_norm": 1.7024151939273084,
      "language_loss": 0.83033264,
      "learning_rate": 3.897880934963007e-06,
      "loss": 0.85274947,
      "num_input_tokens_seen": 22750970,
      "step": 1074,
      "time_per_iteration": 2.6019294261932373
    },
    {
      "auxiliary_loss_clip": 0.01176816,
      "auxiliary_loss_mlp": 0.01044682,
      "balance_loss_clip": 1.05292964,
      "balance_loss_mlp": 1.03173637,
      "epoch": 0.1292611074370228,
      "flos": 20267157081600.0,
      "grad_norm": 2.050166937823833,
      "language_loss": 0.78627151,
      "learning_rate": 3.89763506035154e-06,
      "loss": 0.80848646,
      "num_input_tokens_seen": 22768820,
      "step": 1075,
      "time_per_iteration": 2.6394834518432617
    },
    {
      "auxiliary_loss_clip": 0.01186474,
      "auxiliary_loss_mlp": 0.01040915,
      "balance_loss_clip": 1.0629971,
      "balance_loss_mlp": 1.02678895,
      "epoch": 0.12938135032766188,
      "flos": 27377684668800.0,
      "grad_norm": 3.25265627340935,
      "language_loss": 0.81160069,
      "learning_rate": 3.897388897870343e-06,
      "loss": 0.83387458,
      "num_input_tokens_seen": 22789460,
      "step": 1076,
      "time_per_iteration": 2.5954556465148926
    },
    {
      "auxiliary_loss_clip": 0.01158324,
      "auxiliary_loss_mlp": 0.01051958,
      "balance_loss_clip": 1.05140233,
      "balance_loss_mlp": 1.03714049,
      "epoch": 0.12950159321830096,
      "flos": 29277107861760.0,
      "grad_norm": 1.7707255009338263,
      "language_loss": 0.74820328,
      "learning_rate": 3.89714244755676e-06,
      "loss": 0.77030611,
      "num_input_tokens_seen": 22810820,
      "step": 1077,
      "time_per_iteration": 2.654404401779175
    },
    {
      "auxiliary_loss_clip": 0.01194461,
      "auxiliary_loss_mlp": 0.00767032,
      "balance_loss_clip": 1.06532586,
      "balance_loss_mlp": 1.00021553,
      "epoch": 0.12962183610894007,
      "flos": 24534888629760.0,
      "grad_norm": 2.5821524936448736,
      "language_loss": 0.86584806,
      "learning_rate": 3.896895709448175e-06,
      "loss": 0.885463,
      "num_input_tokens_seen": 22830570,
      "step": 1078,
      "time_per_iteration": 2.5715672969818115
    },
    {
      "auxiliary_loss_clip": 0.01206662,
      "auxiliary_loss_mlp": 0.0104394,
      "balance_loss_clip": 1.05725932,
      "balance_loss_mlp": 1.02942586,
      "epoch": 0.12974207899957915,
      "flos": 11215944552960.0,
      "grad_norm": 2.7802953072617673,
      "language_loss": 0.77236688,
      "learning_rate": 3.896648683582019e-06,
      "loss": 0.79487294,
      "num_input_tokens_seen": 22845905,
      "step": 1079,
      "time_per_iteration": 4.324894189834595
    },
    {
      "auxiliary_loss_clip": 0.01207582,
      "auxiliary_loss_mlp": 0.01040038,
      "balance_loss_clip": 1.06055689,
      "balance_loss_mlp": 1.02690697,
      "epoch": 0.12986232189021824,
      "flos": 24717889445760.0,
      "grad_norm": 1.9378907488322357,
      "language_loss": 0.80983812,
      "learning_rate": 3.896401369995766e-06,
      "loss": 0.83231431,
      "num_input_tokens_seen": 22865710,
      "step": 1080,
      "time_per_iteration": 2.5774264335632324
    },
    {
      "auxiliary_loss_clip": 0.01183347,
      "auxiliary_loss_mlp": 0.00766582,
      "balance_loss_clip": 1.05913901,
      "balance_loss_mlp": 1.00036979,
      "epoch": 0.12998256478085732,
      "flos": 23915357827200.0,
      "grad_norm": 1.7284025464883426,
      "language_loss": 0.79628402,
      "learning_rate": 3.896153768726932e-06,
      "loss": 0.81578332,
      "num_input_tokens_seen": 22886020,
      "step": 1081,
      "time_per_iteration": 3.3377022743225098
    },
    {
      "auxiliary_loss_clip": 0.01196431,
      "auxiliary_loss_mlp": 0.01041559,
      "balance_loss_clip": 1.06294203,
      "balance_loss_mlp": 1.02807665,
      "epoch": 0.13010280767149643,
      "flos": 18624207974400.0,
      "grad_norm": 2.3100597658553697,
      "language_loss": 0.88169396,
      "learning_rate": 3.8959058798130806e-06,
      "loss": 0.90407389,
      "num_input_tokens_seen": 22903995,
      "step": 1082,
      "time_per_iteration": 2.561387538909912
    },
    {
      "auxiliary_loss_clip": 0.01180566,
      "auxiliary_loss_mlp": 0.01038892,
      "balance_loss_clip": 1.05700183,
      "balance_loss_mlp": 1.02396739,
      "epoch": 0.1302230505621355,
      "flos": 22783992174720.0,
      "grad_norm": 1.8597353455277887,
      "language_loss": 0.75039721,
      "learning_rate": 3.895657703291814e-06,
      "loss": 0.77259171,
      "num_input_tokens_seen": 22924100,
      "step": 1083,
      "time_per_iteration": 3.3469653129577637
    },
    {
      "auxiliary_loss_clip": 0.01216254,
      "auxiliary_loss_mlp": 0.01045296,
      "balance_loss_clip": 1.06342697,
      "balance_loss_mlp": 1.02979887,
      "epoch": 0.1303432934527746,
      "flos": 21323612920320.0,
      "grad_norm": 3.463765949857324,
      "language_loss": 0.79841822,
      "learning_rate": 3.895409239200781e-06,
      "loss": 0.82103372,
      "num_input_tokens_seen": 22939985,
      "step": 1084,
      "time_per_iteration": 2.4953439235687256
    },
    {
      "auxiliary_loss_clip": 0.01194797,
      "auxiliary_loss_mlp": 0.01044706,
      "balance_loss_clip": 1.06127763,
      "balance_loss_mlp": 1.03007936,
      "epoch": 0.1304635363434137,
      "flos": 20922490765440.0,
      "grad_norm": 2.774590121848232,
      "language_loss": 0.91793871,
      "learning_rate": 3.895160487577673e-06,
      "loss": 0.94033372,
      "num_input_tokens_seen": 22957555,
      "step": 1085,
      "time_per_iteration": 2.565598726272583
    },
    {
      "auxiliary_loss_clip": 0.01114322,
      "auxiliary_loss_mlp": 0.01006599,
      "balance_loss_clip": 1.04386497,
      "balance_loss_mlp": 1.00383365,
      "epoch": 0.1305837792340528,
      "flos": 63245659080960.0,
      "grad_norm": 0.7916757764629753,
      "language_loss": 0.60924852,
      "learning_rate": 3.894911448460226e-06,
      "loss": 0.6304577,
      "num_input_tokens_seen": 23016870,
      "step": 1086,
      "time_per_iteration": 3.0040197372436523
    },
    {
      "auxiliary_loss_clip": 0.01208677,
      "auxiliary_loss_mlp": 0.01040401,
      "balance_loss_clip": 1.06478381,
      "balance_loss_mlp": 1.02760363,
      "epoch": 0.13070402212469187,
      "flos": 26428852955520.0,
      "grad_norm": 2.1938591050028786,
      "language_loss": 0.72655904,
      "learning_rate": 3.8946621218862195e-06,
      "loss": 0.74904984,
      "num_input_tokens_seen": 23037870,
      "step": 1087,
      "time_per_iteration": 2.623931646347046
    },
    {
      "auxiliary_loss_clip": 0.01227389,
      "auxiliary_loss_mlp": 0.01039551,
      "balance_loss_clip": 1.06879067,
      "balance_loss_mlp": 1.02547872,
      "epoch": 0.13082426501533098,
      "flos": 27673409341440.0,
      "grad_norm": 1.761888229955963,
      "language_loss": 0.89150524,
      "learning_rate": 3.894412507893475e-06,
      "loss": 0.91417462,
      "num_input_tokens_seen": 23058150,
      "step": 1088,
      "time_per_iteration": 2.546581983566284
    },
    {
      "auxiliary_loss_clip": 0.01227085,
      "auxiliary_loss_mlp": 0.0104175,
      "balance_loss_clip": 1.06795609,
      "balance_loss_mlp": 1.02781439,
      "epoch": 0.13094450790597006,
      "flos": 24826770547200.0,
      "grad_norm": 2.003474237592484,
      "language_loss": 0.7237227,
      "learning_rate": 3.894162606519859e-06,
      "loss": 0.74641109,
      "num_input_tokens_seen": 23077100,
      "step": 1089,
      "time_per_iteration": 2.531041145324707
    },
    {
      "auxiliary_loss_clip": 0.01216679,
      "auxiliary_loss_mlp": 0.01044695,
      "balance_loss_clip": 1.06884551,
      "balance_loss_mlp": 1.03144479,
      "epoch": 0.13106475079660915,
      "flos": 19062605468160.0,
      "grad_norm": 1.9289357849707667,
      "language_loss": 0.773072,
      "learning_rate": 3.893912417803282e-06,
      "loss": 0.79568571,
      "num_input_tokens_seen": 23096815,
      "step": 1090,
      "time_per_iteration": 2.5186145305633545
    },
    {
      "auxiliary_loss_clip": 0.011913,
      "auxiliary_loss_mlp": 0.01038028,
      "balance_loss_clip": 1.06192493,
      "balance_loss_mlp": 1.02444386,
      "epoch": 0.13118499368724823,
      "flos": 28913189218560.0,
      "grad_norm": 1.9422684880273917,
      "language_loss": 0.77077872,
      "learning_rate": 3.8936619417816975e-06,
      "loss": 0.79307193,
      "num_input_tokens_seen": 23117145,
      "step": 1091,
      "time_per_iteration": 2.6151657104492188
    },
    {
      "auxiliary_loss_clip": 0.01180994,
      "auxiliary_loss_mlp": 0.00766353,
      "balance_loss_clip": 1.06437349,
      "balance_loss_mlp": 1.00030279,
      "epoch": 0.13130523657788734,
      "flos": 14283398206080.0,
      "grad_norm": 2.2235979236073757,
      "language_loss": 0.71438289,
      "learning_rate": 3.8934111784931015e-06,
      "loss": 0.73385632,
      "num_input_tokens_seen": 23134595,
      "step": 1092,
      "time_per_iteration": 2.5856704711914062
    },
    {
      "auxiliary_loss_clip": 0.0113345,
      "auxiliary_loss_mlp": 0.01002927,
      "balance_loss_clip": 1.04092431,
      "balance_loss_mlp": 1.00019681,
      "epoch": 0.13142547946852642,
      "flos": 70174155519360.0,
      "grad_norm": 0.9092818974386092,
      "language_loss": 0.59057868,
      "learning_rate": 3.893160127975535e-06,
      "loss": 0.61194241,
      "num_input_tokens_seen": 23195285,
      "step": 1093,
      "time_per_iteration": 3.1993348598480225
    },
    {
      "auxiliary_loss_clip": 0.01208218,
      "auxiliary_loss_mlp": 0.0104441,
      "balance_loss_clip": 1.06302667,
      "balance_loss_mlp": 1.03198814,
      "epoch": 0.1315457223591655,
      "flos": 45805998844800.0,
      "grad_norm": 2.360678246295909,
      "language_loss": 0.81013799,
      "learning_rate": 3.8929087902670826e-06,
      "loss": 0.83266425,
      "num_input_tokens_seen": 23216915,
      "step": 1094,
      "time_per_iteration": 2.7434816360473633
    },
    {
      "auxiliary_loss_clip": 0.01090297,
      "auxiliary_loss_mlp": 0.01003923,
      "balance_loss_clip": 1.0349524,
      "balance_loss_mlp": 1.00096631,
      "epoch": 0.13166596524980462,
      "flos": 62881165820160.0,
      "grad_norm": 0.9412134648503369,
      "language_loss": 0.60737789,
      "learning_rate": 3.8926571654058715e-06,
      "loss": 0.6283201,
      "num_input_tokens_seen": 23273560,
      "step": 1095,
      "time_per_iteration": 3.0864174365997314
    },
    {
      "auxiliary_loss_clip": 0.01180686,
      "auxiliary_loss_mlp": 0.01044875,
      "balance_loss_clip": 1.05673146,
      "balance_loss_mlp": 1.03240538,
      "epoch": 0.1317862081404437,
      "flos": 23586523793280.0,
      "grad_norm": 2.6098195233473893,
      "language_loss": 0.76930809,
      "learning_rate": 3.892405253430074e-06,
      "loss": 0.79156369,
      "num_input_tokens_seen": 23291080,
      "step": 1096,
      "time_per_iteration": 2.5801501274108887
    },
    {
      "auxiliary_loss_clip": 0.01148079,
      "auxiliary_loss_mlp": 0.01040417,
      "balance_loss_clip": 1.05751944,
      "balance_loss_mlp": 1.0275898,
      "epoch": 0.13190645103108278,
      "flos": 20260764460800.0,
      "grad_norm": 1.8089845616647064,
      "language_loss": 0.82210159,
      "learning_rate": 3.892153054377904e-06,
      "loss": 0.84398657,
      "num_input_tokens_seen": 23308485,
      "step": 1097,
      "time_per_iteration": 2.6081185340881348
    },
    {
      "auxiliary_loss_clip": 0.01108857,
      "auxiliary_loss_mlp": 0.01003086,
      "balance_loss_clip": 1.03846908,
      "balance_loss_mlp": 1.00026083,
      "epoch": 0.13202669392172187,
      "flos": 53455440136320.0,
      "grad_norm": 0.932914852906167,
      "language_loss": 0.59342825,
      "learning_rate": 3.891900568287619e-06,
      "loss": 0.61454761,
      "num_input_tokens_seen": 23360870,
      "step": 1098,
      "time_per_iteration": 2.9800024032592773
    },
    {
      "auxiliary_loss_clip": 0.01148658,
      "auxiliary_loss_mlp": 0.01045735,
      "balance_loss_clip": 1.05144739,
      "balance_loss_mlp": 1.03139448,
      "epoch": 0.13214693681236098,
      "flos": 15851293845120.0,
      "grad_norm": 2.3871342125002797,
      "language_loss": 0.7183131,
      "learning_rate": 3.891647795197523e-06,
      "loss": 0.74025702,
      "num_input_tokens_seen": 23376910,
      "step": 1099,
      "time_per_iteration": 2.8068671226501465
    },
    {
      "auxiliary_loss_clip": 0.0122708,
      "auxiliary_loss_mlp": 0.01042541,
      "balance_loss_clip": 1.06670141,
      "balance_loss_mlp": 1.02772284,
      "epoch": 0.13226717970300006,
      "flos": 19353840940800.0,
      "grad_norm": 2.1960748307472535,
      "language_loss": 0.68700582,
      "learning_rate": 3.8913947351459605e-06,
      "loss": 0.70970201,
      "num_input_tokens_seen": 23394450,
      "step": 1100,
      "time_per_iteration": 2.839862823486328
    },
    {
      "auxiliary_loss_clip": 0.01171138,
      "auxiliary_loss_mlp": 0.01049439,
      "balance_loss_clip": 1.05611372,
      "balance_loss_mlp": 1.03603423,
      "epoch": 0.13238742259363914,
      "flos": 20698084546560.0,
      "grad_norm": 1.7918124327655796,
      "language_loss": 0.67936587,
      "learning_rate": 3.89114138817132e-06,
      "loss": 0.7015717,
      "num_input_tokens_seen": 23411115,
      "step": 1101,
      "time_per_iteration": 2.590651273727417
    },
    {
      "auxiliary_loss_clip": 0.01206769,
      "auxiliary_loss_mlp": 0.01046747,
      "balance_loss_clip": 1.06067073,
      "balance_loss_mlp": 1.03396773,
      "epoch": 0.13250766548427825,
      "flos": 21032449274880.0,
      "grad_norm": 1.9692813189103746,
      "language_loss": 0.84473538,
      "learning_rate": 3.890887754312035e-06,
      "loss": 0.86727053,
      "num_input_tokens_seen": 23429360,
      "step": 1102,
      "time_per_iteration": 2.5450997352600098
    },
    {
      "auxiliary_loss_clip": 0.0117242,
      "auxiliary_loss_mlp": 0.01044022,
      "balance_loss_clip": 1.05633008,
      "balance_loss_mlp": 1.03048015,
      "epoch": 0.13262790837491734,
      "flos": 22637871648000.0,
      "grad_norm": 3.5687437191965707,
      "language_loss": 0.87843394,
      "learning_rate": 3.890633833606581e-06,
      "loss": 0.90059835,
      "num_input_tokens_seen": 23449050,
      "step": 1103,
      "time_per_iteration": 2.5908615589141846
    },
    {
      "auxiliary_loss_clip": 0.01162696,
      "auxiliary_loss_mlp": 0.01043827,
      "balance_loss_clip": 1.05823588,
      "balance_loss_mlp": 1.03091085,
      "epoch": 0.13274815126555642,
      "flos": 19683141851520.0,
      "grad_norm": 1.862093509349329,
      "language_loss": 0.69631255,
      "learning_rate": 3.890379626093477e-06,
      "loss": 0.71837777,
      "num_input_tokens_seen": 23468800,
      "step": 1104,
      "time_per_iteration": 2.634462594985962
    },
    {
      "auxiliary_loss_clip": 0.01198139,
      "auxiliary_loss_mlp": 0.01040936,
      "balance_loss_clip": 1.05947018,
      "balance_loss_mlp": 1.02573669,
      "epoch": 0.1328683941561955,
      "flos": 21317687176320.0,
      "grad_norm": 2.137378234892296,
      "language_loss": 0.92177927,
      "learning_rate": 3.890125131811287e-06,
      "loss": 0.94417006,
      "num_input_tokens_seen": 23486850,
      "step": 1105,
      "time_per_iteration": 4.032005548477173
    },
    {
      "auxiliary_loss_clip": 0.01201997,
      "auxiliary_loss_mlp": 0.01038113,
      "balance_loss_clip": 1.06247556,
      "balance_loss_mlp": 1.02522635,
      "epoch": 0.1329886370468346,
      "flos": 13699131580800.0,
      "grad_norm": 1.85386277334166,
      "language_loss": 0.75779581,
      "learning_rate": 3.889870350798618e-06,
      "loss": 0.78019691,
      "num_input_tokens_seen": 23504195,
      "step": 1106,
      "time_per_iteration": 2.516993284225464
    },
    {
      "auxiliary_loss_clip": 0.01169062,
      "auxiliary_loss_mlp": 0.01036859,
      "balance_loss_clip": 1.05571938,
      "balance_loss_mlp": 1.02361441,
      "epoch": 0.1331088799374737,
      "flos": 21032413361280.0,
      "grad_norm": 1.5524014457193724,
      "language_loss": 0.78374982,
      "learning_rate": 3.889615283094119e-06,
      "loss": 0.80580902,
      "num_input_tokens_seen": 23523385,
      "step": 1107,
      "time_per_iteration": 3.3486320972442627
    },
    {
      "auxiliary_loss_clip": 0.01190428,
      "auxiliary_loss_mlp": 0.01036333,
      "balance_loss_clip": 1.05887806,
      "balance_loss_mlp": 1.02237344,
      "epoch": 0.13322912282811278,
      "flos": 18260432985600.0,
      "grad_norm": 2.2440070388980544,
      "language_loss": 0.8438983,
      "learning_rate": 3.889359928736485e-06,
      "loss": 0.86616588,
      "num_input_tokens_seen": 23541330,
      "step": 1108,
      "time_per_iteration": 2.5311524868011475
    },
    {
      "auxiliary_loss_clip": 0.01147376,
      "auxiliary_loss_mlp": 0.01039538,
      "balance_loss_clip": 1.05124569,
      "balance_loss_mlp": 1.02558494,
      "epoch": 0.1333493657187519,
      "flos": 24460876656000.0,
      "grad_norm": 4.079622783452121,
      "language_loss": 0.91179734,
      "learning_rate": 3.889104287764451e-06,
      "loss": 0.93366647,
      "num_input_tokens_seen": 23561705,
      "step": 1109,
      "time_per_iteration": 3.4294443130493164
    },
    {
      "auxiliary_loss_clip": 0.01183173,
      "auxiliary_loss_mlp": 0.00766066,
      "balance_loss_clip": 1.05533051,
      "balance_loss_mlp": 1.00024211,
      "epoch": 0.13346960860939097,
      "flos": 22158930677760.0,
      "grad_norm": 2.2617758511087183,
      "language_loss": 0.90380561,
      "learning_rate": 3.888848360216798e-06,
      "loss": 0.923298,
      "num_input_tokens_seen": 23579350,
      "step": 1110,
      "time_per_iteration": 2.573270559310913
    },
    {
      "auxiliary_loss_clip": 0.01084216,
      "auxiliary_loss_mlp": 0.01003348,
      "balance_loss_clip": 1.0375545,
      "balance_loss_mlp": 1.0008086,
      "epoch": 0.13358985150003005,
      "flos": 67931212608000.0,
      "grad_norm": 0.8084575142605429,
      "language_loss": 0.56668991,
      "learning_rate": 3.888592146132351e-06,
      "loss": 0.58756554,
      "num_input_tokens_seen": 23640620,
      "step": 1111,
      "time_per_iteration": 3.2617316246032715
    },
    {
      "auxiliary_loss_clip": 0.01198335,
      "auxiliary_loss_mlp": 0.01044617,
      "balance_loss_clip": 1.06410849,
      "balance_loss_mlp": 1.03154564,
      "epoch": 0.13371009439066917,
      "flos": 26834284742400.0,
      "grad_norm": 1.7813707911172525,
      "language_loss": 0.78622556,
      "learning_rate": 3.888335645549978e-06,
      "loss": 0.80865514,
      "num_input_tokens_seen": 23661040,
      "step": 1112,
      "time_per_iteration": 2.6034839153289795
    },
    {
      "auxiliary_loss_clip": 0.01197701,
      "auxiliary_loss_mlp": 0.01043353,
      "balance_loss_clip": 1.06247711,
      "balance_loss_mlp": 1.02955449,
      "epoch": 0.13383033728130825,
      "flos": 26322844942080.0,
      "grad_norm": 2.2419167870574954,
      "language_loss": 0.81366342,
      "learning_rate": 3.888078858508588e-06,
      "loss": 0.83607399,
      "num_input_tokens_seen": 23680900,
      "step": 1113,
      "time_per_iteration": 2.625227451324463
    },
    {
      "auxiliary_loss_clip": 0.0121262,
      "auxiliary_loss_mlp": 0.01046338,
      "balance_loss_clip": 1.06595469,
      "balance_loss_mlp": 1.03314757,
      "epoch": 0.13395058017194733,
      "flos": 22563931501440.0,
      "grad_norm": 2.1833339470718687,
      "language_loss": 0.84367526,
      "learning_rate": 3.8878217850471365e-06,
      "loss": 0.86626488,
      "num_input_tokens_seen": 23700815,
      "step": 1114,
      "time_per_iteration": 2.557391881942749
    },
    {
      "auxiliary_loss_clip": 0.01229942,
      "auxiliary_loss_mlp": 0.0105061,
      "balance_loss_clip": 1.07048035,
      "balance_loss_mlp": 1.03750873,
      "epoch": 0.13407082306258641,
      "flos": 25810938264960.0,
      "grad_norm": 1.9793071006357847,
      "language_loss": 0.74198174,
      "learning_rate": 3.887564425204621e-06,
      "loss": 0.76478726,
      "num_input_tokens_seen": 23722500,
      "step": 1115,
      "time_per_iteration": 2.532060146331787
    },
    {
      "auxiliary_loss_clip": 0.01105817,
      "auxiliary_loss_mlp": 0.01004277,
      "balance_loss_clip": 1.0369699,
      "balance_loss_mlp": 1.0015831,
      "epoch": 0.13419106595322552,
      "flos": 68338365269760.0,
      "grad_norm": 0.8375067357806419,
      "language_loss": 0.54617047,
      "learning_rate": 3.887306779020083e-06,
      "loss": 0.56727141,
      "num_input_tokens_seen": 23777155,
      "step": 1116,
      "time_per_iteration": 3.0334224700927734
    },
    {
      "auxiliary_loss_clip": 0.01197946,
      "auxiliary_loss_mlp": 0.01038841,
      "balance_loss_clip": 1.05839825,
      "balance_loss_mlp": 1.02503014,
      "epoch": 0.1343113088438646,
      "flos": 20449080489600.0,
      "grad_norm": 2.5443833623615104,
      "language_loss": 0.70434254,
      "learning_rate": 3.887048846532608e-06,
      "loss": 0.72671044,
      "num_input_tokens_seen": 23794130,
      "step": 1117,
      "time_per_iteration": 2.5431957244873047
    },
    {
      "auxiliary_loss_clip": 0.01110717,
      "auxiliary_loss_mlp": 0.0100408,
      "balance_loss_clip": 1.03667116,
      "balance_loss_mlp": 1.00146961,
      "epoch": 0.1344315517345037,
      "flos": 67389784951680.0,
      "grad_norm": 0.7584604502753379,
      "language_loss": 0.5808382,
      "learning_rate": 3.8867906277813224e-06,
      "loss": 0.60198617,
      "num_input_tokens_seen": 23852285,
      "step": 1118,
      "time_per_iteration": 2.9922373294830322
    },
    {
      "auxiliary_loss_clip": 0.0122851,
      "auxiliary_loss_mlp": 0.01042263,
      "balance_loss_clip": 1.06693912,
      "balance_loss_mlp": 1.02975774,
      "epoch": 0.1345517946251428,
      "flos": 40734442788480.0,
      "grad_norm": 1.9957198666362397,
      "language_loss": 0.74397063,
      "learning_rate": 3.886532122805399e-06,
      "loss": 0.76667833,
      "num_input_tokens_seen": 23874765,
      "step": 1119,
      "time_per_iteration": 2.668661117553711
    },
    {
      "auxiliary_loss_clip": 0.01244855,
      "auxiliary_loss_mlp": 0.01041674,
      "balance_loss_clip": 1.06995749,
      "balance_loss_mlp": 1.02819192,
      "epoch": 0.13467203751578188,
      "flos": 22816850140800.0,
      "grad_norm": 2.4004090034133574,
      "language_loss": 0.89666444,
      "learning_rate": 3.886273331644053e-06,
      "loss": 0.9195298,
      "num_input_tokens_seen": 23893635,
      "step": 1120,
      "time_per_iteration": 2.4546267986297607
    },
    {
      "auxiliary_loss_clip": 0.01154614,
      "auxiliary_loss_mlp": 0.01042436,
      "balance_loss_clip": 1.05648351,
      "balance_loss_mlp": 1.02930474,
      "epoch": 0.13479228040642097,
      "flos": 17091576512640.0,
      "grad_norm": 1.9042177988578142,
      "language_loss": 0.81959808,
      "learning_rate": 3.886014254336542e-06,
      "loss": 0.84156859,
      "num_input_tokens_seen": 23910110,
      "step": 1121,
      "time_per_iteration": 2.6014654636383057
    },
    {
      "auxiliary_loss_clip": 0.01182168,
      "auxiliary_loss_mlp": 0.01039514,
      "balance_loss_clip": 1.06122804,
      "balance_loss_mlp": 1.02677608,
      "epoch": 0.13491252329706005,
      "flos": 23730525417600.0,
      "grad_norm": 2.0257536477799527,
      "language_loss": 0.92580396,
      "learning_rate": 3.885754890922168e-06,
      "loss": 0.94802082,
      "num_input_tokens_seen": 23930440,
      "step": 1122,
      "time_per_iteration": 2.5711748600006104
    },
    {
      "auxiliary_loss_clip": 0.01198643,
      "auxiliary_loss_mlp": 0.01046032,
      "balance_loss_clip": 1.06172526,
      "balance_loss_mlp": 1.03141117,
      "epoch": 0.13503276618769916,
      "flos": 34127058960000.0,
      "grad_norm": 4.917909856126611,
      "language_loss": 0.78660369,
      "learning_rate": 3.885495241440277e-06,
      "loss": 0.80905044,
      "num_input_tokens_seen": 23954535,
      "step": 1123,
      "time_per_iteration": 2.674233913421631
    },
    {
      "auxiliary_loss_clip": 0.01179818,
      "auxiliary_loss_mlp": 0.01048662,
      "balance_loss_clip": 1.06222129,
      "balance_loss_mlp": 1.03480959,
      "epoch": 0.13515300907833824,
      "flos": 17712328377600.0,
      "grad_norm": 2.0023050823090096,
      "language_loss": 0.73953569,
      "learning_rate": 3.885235305930257e-06,
      "loss": 0.7618205,
      "num_input_tokens_seen": 23972735,
      "step": 1124,
      "time_per_iteration": 2.554313898086548
    },
    {
      "auxiliary_loss_clip": 0.01194426,
      "auxiliary_loss_mlp": 0.01051573,
      "balance_loss_clip": 1.06043673,
      "balance_loss_mlp": 1.0390265,
      "epoch": 0.13527325196897733,
      "flos": 20260872201600.0,
      "grad_norm": 1.9697145941857137,
      "language_loss": 0.85385478,
      "learning_rate": 3.884975084431539e-06,
      "loss": 0.87631476,
      "num_input_tokens_seen": 23987685,
      "step": 1125,
      "time_per_iteration": 2.5437698364257812
    },
    {
      "auxiliary_loss_clip": 0.01188535,
      "auxiliary_loss_mlp": 0.00766389,
      "balance_loss_clip": 1.05958486,
      "balance_loss_mlp": 1.00027788,
      "epoch": 0.13539349485961644,
      "flos": 18186492839040.0,
      "grad_norm": 2.422895490341595,
      "language_loss": 0.91478294,
      "learning_rate": 3.8847145769836e-06,
      "loss": 0.93433213,
      "num_input_tokens_seen": 24004105,
      "step": 1126,
      "time_per_iteration": 2.54579496383667
    },
    {
      "auxiliary_loss_clip": 0.01197461,
      "auxiliary_loss_mlp": 0.00766461,
      "balance_loss_clip": 1.0619874,
      "balance_loss_mlp": 1.00030684,
      "epoch": 0.13551373775025552,
      "flos": 19317463441920.0,
      "grad_norm": 2.2124961732820387,
      "language_loss": 0.66089928,
      "learning_rate": 3.884453783625959e-06,
      "loss": 0.68053854,
      "num_input_tokens_seen": 24021715,
      "step": 1127,
      "time_per_iteration": 2.55527663230896
    },
    {
      "auxiliary_loss_clip": 0.01181692,
      "auxiliary_loss_mlp": 0.01044051,
      "balance_loss_clip": 1.06082702,
      "balance_loss_mlp": 1.03106284,
      "epoch": 0.1356339806408946,
      "flos": 20850813175680.0,
      "grad_norm": 2.539092537209517,
      "language_loss": 0.85010266,
      "learning_rate": 3.884192704398176e-06,
      "loss": 0.87236005,
      "num_input_tokens_seen": 24038915,
      "step": 1128,
      "time_per_iteration": 2.5501911640167236
    },
    {
      "auxiliary_loss_clip": 0.01161575,
      "auxiliary_loss_mlp": 0.01038737,
      "balance_loss_clip": 1.06301594,
      "balance_loss_mlp": 1.02623188,
      "epoch": 0.13575422353153369,
      "flos": 50476037696640.0,
      "grad_norm": 1.7376150086511697,
      "language_loss": 0.74573529,
      "learning_rate": 3.883931339339858e-06,
      "loss": 0.7677384,
      "num_input_tokens_seen": 24063300,
      "step": 1129,
      "time_per_iteration": 2.8919591903686523
    },
    {
      "auxiliary_loss_clip": 0.01219053,
      "auxiliary_loss_mlp": 0.01043161,
      "balance_loss_clip": 1.06513858,
      "balance_loss_mlp": 1.02874267,
      "epoch": 0.1358744664221728,
      "flos": 18150797698560.0,
      "grad_norm": 1.8173211864526513,
      "language_loss": 0.78527236,
      "learning_rate": 3.883669688490654e-06,
      "loss": 0.80789453,
      "num_input_tokens_seen": 24081070,
      "step": 1130,
      "time_per_iteration": 2.4780266284942627
    },
    {
      "auxiliary_loss_clip": 0.01198723,
      "auxiliary_loss_mlp": 0.01039518,
      "balance_loss_clip": 1.06210184,
      "balance_loss_mlp": 1.0280205,
      "epoch": 0.13599470931281188,
      "flos": 18442966924800.0,
      "grad_norm": 3.1149684974096497,
      "language_loss": 0.85423452,
      "learning_rate": 3.883407751890256e-06,
      "loss": 0.8766169,
      "num_input_tokens_seen": 24099675,
      "step": 1131,
      "time_per_iteration": 4.077016115188599
    },
    {
      "auxiliary_loss_clip": 0.01212822,
      "auxiliary_loss_mlp": 0.01050029,
      "balance_loss_clip": 1.06557608,
      "balance_loss_mlp": 1.03640962,
      "epoch": 0.13611495220345096,
      "flos": 26680766014080.0,
      "grad_norm": 1.8700927122881257,
      "language_loss": 0.85774148,
      "learning_rate": 3.8831455295783994e-06,
      "loss": 0.88037002,
      "num_input_tokens_seen": 24118925,
      "step": 1132,
      "time_per_iteration": 2.589080810546875
    },
    {
      "auxiliary_loss_clip": 0.01150459,
      "auxiliary_loss_mlp": 0.01046207,
      "balance_loss_clip": 1.0531019,
      "balance_loss_mlp": 1.03051353,
      "epoch": 0.13623519509409007,
      "flos": 21686238673920.0,
      "grad_norm": 1.8271924794271641,
      "language_loss": 0.74064612,
      "learning_rate": 3.882883021594864e-06,
      "loss": 0.76261282,
      "num_input_tokens_seen": 24137065,
      "step": 1133,
      "time_per_iteration": 3.358060121536255
    },
    {
      "auxiliary_loss_clip": 0.01241198,
      "auxiliary_loss_mlp": 0.01041159,
      "balance_loss_clip": 1.07006943,
      "balance_loss_mlp": 1.0281409,
      "epoch": 0.13635543798472916,
      "flos": 14830389492480.0,
      "grad_norm": 1.9570007380898926,
      "language_loss": 0.86363602,
      "learning_rate": 3.8826202279794705e-06,
      "loss": 0.88645965,
      "num_input_tokens_seen": 24154125,
      "step": 1134,
      "time_per_iteration": 3.222118616104126
    },
    {
      "auxiliary_loss_clip": 0.01178203,
      "auxiliary_loss_mlp": 0.01043859,
      "balance_loss_clip": 1.06202233,
      "balance_loss_mlp": 1.03069246,
      "epoch": 0.13647568087536824,
      "flos": 22890323410560.0,
      "grad_norm": 2.164222103075469,
      "language_loss": 0.70665383,
      "learning_rate": 3.882357148772085e-06,
      "loss": 0.72887439,
      "num_input_tokens_seen": 24171550,
      "step": 1135,
      "time_per_iteration": 2.6186587810516357
    },
    {
      "auxiliary_loss_clip": 0.01189469,
      "auxiliary_loss_mlp": 0.01044772,
      "balance_loss_clip": 1.06288695,
      "balance_loss_mlp": 1.03115201,
      "epoch": 0.13659592376600732,
      "flos": 19937927998080.0,
      "grad_norm": 2.2144827172434662,
      "language_loss": 0.84279948,
      "learning_rate": 3.882093784012617e-06,
      "loss": 0.86514187,
      "num_input_tokens_seen": 24190190,
      "step": 1136,
      "time_per_iteration": 2.547165632247925
    },
    {
      "auxiliary_loss_clip": 0.01210232,
      "auxiliary_loss_mlp": 0.01043336,
      "balance_loss_clip": 1.06576347,
      "balance_loss_mlp": 1.02923346,
      "epoch": 0.13671616665664643,
      "flos": 21428579439360.0,
      "grad_norm": 1.6885435145232794,
      "language_loss": 0.84251833,
      "learning_rate": 3.881830133741019e-06,
      "loss": 0.86505401,
      "num_input_tokens_seen": 24209055,
      "step": 1137,
      "time_per_iteration": 2.5523879528045654
    },
    {
      "auxiliary_loss_clip": 0.01223693,
      "auxiliary_loss_mlp": 0.01056994,
      "balance_loss_clip": 1.06631303,
      "balance_loss_mlp": 1.0432198,
      "epoch": 0.13683640954728551,
      "flos": 22778138257920.0,
      "grad_norm": 3.65780011740421,
      "language_loss": 0.76572955,
      "learning_rate": 3.881566197997285e-06,
      "loss": 0.78853643,
      "num_input_tokens_seen": 24225490,
      "step": 1138,
      "time_per_iteration": 2.5015006065368652
    },
    {
      "auxiliary_loss_clip": 0.01220016,
      "auxiliary_loss_mlp": 0.01042446,
      "balance_loss_clip": 1.06653059,
      "balance_loss_mlp": 1.03093052,
      "epoch": 0.1369566524379246,
      "flos": 21725884310400.0,
      "grad_norm": 1.5552426609784487,
      "language_loss": 0.74741185,
      "learning_rate": 3.881301976821456e-06,
      "loss": 0.77003646,
      "num_input_tokens_seen": 24245520,
      "step": 1139,
      "time_per_iteration": 2.5184338092803955
    },
    {
      "auxiliary_loss_clip": 0.01187837,
      "auxiliary_loss_mlp": 0.0103553,
      "balance_loss_clip": 1.0605675,
      "balance_loss_mlp": 1.02279854,
      "epoch": 0.1370768953285637,
      "flos": 18624459369600.0,
      "grad_norm": 1.8102588812779057,
      "language_loss": 0.90468705,
      "learning_rate": 3.881037470253612e-06,
      "loss": 0.92692071,
      "num_input_tokens_seen": 24265035,
      "step": 1140,
      "time_per_iteration": 2.5392277240753174
    },
    {
      "auxiliary_loss_clip": 0.01202902,
      "auxiliary_loss_mlp": 0.01037375,
      "balance_loss_clip": 1.06046093,
      "balance_loss_mlp": 1.02454817,
      "epoch": 0.1371971382192028,
      "flos": 14939521989120.0,
      "grad_norm": 2.7441811171563004,
      "language_loss": 0.79433626,
      "learning_rate": 3.88077267833388e-06,
      "loss": 0.81673908,
      "num_input_tokens_seen": 24281550,
      "step": 1141,
      "time_per_iteration": 2.5218610763549805
    },
    {
      "auxiliary_loss_clip": 0.01167557,
      "auxiliary_loss_mlp": 0.0104059,
      "balance_loss_clip": 1.05817282,
      "balance_loss_mlp": 1.02750099,
      "epoch": 0.13731738110984187,
      "flos": 19023785844480.0,
      "grad_norm": 4.423704524592931,
      "language_loss": 0.84030366,
      "learning_rate": 3.880507601102427e-06,
      "loss": 0.86238509,
      "num_input_tokens_seen": 24299485,
      "step": 1142,
      "time_per_iteration": 2.6125385761260986
    },
    {
      "auxiliary_loss_clip": 0.01173192,
      "auxiliary_loss_mlp": 0.01044411,
      "balance_loss_clip": 1.05916405,
      "balance_loss_mlp": 1.03113079,
      "epoch": 0.13743762400048098,
      "flos": 18187462506240.0,
      "grad_norm": 1.810384334541237,
      "language_loss": 0.82257307,
      "learning_rate": 3.880242238599467e-06,
      "loss": 0.84474903,
      "num_input_tokens_seen": 24316010,
      "step": 1143,
      "time_per_iteration": 2.5737850666046143
    },
    {
      "auxiliary_loss_clip": 0.01156968,
      "auxiliary_loss_mlp": 0.010414,
      "balance_loss_clip": 1.05921662,
      "balance_loss_mlp": 1.02935338,
      "epoch": 0.13755786689112007,
      "flos": 21031982398080.0,
      "grad_norm": 1.6449225935521272,
      "language_loss": 0.83074152,
      "learning_rate": 3.879976590865254e-06,
      "loss": 0.85272521,
      "num_input_tokens_seen": 24335465,
      "step": 1144,
      "time_per_iteration": 2.5708396434783936
    },
    {
      "auxiliary_loss_clip": 0.01139788,
      "auxiliary_loss_mlp": 0.01042624,
      "balance_loss_clip": 1.05176711,
      "balance_loss_mlp": 1.03016686,
      "epoch": 0.13767810978175915,
      "flos": 21360636864000.0,
      "grad_norm": 2.290732796529285,
      "language_loss": 0.87292171,
      "learning_rate": 3.879710657940087e-06,
      "loss": 0.89474583,
      "num_input_tokens_seen": 24354415,
      "step": 1145,
      "time_per_iteration": 2.7038159370422363
    },
    {
      "auxiliary_loss_clip": 0.01222571,
      "auxiliary_loss_mlp": 0.01045588,
      "balance_loss_clip": 1.06629181,
      "balance_loss_mlp": 1.0317595,
      "epoch": 0.13779835267239823,
      "flos": 30592084861440.0,
      "grad_norm": 1.880922414117104,
      "language_loss": 0.70313221,
      "learning_rate": 3.879444439864308e-06,
      "loss": 0.72581381,
      "num_input_tokens_seen": 24373990,
      "step": 1146,
      "time_per_iteration": 2.755586862564087
    },
    {
      "auxiliary_loss_clip": 0.01227883,
      "auxiliary_loss_mlp": 0.0104272,
      "balance_loss_clip": 1.06617761,
      "balance_loss_mlp": 1.02942777,
      "epoch": 0.13791859556303734,
      "flos": 22669867687680.0,
      "grad_norm": 4.176801729979391,
      "language_loss": 0.85969508,
      "learning_rate": 3.879177936678301e-06,
      "loss": 0.88240111,
      "num_input_tokens_seen": 24392995,
      "step": 1147,
      "time_per_iteration": 2.527067184448242
    },
    {
      "auxiliary_loss_clip": 0.01192803,
      "auxiliary_loss_mlp": 0.0076653,
      "balance_loss_clip": 1.0613637,
      "balance_loss_mlp": 1.0003767,
      "epoch": 0.13803883845367643,
      "flos": 35224166016000.0,
      "grad_norm": 1.9796716468379427,
      "language_loss": 0.77300048,
      "learning_rate": 3.878911148422496e-06,
      "loss": 0.79259384,
      "num_input_tokens_seen": 24414470,
      "step": 1148,
      "time_per_iteration": 2.694110631942749
    },
    {
      "auxiliary_loss_clip": 0.01129252,
      "auxiliary_loss_mlp": 0.01035202,
      "balance_loss_clip": 1.04856277,
      "balance_loss_mlp": 1.02102757,
      "epoch": 0.1381590813443155,
      "flos": 32014542332160.0,
      "grad_norm": 3.332715676713788,
      "language_loss": 0.70472813,
      "learning_rate": 3.878644075137364e-06,
      "loss": 0.7263726,
      "num_input_tokens_seen": 24435120,
      "step": 1149,
      "time_per_iteration": 2.786638021469116
    },
    {
      "auxiliary_loss_clip": 0.01156805,
      "auxiliary_loss_mlp": 0.01047427,
      "balance_loss_clip": 1.05218792,
      "balance_loss_mlp": 1.03429651,
      "epoch": 0.13827932423495462,
      "flos": 17821855923840.0,
      "grad_norm": 1.9770739589045583,
      "language_loss": 0.79008323,
      "learning_rate": 3.878376716863418e-06,
      "loss": 0.81212556,
      "num_input_tokens_seen": 24451420,
      "step": 1150,
      "time_per_iteration": 2.5664453506469727
    },
    {
      "auxiliary_loss_clip": 0.01173974,
      "auxiliary_loss_mlp": 0.01037988,
      "balance_loss_clip": 1.06013048,
      "balance_loss_mlp": 1.02464223,
      "epoch": 0.1383995671255937,
      "flos": 19427098728960.0,
      "grad_norm": 2.5732165891036085,
      "language_loss": 0.71853453,
      "learning_rate": 3.878109073641219e-06,
      "loss": 0.74065411,
      "num_input_tokens_seen": 24470450,
      "step": 1151,
      "time_per_iteration": 2.641545295715332
    },
    {
      "auxiliary_loss_clip": 0.01142405,
      "auxiliary_loss_mlp": 0.0104214,
      "balance_loss_clip": 1.0544343,
      "balance_loss_mlp": 1.02965319,
      "epoch": 0.13851981001623279,
      "flos": 28296603331200.0,
      "grad_norm": 1.5629986792165662,
      "language_loss": 0.81431079,
      "learning_rate": 3.877841145511366e-06,
      "loss": 0.83615625,
      "num_input_tokens_seen": 24493190,
      "step": 1152,
      "time_per_iteration": 2.7176029682159424
    },
    {
      "auxiliary_loss_clip": 0.01197593,
      "auxiliary_loss_mlp": 0.0103981,
      "balance_loss_clip": 1.06830692,
      "balance_loss_mlp": 1.02596438,
      "epoch": 0.13864005290687187,
      "flos": 21213079793280.0,
      "grad_norm": 1.796535957972904,
      "language_loss": 0.82924229,
      "learning_rate": 3.8775729325145035e-06,
      "loss": 0.85161638,
      "num_input_tokens_seen": 24512425,
      "step": 1153,
      "time_per_iteration": 2.6115031242370605
    },
    {
      "auxiliary_loss_clip": 0.01075153,
      "auxiliary_loss_mlp": 0.00759454,
      "balance_loss_clip": 1.03535342,
      "balance_loss_mlp": 1.00072038,
      "epoch": 0.13876029579751098,
      "flos": 71653389413760.0,
      "grad_norm": 0.8141235076892157,
      "language_loss": 0.64780509,
      "learning_rate": 3.877304434691321e-06,
      "loss": 0.66615117,
      "num_input_tokens_seen": 24579275,
      "step": 1154,
      "time_per_iteration": 3.3377926349639893
    },
    {
      "auxiliary_loss_clip": 0.01225629,
      "auxiliary_loss_mlp": 0.01045042,
      "balance_loss_clip": 1.06926405,
      "balance_loss_mlp": 1.03301978,
      "epoch": 0.13888053868815006,
      "flos": 21941348042880.0,
      "grad_norm": 1.926808375817095,
      "language_loss": 0.7999872,
      "learning_rate": 3.877035652082548e-06,
      "loss": 0.82269388,
      "num_input_tokens_seen": 24598720,
      "step": 1155,
      "time_per_iteration": 2.5243639945983887
    },
    {
      "auxiliary_loss_clip": 0.01243067,
      "auxiliary_loss_mlp": 0.0104594,
      "balance_loss_clip": 1.07062292,
      "balance_loss_mlp": 1.03203964,
      "epoch": 0.13900078157878915,
      "flos": 19608627087360.0,
      "grad_norm": 1.9108906641418943,
      "language_loss": 0.85567969,
      "learning_rate": 3.87676658472896e-06,
      "loss": 0.87856972,
      "num_input_tokens_seen": 24617530,
      "step": 1156,
      "time_per_iteration": 3.3406221866607666
    },
    {
      "auxiliary_loss_clip": 0.01219875,
      "auxiliary_loss_mlp": 0.01040134,
      "balance_loss_clip": 1.06264973,
      "balance_loss_mlp": 1.02589428,
      "epoch": 0.13912102446942826,
      "flos": 22638051216000.0,
      "grad_norm": 7.507225737900705,
      "language_loss": 0.85107231,
      "learning_rate": 3.876497232671372e-06,
      "loss": 0.87367237,
      "num_input_tokens_seen": 24637485,
      "step": 1157,
      "time_per_iteration": 3.2506749629974365
    },
    {
      "auxiliary_loss_clip": 0.01147359,
      "auxiliary_loss_mlp": 0.01038503,
      "balance_loss_clip": 1.05842507,
      "balance_loss_mlp": 1.02552152,
      "epoch": 0.13924126736006734,
      "flos": 29643324975360.0,
      "grad_norm": 2.309560496658646,
      "language_loss": 0.83832973,
      "learning_rate": 3.876227595950647e-06,
      "loss": 0.86018836,
      "num_input_tokens_seen": 24656915,
      "step": 1158,
      "time_per_iteration": 2.891754627227783
    },
    {
      "auxiliary_loss_clip": 0.01137707,
      "auxiliary_loss_mlp": 0.00766179,
      "balance_loss_clip": 1.05687189,
      "balance_loss_mlp": 1.00032175,
      "epoch": 0.13936151025070642,
      "flos": 27417653527680.0,
      "grad_norm": 1.5973107373168725,
      "language_loss": 0.7922042,
      "learning_rate": 3.875957674607686e-06,
      "loss": 0.81124306,
      "num_input_tokens_seen": 24679190,
      "step": 1159,
      "time_per_iteration": 3.749920129776001
    },
    {
      "auxiliary_loss_clip": 0.01136532,
      "auxiliary_loss_mlp": 0.01040808,
      "balance_loss_clip": 1.04850554,
      "balance_loss_mlp": 1.02711141,
      "epoch": 0.1394817531413455,
      "flos": 16399326625920.0,
      "grad_norm": 2.3152182552246536,
      "language_loss": 0.88196737,
      "learning_rate": 3.8756874686834386e-06,
      "loss": 0.9037407,
      "num_input_tokens_seen": 24697405,
      "step": 1160,
      "time_per_iteration": 3.6124117374420166
    },
    {
      "auxiliary_loss_clip": 0.01239558,
      "auxiliary_loss_mlp": 0.01042428,
      "balance_loss_clip": 1.06814528,
      "balance_loss_mlp": 1.02884376,
      "epoch": 0.13960199603198462,
      "flos": 30922319525760.0,
      "grad_norm": 1.879566109183894,
      "language_loss": 0.80451328,
      "learning_rate": 3.875416978218893e-06,
      "loss": 0.82733309,
      "num_input_tokens_seen": 24720600,
      "step": 1161,
      "time_per_iteration": 2.6342194080352783
    },
    {
      "auxiliary_loss_clip": 0.01202833,
      "auxiliary_loss_mlp": 0.00766578,
      "balance_loss_clip": 1.05931902,
      "balance_loss_mlp": 1.0003258,
      "epoch": 0.1397222389226237,
      "flos": 18113773754880.0,
      "grad_norm": 3.469545914413374,
      "language_loss": 0.82934815,
      "learning_rate": 3.8751462032550835e-06,
      "loss": 0.84904224,
      "num_input_tokens_seen": 24737605,
      "step": 1162,
      "time_per_iteration": 2.6006264686584473
    },
    {
      "auxiliary_loss_clip": 0.01199019,
      "auxiliary_loss_mlp": 0.01041162,
      "balance_loss_clip": 1.05976272,
      "balance_loss_mlp": 1.02781677,
      "epoch": 0.13984248181326278,
      "flos": 16872772815360.0,
      "grad_norm": 2.0298587266445813,
      "language_loss": 0.8310554,
      "learning_rate": 3.874875143833085e-06,
      "loss": 0.85345721,
      "num_input_tokens_seen": 24755845,
      "step": 1163,
      "time_per_iteration": 2.556870698928833
    },
    {
      "auxiliary_loss_clip": 0.01189645,
      "auxiliary_loss_mlp": 0.01038516,
      "balance_loss_clip": 1.05784202,
      "balance_loss_mlp": 1.02396703,
      "epoch": 0.1399627247039019,
      "flos": 54121401267840.0,
      "grad_norm": 2.361611078463927,
      "language_loss": 0.69073391,
      "learning_rate": 3.874603799994019e-06,
      "loss": 0.71301556,
      "num_input_tokens_seen": 24779380,
      "step": 1164,
      "time_per_iteration": 2.8882899284362793
    },
    {
      "auxiliary_loss_clip": 0.01192269,
      "auxiliary_loss_mlp": 0.00765527,
      "balance_loss_clip": 1.06482506,
      "balance_loss_mlp": 1.00026786,
      "epoch": 0.14008296759454097,
      "flos": 11765521618560.0,
      "grad_norm": 2.0345256684444517,
      "language_loss": 0.86933154,
      "learning_rate": 3.874332171779046e-06,
      "loss": 0.88890958,
      "num_input_tokens_seen": 24794260,
      "step": 1165,
      "time_per_iteration": 2.553415536880493
    },
    {
      "auxiliary_loss_clip": 0.01176648,
      "auxiliary_loss_mlp": 0.01040736,
      "balance_loss_clip": 1.05532086,
      "balance_loss_mlp": 1.02782583,
      "epoch": 0.14020321048518006,
      "flos": 22017514832640.0,
      "grad_norm": 1.8761618101435935,
      "language_loss": 0.75919133,
      "learning_rate": 3.874060259229373e-06,
      "loss": 0.78136516,
      "num_input_tokens_seen": 24815835,
      "step": 1166,
      "time_per_iteration": 2.702584743499756
    },
    {
      "auxiliary_loss_clip": 0.01195657,
      "auxiliary_loss_mlp": 0.01042208,
      "balance_loss_clip": 1.06166863,
      "balance_loss_mlp": 1.02781367,
      "epoch": 0.14032345337581917,
      "flos": 23404313076480.0,
      "grad_norm": 2.0357577189408596,
      "language_loss": 0.93438232,
      "learning_rate": 3.873788062386249e-06,
      "loss": 0.95676094,
      "num_input_tokens_seen": 24834095,
      "step": 1167,
      "time_per_iteration": 2.5725719928741455
    },
    {
      "auxiliary_loss_clip": 0.01166524,
      "auxiliary_loss_mlp": 0.01046252,
      "balance_loss_clip": 1.05944467,
      "balance_loss_mlp": 1.03138709,
      "epoch": 0.14044369626645825,
      "flos": 29645767100160.0,
      "grad_norm": 2.070026360942372,
      "language_loss": 0.82091975,
      "learning_rate": 3.873515581290965e-06,
      "loss": 0.8430475,
      "num_input_tokens_seen": 24858900,
      "step": 1168,
      "time_per_iteration": 2.7126848697662354
    },
    {
      "auxiliary_loss_clip": 0.01226234,
      "auxiliary_loss_mlp": 0.01042803,
      "balance_loss_clip": 1.06765723,
      "balance_loss_mlp": 1.0296061,
      "epoch": 0.14056393915709733,
      "flos": 18332972501760.0,
      "grad_norm": 2.6422968408563396,
      "language_loss": 0.75876099,
      "learning_rate": 3.8732428159848575e-06,
      "loss": 0.78145134,
      "num_input_tokens_seen": 24877875,
      "step": 1169,
      "time_per_iteration": 2.5070877075195312
    },
    {
      "auxiliary_loss_clip": 0.0116234,
      "auxiliary_loss_mlp": 0.01044044,
      "balance_loss_clip": 1.05720234,
      "balance_loss_mlp": 1.02957201,
      "epoch": 0.14068418204773642,
      "flos": 26687517770880.0,
      "grad_norm": 3.04528980393777,
      "language_loss": 0.78569031,
      "learning_rate": 3.872969766509304e-06,
      "loss": 0.8077541,
      "num_input_tokens_seen": 24898430,
      "step": 1170,
      "time_per_iteration": 2.72815203666687
    },
    {
      "auxiliary_loss_clip": 0.01077514,
      "auxiliary_loss_mlp": 0.01011362,
      "balance_loss_clip": 1.03972983,
      "balance_loss_mlp": 1.00872755,
      "epoch": 0.14080442493837553,
      "flos": 65259314501760.0,
      "grad_norm": 0.7619756836123063,
      "language_loss": 0.55639899,
      "learning_rate": 3.872696432905726e-06,
      "loss": 0.57728779,
      "num_input_tokens_seen": 24959250,
      "step": 1171,
      "time_per_iteration": 3.2428271770477295
    },
    {
      "auxiliary_loss_clip": 0.01175249,
      "auxiliary_loss_mlp": 0.00766803,
      "balance_loss_clip": 1.05442333,
      "balance_loss_mlp": 1.00026917,
      "epoch": 0.1409246678290146,
      "flos": 25776715582080.0,
      "grad_norm": 3.4381766411050028,
      "language_loss": 0.71724164,
      "learning_rate": 3.872422815215589e-06,
      "loss": 0.73666221,
      "num_input_tokens_seen": 24978330,
      "step": 1172,
      "time_per_iteration": 2.6412277221679688
    },
    {
      "auxiliary_loss_clip": 0.01203849,
      "auxiliary_loss_mlp": 0.0104783,
      "balance_loss_clip": 1.05814004,
      "balance_loss_mlp": 1.03283322,
      "epoch": 0.1410449107196537,
      "flos": 21868521217920.0,
      "grad_norm": 1.8025612171768772,
      "language_loss": 0.74420559,
      "learning_rate": 3.8721489134803994e-06,
      "loss": 0.76672244,
      "num_input_tokens_seen": 24997120,
      "step": 1173,
      "time_per_iteration": 2.5195631980895996
    },
    {
      "auxiliary_loss_clip": 0.01204042,
      "auxiliary_loss_mlp": 0.01036597,
      "balance_loss_clip": 1.06361246,
      "balance_loss_mlp": 1.02266788,
      "epoch": 0.1411651536102928,
      "flos": 16684133564160.0,
      "grad_norm": 6.064773751889712,
      "language_loss": 0.72919381,
      "learning_rate": 3.871874727741707e-06,
      "loss": 0.75160027,
      "num_input_tokens_seen": 25014350,
      "step": 1174,
      "time_per_iteration": 2.5044164657592773
    },
    {
      "auxiliary_loss_clip": 0.01151367,
      "auxiliary_loss_mlp": 0.0104155,
      "balance_loss_clip": 1.05249417,
      "balance_loss_mlp": 1.02979565,
      "epoch": 0.1412853965009319,
      "flos": 20992264934400.0,
      "grad_norm": 1.8518187383246159,
      "language_loss": 0.9655751,
      "learning_rate": 3.871600258041108e-06,
      "loss": 0.98750424,
      "num_input_tokens_seen": 25033875,
      "step": 1175,
      "time_per_iteration": 2.6010029315948486
    },
    {
      "auxiliary_loss_clip": 0.01201871,
      "auxiliary_loss_mlp": 0.01039688,
      "balance_loss_clip": 1.06187487,
      "balance_loss_mlp": 1.02489364,
      "epoch": 0.14140563939157097,
      "flos": 20335279224960.0,
      "grad_norm": 2.246532599933768,
      "language_loss": 0.8585484,
      "learning_rate": 3.871325504420238e-06,
      "loss": 0.88096398,
      "num_input_tokens_seen": 25052865,
      "step": 1176,
      "time_per_iteration": 2.542079448699951
    },
    {
      "auxiliary_loss_clip": 0.01195745,
      "auxiliary_loss_mlp": 0.01041015,
      "balance_loss_clip": 1.06537151,
      "balance_loss_mlp": 1.02790236,
      "epoch": 0.14152588228221005,
      "flos": 21068826773760.0,
      "grad_norm": 2.003739080458816,
      "language_loss": 0.81978488,
      "learning_rate": 3.871050466920776e-06,
      "loss": 0.84215248,
      "num_input_tokens_seen": 25072770,
      "step": 1177,
      "time_per_iteration": 2.5570883750915527
    },
    {
      "auxiliary_loss_clip": 0.01217138,
      "auxiliary_loss_mlp": 0.01045435,
      "balance_loss_clip": 1.06364417,
      "balance_loss_mlp": 1.0322684,
      "epoch": 0.14164612517284916,
      "flos": 18223157646720.0,
      "grad_norm": 1.9536526555043767,
      "language_loss": 0.79836994,
      "learning_rate": 3.870775145584447e-06,
      "loss": 0.82099569,
      "num_input_tokens_seen": 25090550,
      "step": 1178,
      "time_per_iteration": 2.491499900817871
    },
    {
      "auxiliary_loss_clip": 0.01238829,
      "auxiliary_loss_mlp": 0.01053831,
      "balance_loss_clip": 1.06752264,
      "balance_loss_mlp": 1.03716516,
      "epoch": 0.14176636806348825,
      "flos": 22744454279040.0,
      "grad_norm": 4.574011390833995,
      "language_loss": 0.65109557,
      "learning_rate": 3.8704995404530145e-06,
      "loss": 0.6740222,
      "num_input_tokens_seen": 25106175,
      "step": 1179,
      "time_per_iteration": 2.4535443782806396
    },
    {
      "auxiliary_loss_clip": 0.01135919,
      "auxiliary_loss_mlp": 0.01037303,
      "balance_loss_clip": 1.05541897,
      "balance_loss_mlp": 1.02607393,
      "epoch": 0.14188661095412733,
      "flos": 22091095843200.0,
      "grad_norm": 1.6794357159815112,
      "language_loss": 0.85058945,
      "learning_rate": 3.87022365156829e-06,
      "loss": 0.87232167,
      "num_input_tokens_seen": 25126890,
      "step": 1180,
      "time_per_iteration": 2.6518971920013428
    },
    {
      "auxiliary_loss_clip": 0.01233257,
      "auxiliary_loss_mlp": 0.010398,
      "balance_loss_clip": 1.06571627,
      "balance_loss_mlp": 1.02706826,
      "epoch": 0.14200685384476644,
      "flos": 24352390604160.0,
      "grad_norm": 1.883816853591891,
      "language_loss": 0.81280589,
      "learning_rate": 3.869947478972123e-06,
      "loss": 0.83553648,
      "num_input_tokens_seen": 25147915,
      "step": 1181,
      "time_per_iteration": 2.5604188442230225
    },
    {
      "auxiliary_loss_clip": 0.01193727,
      "auxiliary_loss_mlp": 0.01036279,
      "balance_loss_clip": 1.05713832,
      "balance_loss_mlp": 1.02323711,
      "epoch": 0.14212709673540552,
      "flos": 24022048199040.0,
      "grad_norm": 2.336700663070734,
      "language_loss": 0.8254118,
      "learning_rate": 3.869671022706412e-06,
      "loss": 0.84771186,
      "num_input_tokens_seen": 25166645,
      "step": 1182,
      "time_per_iteration": 3.568490982055664
    },
    {
      "auxiliary_loss_clip": 0.01191435,
      "auxiliary_loss_mlp": 0.01042186,
      "balance_loss_clip": 1.06442833,
      "balance_loss_mlp": 1.02968132,
      "epoch": 0.1422473396260446,
      "flos": 26431797870720.0,
      "grad_norm": 1.6474915092111553,
      "language_loss": 0.64711332,
      "learning_rate": 3.869394282813092e-06,
      "loss": 0.66944957,
      "num_input_tokens_seen": 25185845,
      "step": 1183,
      "time_per_iteration": 3.343153476715088
    },
    {
      "auxiliary_loss_clip": 0.01208279,
      "auxiliary_loss_mlp": 0.01045794,
      "balance_loss_clip": 1.06232047,
      "balance_loss_mlp": 1.03113103,
      "epoch": 0.1423675825166837,
      "flos": 17055306754560.0,
      "grad_norm": 3.016744508247988,
      "language_loss": 0.88937533,
      "learning_rate": 3.869117259334147e-06,
      "loss": 0.91191602,
      "num_input_tokens_seen": 25203770,
      "step": 1184,
      "time_per_iteration": 2.515251398086548
    },
    {
      "auxiliary_loss_clip": 0.01147553,
      "auxiliary_loss_mlp": 0.01042148,
      "balance_loss_clip": 1.0539844,
      "balance_loss_mlp": 1.02862978,
      "epoch": 0.1424878254073228,
      "flos": 17929480049280.0,
      "grad_norm": 1.7399838158022731,
      "language_loss": 0.82055211,
      "learning_rate": 3.868839952311599e-06,
      "loss": 0.84244919,
      "num_input_tokens_seen": 25221725,
      "step": 1185,
      "time_per_iteration": 3.3740482330322266
    },
    {
      "auxiliary_loss_clip": 0.01171915,
      "auxiliary_loss_mlp": 0.01042421,
      "balance_loss_clip": 1.05416632,
      "balance_loss_mlp": 1.02875948,
      "epoch": 0.14260806829796188,
      "flos": 20303606407680.0,
      "grad_norm": 2.159629321449262,
      "language_loss": 0.80587924,
      "learning_rate": 3.868562361787516e-06,
      "loss": 0.82802254,
      "num_input_tokens_seen": 25240855,
      "step": 1186,
      "time_per_iteration": 3.2810583114624023
    },
    {
      "auxiliary_loss_clip": 0.01136218,
      "auxiliary_loss_mlp": 0.01042503,
      "balance_loss_clip": 1.05128312,
      "balance_loss_mlp": 1.02901459,
      "epoch": 0.14272831118860096,
      "flos": 23185724860800.0,
      "grad_norm": 1.9504349829792278,
      "language_loss": 0.69123822,
      "learning_rate": 3.868284487804009e-06,
      "loss": 0.71302545,
      "num_input_tokens_seen": 25260085,
      "step": 1187,
      "time_per_iteration": 2.6558120250701904
    },
    {
      "auxiliary_loss_clip": 0.0108955,
      "auxiliary_loss_mlp": 0.01036861,
      "balance_loss_clip": 1.04424465,
      "balance_loss_mlp": 1.02355778,
      "epoch": 0.14284855407924008,
      "flos": 27232210586880.0,
      "grad_norm": 1.6684302968611306,
      "language_loss": 0.78375727,
      "learning_rate": 3.86800633040323e-06,
      "loss": 0.80502141,
      "num_input_tokens_seen": 25280675,
      "step": 1188,
      "time_per_iteration": 2.932239532470703
    },
    {
      "auxiliary_loss_clip": 0.01184617,
      "auxiliary_loss_mlp": 0.00766116,
      "balance_loss_clip": 1.0594058,
      "balance_loss_mlp": 1.00049281,
      "epoch": 0.14296879696987916,
      "flos": 28184202696960.0,
      "grad_norm": 2.1111920840942253,
      "language_loss": 0.78267586,
      "learning_rate": 3.867727889627376e-06,
      "loss": 0.80218315,
      "num_input_tokens_seen": 25300290,
      "step": 1189,
      "time_per_iteration": 3.011594295501709
    },
    {
      "auxiliary_loss_clip": 0.01144882,
      "auxiliary_loss_mlp": 0.01035934,
      "balance_loss_clip": 1.0545876,
      "balance_loss_mlp": 1.02209926,
      "epoch": 0.14308903986051824,
      "flos": 19390290266880.0,
      "grad_norm": 2.325624085026958,
      "language_loss": 0.78624505,
      "learning_rate": 3.867449165518687e-06,
      "loss": 0.80805314,
      "num_input_tokens_seen": 25316760,
      "step": 1190,
      "time_per_iteration": 2.611903667449951
    },
    {
      "auxiliary_loss_clip": 0.01172703,
      "auxiliary_loss_mlp": 0.01041216,
      "balance_loss_clip": 1.05268538,
      "balance_loss_mlp": 1.0268991,
      "epoch": 0.14320928275115732,
      "flos": 17457506317440.0,
      "grad_norm": 1.9554065094226953,
      "language_loss": 0.71376443,
      "learning_rate": 3.867170158119444e-06,
      "loss": 0.73590356,
      "num_input_tokens_seen": 25335760,
      "step": 1191,
      "time_per_iteration": 2.5650558471679688
    },
    {
      "auxiliary_loss_clip": 0.01190912,
      "auxiliary_loss_mlp": 0.00766015,
      "balance_loss_clip": 1.0621922,
      "balance_loss_mlp": 1.0004487,
      "epoch": 0.14332952564179643,
      "flos": 21466070259840.0,
      "grad_norm": 1.9230715178562123,
      "language_loss": 0.75812435,
      "learning_rate": 3.866890867471972e-06,
      "loss": 0.77769363,
      "num_input_tokens_seen": 25354230,
      "step": 1192,
      "time_per_iteration": 2.558349132537842
    },
    {
      "auxiliary_loss_clip": 0.01174993,
      "auxiliary_loss_mlp": 0.00766217,
      "balance_loss_clip": 1.05824494,
      "balance_loss_mlp": 1.00044799,
      "epoch": 0.14344976853243552,
      "flos": 16396992241920.0,
      "grad_norm": 2.580881749426725,
      "language_loss": 0.89656234,
      "learning_rate": 3.86661129361864e-06,
      "loss": 0.91597444,
      "num_input_tokens_seen": 25368720,
      "step": 1193,
      "time_per_iteration": 2.5648348331451416
    },
    {
      "auxiliary_loss_clip": 0.01206511,
      "auxiliary_loss_mlp": 0.01041103,
      "balance_loss_clip": 1.05838609,
      "balance_loss_mlp": 1.02759647,
      "epoch": 0.1435700114230746,
      "flos": 18916736336640.0,
      "grad_norm": 1.8846909375502126,
      "language_loss": 0.86117154,
      "learning_rate": 3.866331436601859e-06,
      "loss": 0.88364768,
      "num_input_tokens_seen": 25386715,
      "step": 1194,
      "time_per_iteration": 2.508535623550415
    },
    {
      "auxiliary_loss_clip": 0.01179784,
      "auxiliary_loss_mlp": 0.01038762,
      "balance_loss_clip": 1.06095147,
      "balance_loss_mlp": 1.02452838,
      "epoch": 0.1436902543137137,
      "flos": 19755394058880.0,
      "grad_norm": 2.842043486878374,
      "language_loss": 0.73965043,
      "learning_rate": 3.866051296464083e-06,
      "loss": 0.76183593,
      "num_input_tokens_seen": 25405550,
      "step": 1195,
      "time_per_iteration": 2.574517011642456
    },
    {
      "auxiliary_loss_clip": 0.01182899,
      "auxiliary_loss_mlp": 0.01042227,
      "balance_loss_clip": 1.05720949,
      "balance_loss_mlp": 1.02852988,
      "epoch": 0.1438104972043528,
      "flos": 14684807669760.0,
      "grad_norm": 2.1703806652286923,
      "language_loss": 0.85449851,
      "learning_rate": 3.86577087324781e-06,
      "loss": 0.87674975,
      "num_input_tokens_seen": 25422040,
      "step": 1196,
      "time_per_iteration": 2.532907247543335
    },
    {
      "auxiliary_loss_clip": 0.01168112,
      "auxiliary_loss_mlp": 0.01037537,
      "balance_loss_clip": 1.05945992,
      "balance_loss_mlp": 1.02535343,
      "epoch": 0.14393074009499188,
      "flos": 17092330698240.0,
      "grad_norm": 2.075404160684189,
      "language_loss": 0.77655566,
      "learning_rate": 3.865490166995578e-06,
      "loss": 0.79861212,
      "num_input_tokens_seen": 25440270,
      "step": 1197,
      "time_per_iteration": 2.6029582023620605
    },
    {
      "auxiliary_loss_clip": 0.01142023,
      "auxiliary_loss_mlp": 0.01044857,
      "balance_loss_clip": 1.051211,
      "balance_loss_mlp": 1.0313623,
      "epoch": 0.144050982985631,
      "flos": 30476200608000.0,
      "grad_norm": 2.6772004149860833,
      "language_loss": 0.84302932,
      "learning_rate": 3.86520917774997e-06,
      "loss": 0.86489809,
      "num_input_tokens_seen": 25459705,
      "step": 1198,
      "time_per_iteration": 2.684657335281372
    },
    {
      "auxiliary_loss_clip": 0.01189374,
      "auxiliary_loss_mlp": 0.0104167,
      "balance_loss_clip": 1.05716443,
      "balance_loss_mlp": 1.02981448,
      "epoch": 0.14417122587627007,
      "flos": 17858484817920.0,
      "grad_norm": 2.7207712318956894,
      "language_loss": 0.75141215,
      "learning_rate": 3.864927905553614e-06,
      "loss": 0.77372265,
      "num_input_tokens_seen": 25477615,
      "step": 1199,
      "time_per_iteration": 2.5745849609375
    },
    {
      "auxiliary_loss_clip": 0.0118278,
      "auxiliary_loss_mlp": 0.01037464,
      "balance_loss_clip": 1.0572089,
      "balance_loss_mlp": 1.02518535,
      "epoch": 0.14429146876690915,
      "flos": 21613914639360.0,
      "grad_norm": 1.7151728641338395,
      "language_loss": 0.88779438,
      "learning_rate": 3.8646463504491765e-06,
      "loss": 0.90999687,
      "num_input_tokens_seen": 25497750,
      "step": 1200,
      "time_per_iteration": 2.5960488319396973
    },
    {
      "auxiliary_loss_clip": 0.01235551,
      "auxiliary_loss_mlp": 0.01038796,
      "balance_loss_clip": 1.06662548,
      "balance_loss_mlp": 1.02597475,
      "epoch": 0.14441171165754824,
      "flos": 23258120722560.0,
      "grad_norm": 1.6935849675692214,
      "language_loss": 0.83272433,
      "learning_rate": 3.8643645124793705e-06,
      "loss": 0.8554678,
      "num_input_tokens_seen": 25516650,
      "step": 1201,
      "time_per_iteration": 2.5196025371551514
    },
    {
      "auxiliary_loss_clip": 0.01202,
      "auxiliary_loss_mlp": 0.01044701,
      "balance_loss_clip": 1.06213367,
      "balance_loss_mlp": 1.03254735,
      "epoch": 0.14453195454818735,
      "flos": 42854213963520.0,
      "grad_norm": 1.5649486540412099,
      "language_loss": 0.74877262,
      "learning_rate": 3.8640823916869515e-06,
      "loss": 0.77123964,
      "num_input_tokens_seen": 25540960,
      "step": 1202,
      "time_per_iteration": 2.732100248336792
    },
    {
      "auxiliary_loss_clip": 0.01167408,
      "auxiliary_loss_mlp": 0.0103893,
      "balance_loss_clip": 1.0547477,
      "balance_loss_mlp": 1.02675319,
      "epoch": 0.14465219743882643,
      "flos": 27235873774080.0,
      "grad_norm": 1.5017680679775627,
      "language_loss": 0.78572738,
      "learning_rate": 3.863799988114714e-06,
      "loss": 0.80779082,
      "num_input_tokens_seen": 25562990,
      "step": 1203,
      "time_per_iteration": 2.6371898651123047
    },
    {
      "auxiliary_loss_clip": 0.0114934,
      "auxiliary_loss_mlp": 0.01044911,
      "balance_loss_clip": 1.05303597,
      "balance_loss_mlp": 1.0314343,
      "epoch": 0.1447724403294655,
      "flos": 16690705752960.0,
      "grad_norm": 2.2413732739982195,
      "language_loss": 0.70762187,
      "learning_rate": 3.863517301805502e-06,
      "loss": 0.72956431,
      "num_input_tokens_seen": 25581380,
      "step": 1204,
      "time_per_iteration": 2.584981918334961
    },
    {
      "auxiliary_loss_clip": 0.01102562,
      "auxiliary_loss_mlp": 0.01043693,
      "balance_loss_clip": 1.05010593,
      "balance_loss_mlp": 1.0289706,
      "epoch": 0.14489268322010462,
      "flos": 20073741321600.0,
      "grad_norm": 2.3416226907929665,
      "language_loss": 0.97202682,
      "learning_rate": 3.863234332802196e-06,
      "loss": 0.99348938,
      "num_input_tokens_seen": 25593585,
      "step": 1205,
      "time_per_iteration": 2.6910288333892822
    },
    {
      "auxiliary_loss_clip": 0.01136818,
      "auxiliary_loss_mlp": 0.01038909,
      "balance_loss_clip": 1.05618763,
      "balance_loss_mlp": 1.02534294,
      "epoch": 0.1450129261107437,
      "flos": 27125627955840.0,
      "grad_norm": 13.748728570849034,
      "language_loss": 0.73887813,
      "learning_rate": 3.862951081147723e-06,
      "loss": 0.76063538,
      "num_input_tokens_seen": 25613750,
      "step": 1206,
      "time_per_iteration": 2.685216188430786
    },
    {
      "auxiliary_loss_clip": 0.01199059,
      "auxiliary_loss_mlp": 0.0103878,
      "balance_loss_clip": 1.06375551,
      "balance_loss_mlp": 1.02722859,
      "epoch": 0.1451331690013828,
      "flos": 25702344472320.0,
      "grad_norm": 2.0478070636812804,
      "language_loss": 0.7837761,
      "learning_rate": 3.862667546885053e-06,
      "loss": 0.80615449,
      "num_input_tokens_seen": 25632300,
      "step": 1207,
      "time_per_iteration": 2.582555055618286
    },
    {
      "auxiliary_loss_clip": 0.0115781,
      "auxiliary_loss_mlp": 0.01044626,
      "balance_loss_clip": 1.05115986,
      "balance_loss_mlp": 1.03164411,
      "epoch": 0.14525341189202187,
      "flos": 25737393168000.0,
      "grad_norm": 1.8771978810073766,
      "language_loss": 0.73961961,
      "learning_rate": 3.8623837300571965e-06,
      "loss": 0.76164389,
      "num_input_tokens_seen": 25651285,
      "step": 1208,
      "time_per_iteration": 3.3560616970062256
    },
    {
      "auxiliary_loss_clip": 0.01182894,
      "auxiliary_loss_mlp": 0.01046379,
      "balance_loss_clip": 1.05603039,
      "balance_loss_mlp": 1.03309953,
      "epoch": 0.14537365478266098,
      "flos": 23073898844160.0,
      "grad_norm": 2.6159816642522276,
      "language_loss": 0.84164232,
      "learning_rate": 3.8620996307072085e-06,
      "loss": 0.86393505,
      "num_input_tokens_seen": 25671990,
      "step": 1209,
      "time_per_iteration": 3.7661333084106445
    },
    {
      "auxiliary_loss_clip": 0.01216198,
      "auxiliary_loss_mlp": 0.0104032,
      "balance_loss_clip": 1.06397521,
      "balance_loss_mlp": 1.02790451,
      "epoch": 0.14549389767330007,
      "flos": 20595021448320.0,
      "grad_norm": 1.7963869031293451,
      "language_loss": 0.64596045,
      "learning_rate": 3.861815248878188e-06,
      "loss": 0.66852564,
      "num_input_tokens_seen": 25689475,
      "step": 1210,
      "time_per_iteration": 2.5200400352478027
    },
    {
      "auxiliary_loss_clip": 0.01140322,
      "auxiliary_loss_mlp": 0.01043125,
      "balance_loss_clip": 1.0591011,
      "balance_loss_mlp": 1.03039956,
      "epoch": 0.14561414056393915,
      "flos": 15121804533120.0,
      "grad_norm": 2.3319716638155183,
      "language_loss": 0.79715085,
      "learning_rate": 3.861530584613274e-06,
      "loss": 0.81898534,
      "num_input_tokens_seen": 25707475,
      "step": 1211,
      "time_per_iteration": 3.5619447231292725
    },
    {
      "auxiliary_loss_clip": 0.01163473,
      "auxiliary_loss_mlp": 0.0076618,
      "balance_loss_clip": 1.05953431,
      "balance_loss_mlp": 1.00058603,
      "epoch": 0.14573438345457826,
      "flos": 19427493778560.0,
      "grad_norm": 2.9964538589932412,
      "language_loss": 0.82394212,
      "learning_rate": 3.86124563795565e-06,
      "loss": 0.84323865,
      "num_input_tokens_seen": 25726290,
      "step": 1212,
      "time_per_iteration": 3.374706745147705
    },
    {
      "auxiliary_loss_clip": 0.01231257,
      "auxiliary_loss_mlp": 0.01042271,
      "balance_loss_clip": 1.06582737,
      "balance_loss_mlp": 1.0304935,
      "epoch": 0.14585462634521734,
      "flos": 24828422572800.0,
      "grad_norm": 1.7095889996159979,
      "language_loss": 0.70211542,
      "learning_rate": 3.860960408948543e-06,
      "loss": 0.72485065,
      "num_input_tokens_seen": 25748040,
      "step": 1213,
      "time_per_iteration": 2.5211620330810547
    },
    {
      "auxiliary_loss_clip": 0.01112577,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.04689705,
      "balance_loss_mlp": 1.02034521,
      "epoch": 0.14597486923585642,
      "flos": 15448627405440.0,
      "grad_norm": 2.299958113206227,
      "language_loss": 0.89669162,
      "learning_rate": 3.860674897635222e-06,
      "loss": 0.91813803,
      "num_input_tokens_seen": 25764525,
      "step": 1214,
      "time_per_iteration": 2.6296000480651855
    },
    {
      "auxiliary_loss_clip": 0.01219358,
      "auxiliary_loss_mlp": 0.01038264,
      "balance_loss_clip": 1.06612277,
      "balance_loss_mlp": 1.02609265,
      "epoch": 0.1460951121264955,
      "flos": 16655154266880.0,
      "grad_norm": 1.7635022034603138,
      "language_loss": 0.83462954,
      "learning_rate": 3.860389104058998e-06,
      "loss": 0.85720569,
      "num_input_tokens_seen": 25782755,
      "step": 1215,
      "time_per_iteration": 2.489319086074829
    },
    {
      "auxiliary_loss_clip": 0.01172496,
      "auxiliary_loss_mlp": 0.01040464,
      "balance_loss_clip": 1.05120492,
      "balance_loss_mlp": 1.02710032,
      "epoch": 0.14621535501713462,
      "flos": 24863291700480.0,
      "grad_norm": 2.1776767020706815,
      "language_loss": 0.72561926,
      "learning_rate": 3.860103028263227e-06,
      "loss": 0.74774885,
      "num_input_tokens_seen": 25805860,
      "step": 1216,
      "time_per_iteration": 2.5980687141418457
    },
    {
      "auxiliary_loss_clip": 0.01160752,
      "auxiliary_loss_mlp": 0.01047096,
      "balance_loss_clip": 1.0544312,
      "balance_loss_mlp": 1.03431082,
      "epoch": 0.1463355979077737,
      "flos": 25228000442880.0,
      "grad_norm": 4.895051825922481,
      "language_loss": 0.70260692,
      "learning_rate": 3.859816670291304e-06,
      "loss": 0.72468537,
      "num_input_tokens_seen": 25824955,
      "step": 1217,
      "time_per_iteration": 2.6237518787384033
    },
    {
      "auxiliary_loss_clip": 0.01169726,
      "auxiliary_loss_mlp": 0.01042299,
      "balance_loss_clip": 1.06231856,
      "balance_loss_mlp": 1.02958512,
      "epoch": 0.14645584079841278,
      "flos": 22054143726720.0,
      "grad_norm": 1.95702419250494,
      "language_loss": 0.8986823,
      "learning_rate": 3.859530030186672e-06,
      "loss": 0.92080253,
      "num_input_tokens_seen": 25841965,
      "step": 1218,
      "time_per_iteration": 2.5873336791992188
    },
    {
      "auxiliary_loss_clip": 0.01210328,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.06092811,
      "balance_loss_mlp": 1.01989579,
      "epoch": 0.1465760836890519,
      "flos": 23623870959360.0,
      "grad_norm": 3.419120758421288,
      "language_loss": 0.83571333,
      "learning_rate": 3.859243107992813e-06,
      "loss": 0.85813951,
      "num_input_tokens_seen": 25860770,
      "step": 1219,
      "time_per_iteration": 2.52917742729187
    },
    {
      "auxiliary_loss_clip": 0.01185521,
      "auxiliary_loss_mlp": 0.01041257,
      "balance_loss_clip": 1.05875206,
      "balance_loss_mlp": 1.0276556,
      "epoch": 0.14669632657969098,
      "flos": 37407893356800.0,
      "grad_norm": 2.4858196094448393,
      "language_loss": 0.78117371,
      "learning_rate": 3.858955903753252e-06,
      "loss": 0.80344152,
      "num_input_tokens_seen": 25879410,
      "step": 1220,
      "time_per_iteration": 2.663583278656006
    },
    {
      "auxiliary_loss_clip": 0.01136342,
      "auxiliary_loss_mlp": 0.01040528,
      "balance_loss_clip": 1.04926109,
      "balance_loss_mlp": 1.02847564,
      "epoch": 0.14681656947033006,
      "flos": 28365910623360.0,
      "grad_norm": 1.4722632108758873,
      "language_loss": 0.83375072,
      "learning_rate": 3.858668417511559e-06,
      "loss": 0.85551935,
      "num_input_tokens_seen": 25902160,
      "step": 1221,
      "time_per_iteration": 2.6864514350891113
    },
    {
      "auxiliary_loss_clip": 0.01204449,
      "auxiliary_loss_mlp": 0.01039054,
      "balance_loss_clip": 1.06796503,
      "balance_loss_mlp": 1.02589965,
      "epoch": 0.14693681236096917,
      "flos": 18479488078080.0,
      "grad_norm": 2.002332643743365,
      "language_loss": 0.76341736,
      "learning_rate": 3.8583806493113445e-06,
      "loss": 0.78585237,
      "num_input_tokens_seen": 25920505,
      "step": 1222,
      "time_per_iteration": 2.5328099727630615
    },
    {
      "auxiliary_loss_clip": 0.01180216,
      "auxiliary_loss_mlp": 0.01039779,
      "balance_loss_clip": 1.05583394,
      "balance_loss_mlp": 1.02716708,
      "epoch": 0.14705705525160825,
      "flos": 20777806782720.0,
      "grad_norm": 2.387484921146313,
      "language_loss": 0.82565999,
      "learning_rate": 3.858092599196263e-06,
      "loss": 0.84785992,
      "num_input_tokens_seen": 25938460,
      "step": 1223,
      "time_per_iteration": 2.5774245262145996
    },
    {
      "auxiliary_loss_clip": 0.01153065,
      "auxiliary_loss_mlp": 0.01031257,
      "balance_loss_clip": 1.05858803,
      "balance_loss_mlp": 1.01897216,
      "epoch": 0.14717729814224734,
      "flos": 29932944336000.0,
      "grad_norm": 2.8154814087432634,
      "language_loss": 0.82262063,
      "learning_rate": 3.857804267210012e-06,
      "loss": 0.84446383,
      "num_input_tokens_seen": 25957760,
      "step": 1224,
      "time_per_iteration": 2.682328462600708
    },
    {
      "auxiliary_loss_clip": 0.01191951,
      "auxiliary_loss_mlp": 0.0103704,
      "balance_loss_clip": 1.05720377,
      "balance_loss_mlp": 1.02600145,
      "epoch": 0.14729754103288642,
      "flos": 20047491457920.0,
      "grad_norm": 2.4193163898642975,
      "language_loss": 0.88492209,
      "learning_rate": 3.857515653396331e-06,
      "loss": 0.90721202,
      "num_input_tokens_seen": 25974970,
      "step": 1225,
      "time_per_iteration": 2.5457067489624023
    },
    {
      "auxiliary_loss_clip": 0.01155803,
      "auxiliary_loss_mlp": 0.01035762,
      "balance_loss_clip": 1.05286837,
      "balance_loss_mlp": 1.02339387,
      "epoch": 0.14741778392352553,
      "flos": 19281516906240.0,
      "grad_norm": 2.181412929299813,
      "language_loss": 0.86753857,
      "learning_rate": 3.857226757799002e-06,
      "loss": 0.88945425,
      "num_input_tokens_seen": 25992525,
      "step": 1226,
      "time_per_iteration": 2.6218724250793457
    },
    {
      "auxiliary_loss_clip": 0.01195316,
      "auxiliary_loss_mlp": 0.01039914,
      "balance_loss_clip": 1.05627799,
      "balance_loss_mlp": 1.02776647,
      "epoch": 0.1475380268141646,
      "flos": 25411108999680.0,
      "grad_norm": 2.2191092313286958,
      "language_loss": 0.73413616,
      "learning_rate": 3.85693758046185e-06,
      "loss": 0.75648844,
      "num_input_tokens_seen": 26010815,
      "step": 1227,
      "time_per_iteration": 2.5692520141601562
    },
    {
      "auxiliary_loss_clip": 0.01208062,
      "auxiliary_loss_mlp": 0.0103888,
      "balance_loss_clip": 1.06446862,
      "balance_loss_mlp": 1.02616048,
      "epoch": 0.1476582697048037,
      "flos": 20847652778880.0,
      "grad_norm": 1.75938264093044,
      "language_loss": 0.82712317,
      "learning_rate": 3.8566481214287435e-06,
      "loss": 0.84959263,
      "num_input_tokens_seen": 26028935,
      "step": 1228,
      "time_per_iteration": 2.5443978309631348
    },
    {
      "auxiliary_loss_clip": 0.01165833,
      "auxiliary_loss_mlp": 0.00766138,
      "balance_loss_clip": 1.05703008,
      "balance_loss_mlp": 1.00061393,
      "epoch": 0.1477785125954428,
      "flos": 14028109269120.0,
      "grad_norm": 2.1573072684522443,
      "language_loss": 0.91113639,
      "learning_rate": 3.8563583807435935e-06,
      "loss": 0.93045616,
      "num_input_tokens_seen": 26045080,
      "step": 1229,
      "time_per_iteration": 2.6006619930267334
    },
    {
      "auxiliary_loss_clip": 0.01138786,
      "auxiliary_loss_mlp": 0.01041686,
      "balance_loss_clip": 1.04926288,
      "balance_loss_mlp": 1.02801871,
      "epoch": 0.1478987554860819,
      "flos": 20516699842560.0,
      "grad_norm": 2.1814500498491034,
      "language_loss": 0.77930522,
      "learning_rate": 3.856068358450353e-06,
      "loss": 0.80110991,
      "num_input_tokens_seen": 26065030,
      "step": 1230,
      "time_per_iteration": 2.619044542312622
    },
    {
      "auxiliary_loss_clip": 0.01204681,
      "auxiliary_loss_mlp": 0.01045745,
      "balance_loss_clip": 1.06465173,
      "balance_loss_mlp": 1.03416979,
      "epoch": 0.14801899837672097,
      "flos": 17857012360320.0,
      "grad_norm": 1.85027189632473,
      "language_loss": 0.86109668,
      "learning_rate": 3.8557780545930186e-06,
      "loss": 0.88360095,
      "num_input_tokens_seen": 26083445,
      "step": 1231,
      "time_per_iteration": 2.529186248779297
    },
    {
      "auxiliary_loss_clip": 0.01230977,
      "auxiliary_loss_mlp": 0.00765217,
      "balance_loss_clip": 1.06567872,
      "balance_loss_mlp": 1.00058472,
      "epoch": 0.14813924126736006,
      "flos": 20881408584960.0,
      "grad_norm": 1.7615201476735027,
      "language_loss": 0.79454911,
      "learning_rate": 3.855487469215628e-06,
      "loss": 0.81451106,
      "num_input_tokens_seen": 26102375,
      "step": 1232,
      "time_per_iteration": 2.462390661239624
    },
    {
      "auxiliary_loss_clip": 0.01150169,
      "auxiliary_loss_mlp": 0.01035662,
      "balance_loss_clip": 1.05509627,
      "balance_loss_mlp": 1.02290034,
      "epoch": 0.14825948415799917,
      "flos": 37414070496000.0,
      "grad_norm": 1.9826690681117316,
      "language_loss": 0.72429794,
      "learning_rate": 3.855196602362264e-06,
      "loss": 0.74615622,
      "num_input_tokens_seen": 26125295,
      "step": 1233,
      "time_per_iteration": 3.4920084476470947
    },
    {
      "auxiliary_loss_clip": 0.0114935,
      "auxiliary_loss_mlp": 0.0076595,
      "balance_loss_clip": 1.05500364,
      "balance_loss_mlp": 1.00056911,
      "epoch": 0.14837972704863825,
      "flos": 22014641744640.0,
      "grad_norm": 1.98367794906282,
      "language_loss": 0.94381046,
      "learning_rate": 3.854905454077051e-06,
      "loss": 0.9629634,
      "num_input_tokens_seen": 26142905,
      "step": 1234,
      "time_per_iteration": 3.3635573387145996
    },
    {
      "auxiliary_loss_clip": 0.01151392,
      "auxiliary_loss_mlp": 0.01037792,
      "balance_loss_clip": 1.05321407,
      "balance_loss_mlp": 1.02639008,
      "epoch": 0.14849996993927733,
      "flos": 20996323171200.0,
      "grad_norm": 2.117028459758764,
      "language_loss": 0.88468385,
      "learning_rate": 3.854614024404155e-06,
      "loss": 0.90657568,
      "num_input_tokens_seen": 26161215,
      "step": 1235,
      "time_per_iteration": 2.616180419921875
    },
    {
      "auxiliary_loss_clip": 0.01118475,
      "auxiliary_loss_mlp": 0.01035242,
      "balance_loss_clip": 1.03989089,
      "balance_loss_mlp": 1.02345192,
      "epoch": 0.14862021282991644,
      "flos": 20047994248320.0,
      "grad_norm": 1.8163003609175712,
      "language_loss": 0.89255691,
      "learning_rate": 3.8543223133877865e-06,
      "loss": 0.91409403,
      "num_input_tokens_seen": 26179810,
      "step": 1236,
      "time_per_iteration": 2.6517527103424072
    },
    {
      "auxiliary_loss_clip": 0.01175301,
      "auxiliary_loss_mlp": 0.00765894,
      "balance_loss_clip": 1.05842924,
      "balance_loss_mlp": 1.00054502,
      "epoch": 0.14874045572055553,
      "flos": 22712027276160.0,
      "grad_norm": 2.6962980505438945,
      "language_loss": 0.88295197,
      "learning_rate": 3.854030321072198e-06,
      "loss": 0.90236396,
      "num_input_tokens_seen": 26199715,
      "step": 1237,
      "time_per_iteration": 4.137237071990967
    },
    {
      "auxiliary_loss_clip": 0.01212886,
      "auxiliary_loss_mlp": 0.01047869,
      "balance_loss_clip": 1.06326962,
      "balance_loss_mlp": 1.03461266,
      "epoch": 0.1488606986111946,
      "flos": 25411288567680.0,
      "grad_norm": 2.037672402913096,
      "language_loss": 0.73281658,
      "learning_rate": 3.853738047501682e-06,
      "loss": 0.75542414,
      "num_input_tokens_seen": 26220275,
      "step": 1238,
      "time_per_iteration": 2.5271246433258057
    },
    {
      "auxiliary_loss_clip": 0.01151671,
      "auxiliary_loss_mlp": 0.01039842,
      "balance_loss_clip": 1.05325294,
      "balance_loss_mlp": 1.02815914,
      "epoch": 0.1489809415018337,
      "flos": 17018749687680.0,
      "grad_norm": 2.2401881773748173,
      "language_loss": 0.77943146,
      "learning_rate": 3.85344549272058e-06,
      "loss": 0.8013466,
      "num_input_tokens_seen": 26238255,
      "step": 1239,
      "time_per_iteration": 2.606011152267456
    },
    {
      "auxiliary_loss_clip": 0.01201359,
      "auxiliary_loss_mlp": 0.01040248,
      "balance_loss_clip": 1.06092358,
      "balance_loss_mlp": 1.02850544,
      "epoch": 0.1491011843924728,
      "flos": 33659394860160.0,
      "grad_norm": 1.8993942732112328,
      "language_loss": 0.82600379,
      "learning_rate": 3.853152656773269e-06,
      "loss": 0.84841985,
      "num_input_tokens_seen": 26259690,
      "step": 1240,
      "time_per_iteration": 2.652174949645996
    },
    {
      "auxiliary_loss_clip": 0.01185561,
      "auxiliary_loss_mlp": 0.01034712,
      "balance_loss_clip": 1.05641544,
      "balance_loss_mlp": 1.02288103,
      "epoch": 0.14922142728311188,
      "flos": 21179000764800.0,
      "grad_norm": 1.7094716606644225,
      "language_loss": 0.8489356,
      "learning_rate": 3.852859539704174e-06,
      "loss": 0.87113833,
      "num_input_tokens_seen": 26278990,
      "step": 1241,
      "time_per_iteration": 2.5704739093780518
    },
    {
      "auxiliary_loss_clip": 0.01166077,
      "auxiliary_loss_mlp": 0.01043038,
      "balance_loss_clip": 1.05757439,
      "balance_loss_mlp": 1.03114128,
      "epoch": 0.14934167017375097,
      "flos": 29860548474240.0,
      "grad_norm": 2.0732476037098175,
      "language_loss": 0.76238328,
      "learning_rate": 3.85256614155776e-06,
      "loss": 0.78447443,
      "num_input_tokens_seen": 26299120,
      "step": 1242,
      "time_per_iteration": 2.651562213897705
    },
    {
      "auxiliary_loss_clip": 0.01152912,
      "auxiliary_loss_mlp": 0.01038676,
      "balance_loss_clip": 1.05235624,
      "balance_loss_mlp": 1.02644551,
      "epoch": 0.14946191306439008,
      "flos": 17019216564480.0,
      "grad_norm": 2.3070204712816036,
      "language_loss": 0.74765396,
      "learning_rate": 3.852272462378535e-06,
      "loss": 0.76956981,
      "num_input_tokens_seen": 26316995,
      "step": 1243,
      "time_per_iteration": 2.579768180847168
    },
    {
      "auxiliary_loss_clip": 0.01192772,
      "auxiliary_loss_mlp": 0.00765188,
      "balance_loss_clip": 1.05937886,
      "balance_loss_mlp": 1.00043249,
      "epoch": 0.14958215595502916,
      "flos": 15669047214720.0,
      "grad_norm": 1.9176564623818597,
      "language_loss": 0.77598059,
      "learning_rate": 3.85197850221105e-06,
      "loss": 0.79556018,
      "num_input_tokens_seen": 26333295,
      "step": 1244,
      "time_per_iteration": 2.5212345123291016
    },
    {
      "auxiliary_loss_clip": 0.01214719,
      "auxiliary_loss_mlp": 0.01032887,
      "balance_loss_clip": 1.06568182,
      "balance_loss_mlp": 1.02198505,
      "epoch": 0.14970239884566824,
      "flos": 33108560818560.0,
      "grad_norm": 1.748287450097762,
      "language_loss": 0.75859374,
      "learning_rate": 3.851684261099899e-06,
      "loss": 0.78106976,
      "num_input_tokens_seen": 26355035,
      "step": 1245,
      "time_per_iteration": 2.606894016265869
    },
    {
      "auxiliary_loss_clip": 0.01201747,
      "auxiliary_loss_mlp": 0.01046063,
      "balance_loss_clip": 1.06391549,
      "balance_loss_mlp": 1.03438091,
      "epoch": 0.14982264173630733,
      "flos": 17821245392640.0,
      "grad_norm": 2.267507542820709,
      "language_loss": 0.86646634,
      "learning_rate": 3.851389739089718e-06,
      "loss": 0.88894451,
      "num_input_tokens_seen": 26371655,
      "step": 1246,
      "time_per_iteration": 2.4898476600646973
    },
    {
      "auxiliary_loss_clip": 0.01177716,
      "auxiliary_loss_mlp": 0.01045865,
      "balance_loss_clip": 1.06027627,
      "balance_loss_mlp": 1.03329468,
      "epoch": 0.14994288462694644,
      "flos": 32409559175040.0,
      "grad_norm": 1.7803423101486588,
      "language_loss": 0.80454499,
      "learning_rate": 3.851094936225186e-06,
      "loss": 0.8267808,
      "num_input_tokens_seen": 26392540,
      "step": 1247,
      "time_per_iteration": 2.7106359004974365
    },
    {
      "auxiliary_loss_clip": 0.01180386,
      "auxiliary_loss_mlp": 0.01038004,
      "balance_loss_clip": 1.05977726,
      "balance_loss_mlp": 1.02539182,
      "epoch": 0.15006312751758552,
      "flos": 31794661226880.0,
      "grad_norm": 1.4203630084819525,
      "language_loss": 0.76862651,
      "learning_rate": 3.850799852551024e-06,
      "loss": 0.79081047,
      "num_input_tokens_seen": 26414960,
      "step": 1248,
      "time_per_iteration": 2.6961796283721924
    },
    {
      "auxiliary_loss_clip": 0.01129181,
      "auxiliary_loss_mlp": 0.01033827,
      "balance_loss_clip": 1.05494976,
      "balance_loss_mlp": 1.02223432,
      "epoch": 0.1501833704082246,
      "flos": 16618022582400.0,
      "grad_norm": 4.151710838284924,
      "language_loss": 0.86079359,
      "learning_rate": 3.850504488111995e-06,
      "loss": 0.88242376,
      "num_input_tokens_seen": 26431635,
      "step": 1249,
      "time_per_iteration": 2.667489528656006
    },
    {
      "auxiliary_loss_clip": 0.01167162,
      "auxiliary_loss_mlp": 0.01040822,
      "balance_loss_clip": 1.04795146,
      "balance_loss_mlp": 1.02995598,
      "epoch": 0.15030361329886371,
      "flos": 23471178243840.0,
      "grad_norm": 1.9260552110392664,
      "language_loss": 0.8238498,
      "learning_rate": 3.850208842952907e-06,
      "loss": 0.84592962,
      "num_input_tokens_seen": 26450440,
      "step": 1250,
      "time_per_iteration": 2.6233274936676025
    },
    {
      "auxiliary_loss_clip": 0.01169765,
      "auxiliary_loss_mlp": 0.010351,
      "balance_loss_clip": 1.05690551,
      "balance_loss_mlp": 1.02387631,
      "epoch": 0.1504238561895028,
      "flos": 25629409906560.0,
      "grad_norm": 2.9620957233803242,
      "language_loss": 0.79373133,
      "learning_rate": 3.849912917118608e-06,
      "loss": 0.81577992,
      "num_input_tokens_seen": 26471480,
      "step": 1251,
      "time_per_iteration": 2.5981974601745605
    },
    {
      "auxiliary_loss_clip": 0.01091131,
      "auxiliary_loss_mlp": 0.01004976,
      "balance_loss_clip": 1.04781818,
      "balance_loss_mlp": 1.00249696,
      "epoch": 0.15054409908014188,
      "flos": 52095146129280.0,
      "grad_norm": 0.8795314962860479,
      "language_loss": 0.59294051,
      "learning_rate": 3.849616710653992e-06,
      "loss": 0.61390156,
      "num_input_tokens_seen": 26532950,
      "step": 1252,
      "time_per_iteration": 3.20995831489563
    },
    {
      "auxiliary_loss_clip": 0.01199851,
      "auxiliary_loss_mlp": 0.01039221,
      "balance_loss_clip": 1.05790091,
      "balance_loss_mlp": 1.02747929,
      "epoch": 0.150664341970781,
      "flos": 18880251096960.0,
      "grad_norm": 1.7254170761647545,
      "language_loss": 0.75048172,
      "learning_rate": 3.84932022360399e-06,
      "loss": 0.77287245,
      "num_input_tokens_seen": 26551615,
      "step": 1253,
      "time_per_iteration": 2.5108249187469482
    },
    {
      "auxiliary_loss_clip": 0.01207982,
      "auxiliary_loss_mlp": 0.01037563,
      "balance_loss_clip": 1.06315756,
      "balance_loss_mlp": 1.02517128,
      "epoch": 0.15078458486142007,
      "flos": 22163240309760.0,
      "grad_norm": 2.6856193902472127,
      "language_loss": 0.85189241,
      "learning_rate": 3.849023456013581e-06,
      "loss": 0.87434781,
      "num_input_tokens_seen": 26569175,
      "step": 1254,
      "time_per_iteration": 2.5216176509857178
    },
    {
      "auxiliary_loss_clip": 0.01201319,
      "auxiliary_loss_mlp": 0.01041985,
      "balance_loss_clip": 1.06424618,
      "balance_loss_mlp": 1.02976608,
      "epoch": 0.15090482775205916,
      "flos": 26651894457600.0,
      "grad_norm": 2.459171263258759,
      "language_loss": 0.62595636,
      "learning_rate": 3.848726407927784e-06,
      "loss": 0.64838934,
      "num_input_tokens_seen": 26589560,
      "step": 1255,
      "time_per_iteration": 2.566155433654785
    },
    {
      "auxiliary_loss_clip": 0.01172753,
      "auxiliary_loss_mlp": 0.01043622,
      "balance_loss_clip": 1.06525993,
      "balance_loss_mlp": 1.03136754,
      "epoch": 0.15102507064269824,
      "flos": 21798998444160.0,
      "grad_norm": 2.402670860736415,
      "language_loss": 0.86866748,
      "learning_rate": 3.84842907939166e-06,
      "loss": 0.89083123,
      "num_input_tokens_seen": 26608785,
      "step": 1256,
      "time_per_iteration": 2.6117804050445557
    },
    {
      "auxiliary_loss_clip": 0.01157218,
      "auxiliary_loss_mlp": 0.01039779,
      "balance_loss_clip": 1.05443263,
      "balance_loss_mlp": 1.02719069,
      "epoch": 0.15114531353333735,
      "flos": 22820908377600.0,
      "grad_norm": 3.6369549155096936,
      "language_loss": 0.70997012,
      "learning_rate": 3.8481314704503146e-06,
      "loss": 0.73194015,
      "num_input_tokens_seen": 26628615,
      "step": 1257,
      "time_per_iteration": 2.5766875743865967
    },
    {
      "auxiliary_loss_clip": 0.0117332,
      "auxiliary_loss_mlp": 0.01047652,
      "balance_loss_clip": 1.05900145,
      "balance_loss_mlp": 1.03474164,
      "epoch": 0.15126555642397643,
      "flos": 19682674974720.0,
      "grad_norm": 2.1529722010765293,
      "language_loss": 0.88011694,
      "learning_rate": 3.847833581148895e-06,
      "loss": 0.90232658,
      "num_input_tokens_seen": 26647525,
      "step": 1258,
      "time_per_iteration": 2.5721263885498047
    },
    {
      "auxiliary_loss_clip": 0.01193969,
      "auxiliary_loss_mlp": 0.01039281,
      "balance_loss_clip": 1.05527759,
      "balance_loss_mlp": 1.02681136,
      "epoch": 0.15138579931461552,
      "flos": 28726022424960.0,
      "grad_norm": 2.436394635997048,
      "language_loss": 0.81215739,
      "learning_rate": 3.84753541153259e-06,
      "loss": 0.83448988,
      "num_input_tokens_seen": 26667095,
      "step": 1259,
      "time_per_iteration": 2.6030666828155518
    },
    {
      "auxiliary_loss_clip": 0.01184264,
      "auxiliary_loss_mlp": 0.00765137,
      "balance_loss_clip": 1.06094563,
      "balance_loss_mlp": 1.00040853,
      "epoch": 0.15150604220525463,
      "flos": 22127006465280.0,
      "grad_norm": 2.892550591918936,
      "language_loss": 0.83188319,
      "learning_rate": 3.847236961646633e-06,
      "loss": 0.85137725,
      "num_input_tokens_seen": 26686075,
      "step": 1260,
      "time_per_iteration": 4.054657936096191
    },
    {
      "auxiliary_loss_clip": 0.01200677,
      "auxiliary_loss_mlp": 0.01038786,
      "balance_loss_clip": 1.06579185,
      "balance_loss_mlp": 1.02542305,
      "epoch": 0.1516262850958937,
      "flos": 12968708515200.0,
      "grad_norm": 4.653117715132465,
      "language_loss": 0.78138554,
      "learning_rate": 3.846938231536296e-06,
      "loss": 0.8037802,
      "num_input_tokens_seen": 26701695,
      "step": 1261,
      "time_per_iteration": 2.5278093814849854
    },
    {
      "auxiliary_loss_clip": 0.01171214,
      "auxiliary_loss_mlp": 0.01041628,
      "balance_loss_clip": 1.06246746,
      "balance_loss_mlp": 1.02951002,
      "epoch": 0.1517465279865328,
      "flos": 21797130936960.0,
      "grad_norm": 1.8361141548305449,
      "language_loss": 0.81059468,
      "learning_rate": 3.8466392212468995e-06,
      "loss": 0.83272314,
      "num_input_tokens_seen": 26721885,
      "step": 1262,
      "time_per_iteration": 3.4040491580963135
    },
    {
      "auxiliary_loss_clip": 0.01108224,
      "auxiliary_loss_mlp": 0.01002097,
      "balance_loss_clip": 1.04832935,
      "balance_loss_mlp": 0.99983186,
      "epoch": 0.15186677087717187,
      "flos": 58174569901440.0,
      "grad_norm": 0.8244076343411318,
      "language_loss": 0.61944175,
      "learning_rate": 3.8463399308238e-06,
      "loss": 0.64054495,
      "num_input_tokens_seen": 26780990,
      "step": 1263,
      "time_per_iteration": 3.873725414276123
    },
    {
      "auxiliary_loss_clip": 0.01161799,
      "auxiliary_loss_mlp": 0.01035157,
      "balance_loss_clip": 1.05718803,
      "balance_loss_mlp": 1.02222908,
      "epoch": 0.15198701376781099,
      "flos": 32669696448000.0,
      "grad_norm": 3.4316884610500384,
      "language_loss": 0.63923472,
      "learning_rate": 3.846040360312402e-06,
      "loss": 0.66120428,
      "num_input_tokens_seen": 26804250,
      "step": 1264,
      "time_per_iteration": 2.7008779048919678
    },
    {
      "auxiliary_loss_clip": 0.01168045,
      "auxiliary_loss_mlp": 0.01036995,
      "balance_loss_clip": 1.05752838,
      "balance_loss_mlp": 1.02423418,
      "epoch": 0.15210725665845007,
      "flos": 28402575431040.0,
      "grad_norm": 2.5207912892297433,
      "language_loss": 0.81453747,
      "learning_rate": 3.8457405097581485e-06,
      "loss": 0.83658785,
      "num_input_tokens_seen": 26823240,
      "step": 1265,
      "time_per_iteration": 2.64896297454834
    },
    {
      "auxiliary_loss_clip": 0.01125589,
      "auxiliary_loss_mlp": 0.01039471,
      "balance_loss_clip": 1.04709423,
      "balance_loss_mlp": 1.02709746,
      "epoch": 0.15222749954908915,
      "flos": 19938179393280.0,
      "grad_norm": 1.7343541760833403,
      "language_loss": 0.78127253,
      "learning_rate": 3.8454403792065275e-06,
      "loss": 0.80292314,
      "num_input_tokens_seen": 26842060,
      "step": 1266,
      "time_per_iteration": 2.6981053352355957
    },
    {
      "auxiliary_loss_clip": 0.01158585,
      "auxiliary_loss_mlp": 0.01037223,
      "balance_loss_clip": 1.06496012,
      "balance_loss_mlp": 1.02602375,
      "epoch": 0.15234774243972826,
      "flos": 21324223451520.0,
      "grad_norm": 2.052062651077573,
      "language_loss": 0.85320014,
      "learning_rate": 3.845139968703068e-06,
      "loss": 0.87515819,
      "num_input_tokens_seen": 26859580,
      "step": 1267,
      "time_per_iteration": 2.6258511543273926
    },
    {
      "auxiliary_loss_clip": 0.0117685,
      "auxiliary_loss_mlp": 0.01040548,
      "balance_loss_clip": 1.05696344,
      "balance_loss_mlp": 1.02790594,
      "epoch": 0.15246798533036734,
      "flos": 25957812977280.0,
      "grad_norm": 1.7960763843068963,
      "language_loss": 0.82836896,
      "learning_rate": 3.844839278293342e-06,
      "loss": 0.8505429,
      "num_input_tokens_seen": 26880430,
      "step": 1268,
      "time_per_iteration": 2.6070384979248047
    },
    {
      "auxiliary_loss_clip": 0.01221469,
      "auxiliary_loss_mlp": 0.00766572,
      "balance_loss_clip": 1.06873322,
      "balance_loss_mlp": 1.00047302,
      "epoch": 0.15258822822100643,
      "flos": 25811907932160.0,
      "grad_norm": 2.297965074217342,
      "language_loss": 0.7723164,
      "learning_rate": 3.8445383080229654e-06,
      "loss": 0.79219675,
      "num_input_tokens_seen": 26896445,
      "step": 1269,
      "time_per_iteration": 2.5311806201934814
    },
    {
      "auxiliary_loss_clip": 0.01165981,
      "auxiliary_loss_mlp": 0.01038818,
      "balance_loss_clip": 1.05531502,
      "balance_loss_mlp": 1.02513289,
      "epoch": 0.1527084711116455,
      "flos": 25265455349760.0,
      "grad_norm": 2.91387380067459,
      "language_loss": 0.73879302,
      "learning_rate": 3.844237057937593e-06,
      "loss": 0.76084107,
      "num_input_tokens_seen": 26915450,
      "step": 1270,
      "time_per_iteration": 2.6426408290863037
    },
    {
      "auxiliary_loss_clip": 0.01203411,
      "auxiliary_loss_mlp": 0.01047883,
      "balance_loss_clip": 1.06342602,
      "balance_loss_mlp": 1.03581333,
      "epoch": 0.15282871400228462,
      "flos": 29240227572480.0,
      "grad_norm": 2.424609404533091,
      "language_loss": 0.77688944,
      "learning_rate": 3.843935528082926e-06,
      "loss": 0.79940236,
      "num_input_tokens_seen": 26936475,
      "step": 1271,
      "time_per_iteration": 2.5975914001464844
    },
    {
      "auxiliary_loss_clip": 0.01197543,
      "auxiliary_loss_mlp": 0.01042538,
      "balance_loss_clip": 1.06188178,
      "balance_loss_mlp": 1.02882886,
      "epoch": 0.1529489568929237,
      "flos": 20882952869760.0,
      "grad_norm": 1.9671632988394963,
      "language_loss": 0.84916353,
      "learning_rate": 3.843633718504704e-06,
      "loss": 0.87156439,
      "num_input_tokens_seen": 26954920,
      "step": 1272,
      "time_per_iteration": 2.555607795715332
    },
    {
      "auxiliary_loss_clip": 0.01188783,
      "auxiliary_loss_mlp": 0.0103842,
      "balance_loss_clip": 1.06358492,
      "balance_loss_mlp": 1.02635622,
      "epoch": 0.1530691997835628,
      "flos": 20083833043200.0,
      "grad_norm": 2.920111366374998,
      "language_loss": 0.90300441,
      "learning_rate": 3.843331629248715e-06,
      "loss": 0.92527634,
      "num_input_tokens_seen": 26972520,
      "step": 1273,
      "time_per_iteration": 2.555454969406128
    },
    {
      "auxiliary_loss_clip": 0.01137448,
      "auxiliary_loss_mlp": 0.01038815,
      "balance_loss_clip": 1.05478406,
      "balance_loss_mlp": 1.02746606,
      "epoch": 0.1531894426742019,
      "flos": 28759814144640.0,
      "grad_norm": 2.0561039917501955,
      "language_loss": 0.7653147,
      "learning_rate": 3.843029260360782e-06,
      "loss": 0.78707731,
      "num_input_tokens_seen": 26990890,
      "step": 1274,
      "time_per_iteration": 2.7061493396759033
    },
    {
      "auxiliary_loss_clip": 0.01194962,
      "auxiliary_loss_mlp": 0.01040017,
      "balance_loss_clip": 1.06047082,
      "balance_loss_mlp": 1.02681506,
      "epoch": 0.15330968556484098,
      "flos": 22236282616320.0,
      "grad_norm": 1.7184109598286452,
      "language_loss": 0.78494525,
      "learning_rate": 3.8427266118867755e-06,
      "loss": 0.80729508,
      "num_input_tokens_seen": 27010640,
      "step": 1275,
      "time_per_iteration": 2.5448474884033203
    },
    {
      "auxiliary_loss_clip": 0.01182437,
      "auxiliary_loss_mlp": 0.01038453,
      "balance_loss_clip": 1.05903912,
      "balance_loss_mlp": 1.02622855,
      "epoch": 0.15342992845548006,
      "flos": 27527504296320.0,
      "grad_norm": 1.918266000527644,
      "language_loss": 0.82724404,
      "learning_rate": 3.842423683872608e-06,
      "loss": 0.84945297,
      "num_input_tokens_seen": 27031215,
      "step": 1276,
      "time_per_iteration": 2.611635684967041
    },
    {
      "auxiliary_loss_clip": 0.01189562,
      "auxiliary_loss_mlp": 0.01046514,
      "balance_loss_clip": 1.05996871,
      "balance_loss_mlp": 1.03293049,
      "epoch": 0.15355017134611917,
      "flos": 19609596754560.0,
      "grad_norm": 2.4454760650779086,
      "language_loss": 0.77883464,
      "learning_rate": 3.842120476364232e-06,
      "loss": 0.80119538,
      "num_input_tokens_seen": 27049665,
      "step": 1277,
      "time_per_iteration": 2.5336766242980957
    },
    {
      "auxiliary_loss_clip": 0.01188215,
      "auxiliary_loss_mlp": 0.01039036,
      "balance_loss_clip": 1.05980968,
      "balance_loss_mlp": 1.02552974,
      "epoch": 0.15367041423675826,
      "flos": 18478590238080.0,
      "grad_norm": 2.0293957750591507,
      "language_loss": 0.83792347,
      "learning_rate": 3.841816989407644e-06,
      "loss": 0.86019593,
      "num_input_tokens_seen": 27065155,
      "step": 1278,
      "time_per_iteration": 2.5481834411621094
    },
    {
      "auxiliary_loss_clip": 0.01161607,
      "auxiliary_loss_mlp": 0.01040664,
      "balance_loss_clip": 1.05467653,
      "balance_loss_mlp": 1.02855265,
      "epoch": 0.15379065712739734,
      "flos": 41427662342400.0,
      "grad_norm": 1.9474550293231425,
      "language_loss": 0.76631171,
      "learning_rate": 3.841513223048884e-06,
      "loss": 0.78833437,
      "num_input_tokens_seen": 27085840,
      "step": 1279,
      "time_per_iteration": 2.7704689502716064
    },
    {
      "auxiliary_loss_clip": 0.01099812,
      "auxiliary_loss_mlp": 0.00765828,
      "balance_loss_clip": 1.04328775,
      "balance_loss_mlp": 1.0004015,
      "epoch": 0.15391090001803642,
      "flos": 22054215553920.0,
      "grad_norm": 2.5623707779107203,
      "language_loss": 0.78922749,
      "learning_rate": 3.841209177334031e-06,
      "loss": 0.80788392,
      "num_input_tokens_seen": 27104200,
      "step": 1280,
      "time_per_iteration": 2.6924219131469727
    },
    {
      "auxiliary_loss_clip": 0.01169294,
      "auxiliary_loss_mlp": 0.00765429,
      "balance_loss_clip": 1.05285037,
      "balance_loss_mlp": 1.00041318,
      "epoch": 0.15403114290867553,
      "flos": 15450351258240.0,
      "grad_norm": 2.119963363068517,
      "language_loss": 0.74681079,
      "learning_rate": 3.84090485230921e-06,
      "loss": 0.7661581,
      "num_input_tokens_seen": 27122440,
      "step": 1281,
      "time_per_iteration": 2.5888750553131104
    },
    {
      "auxiliary_loss_clip": 0.01214316,
      "auxiliary_loss_mlp": 0.0104481,
      "balance_loss_clip": 1.06369388,
      "balance_loss_mlp": 1.03202462,
      "epoch": 0.15415138579931462,
      "flos": 17929156826880.0,
      "grad_norm": 3.1254961716186336,
      "language_loss": 0.76108861,
      "learning_rate": 3.840600248020588e-06,
      "loss": 0.7836799,
      "num_input_tokens_seen": 27139380,
      "step": 1282,
      "time_per_iteration": 2.4955947399139404
    },
    {
      "auxiliary_loss_clip": 0.01211355,
      "auxiliary_loss_mlp": 0.01038824,
      "balance_loss_clip": 1.0613656,
      "balance_loss_mlp": 1.02465594,
      "epoch": 0.1542716286899537,
      "flos": 11429325296640.0,
      "grad_norm": 2.1603032506237496,
      "language_loss": 0.7993933,
      "learning_rate": 3.840295364514371e-06,
      "loss": 0.82189512,
      "num_input_tokens_seen": 27156760,
      "step": 1283,
      "time_per_iteration": 2.525277853012085
    },
    {
      "auxiliary_loss_clip": 0.01134461,
      "auxiliary_loss_mlp": 0.00766019,
      "balance_loss_clip": 1.05576706,
      "balance_loss_mlp": 1.00044966,
      "epoch": 0.1543918715805928,
      "flos": 17420338719360.0,
      "grad_norm": 2.8177243698626797,
      "language_loss": 0.79035127,
      "learning_rate": 3.83999020183681e-06,
      "loss": 0.80935609,
      "num_input_tokens_seen": 27175455,
      "step": 1284,
      "time_per_iteration": 2.654604196548462
    },
    {
      "auxiliary_loss_clip": 0.01226407,
      "auxiliary_loss_mlp": 0.00765838,
      "balance_loss_clip": 1.06838667,
      "balance_loss_mlp": 1.00054121,
      "epoch": 0.1545121144712319,
      "flos": 17786376264960.0,
      "grad_norm": 2.154810638732837,
      "language_loss": 0.78626972,
      "learning_rate": 3.839684760034199e-06,
      "loss": 0.80619222,
      "num_input_tokens_seen": 27193660,
      "step": 1285,
      "time_per_iteration": 3.797245502471924
    },
    {
      "auxiliary_loss_clip": 0.01199865,
      "auxiliary_loss_mlp": 0.01038062,
      "balance_loss_clip": 1.06201506,
      "balance_loss_mlp": 1.02551544,
      "epoch": 0.15463235736187098,
      "flos": 28220185146240.0,
      "grad_norm": 2.628839357851128,
      "language_loss": 0.65119082,
      "learning_rate": 3.8393790391528716e-06,
      "loss": 0.67357004,
      "num_input_tokens_seen": 27214355,
      "step": 1286,
      "time_per_iteration": 3.31315541267395
    },
    {
      "auxiliary_loss_clip": 0.01189872,
      "auxiliary_loss_mlp": 0.01041915,
      "balance_loss_clip": 1.05675769,
      "balance_loss_mlp": 1.02942741,
      "epoch": 0.15475260025251006,
      "flos": 22856890826880.0,
      "grad_norm": 3.0560527264309436,
      "language_loss": 0.89250398,
      "learning_rate": 3.8390730392392075e-06,
      "loss": 0.91482186,
      "num_input_tokens_seen": 27234335,
      "step": 1287,
      "time_per_iteration": 2.546253204345703
    },
    {
      "auxiliary_loss_clip": 0.01164621,
      "auxiliary_loss_mlp": 0.01040704,
      "balance_loss_clip": 1.05698824,
      "balance_loss_mlp": 1.02896833,
      "epoch": 0.15487284314314917,
      "flos": 17602872658560.0,
      "grad_norm": 2.4052591857383523,
      "language_loss": 0.79534572,
      "learning_rate": 3.838766760339626e-06,
      "loss": 0.81739902,
      "num_input_tokens_seen": 27252860,
      "step": 1288,
      "time_per_iteration": 3.355161666870117
    },
    {
      "auxiliary_loss_clip": 0.01118411,
      "auxiliary_loss_mlp": 0.01040125,
      "balance_loss_clip": 1.05226684,
      "balance_loss_mlp": 1.02772748,
      "epoch": 0.15499308603378825,
      "flos": 20082037363200.0,
      "grad_norm": 2.503123547809381,
      "language_loss": 0.7910527,
      "learning_rate": 3.838460202500587e-06,
      "loss": 0.81263804,
      "num_input_tokens_seen": 27268650,
      "step": 1289,
      "time_per_iteration": 3.5399234294891357
    },
    {
      "auxiliary_loss_clip": 0.01235194,
      "auxiliary_loss_mlp": 0.01043303,
      "balance_loss_clip": 1.06774473,
      "balance_loss_mlp": 1.03036928,
      "epoch": 0.15511332892442733,
      "flos": 15918051271680.0,
      "grad_norm": 2.152971061098414,
      "language_loss": 0.74435657,
      "learning_rate": 3.838153365768599e-06,
      "loss": 0.76714152,
      "num_input_tokens_seen": 27285160,
      "step": 1290,
      "time_per_iteration": 2.842212677001953
    },
    {
      "auxiliary_loss_clip": 0.01221776,
      "auxiliary_loss_mlp": 0.01048011,
      "balance_loss_clip": 1.07101905,
      "balance_loss_mlp": 1.03408766,
      "epoch": 0.15523357181506645,
      "flos": 41282475569280.0,
      "grad_norm": 3.481210447624527,
      "language_loss": 0.75525707,
      "learning_rate": 3.837846250190206e-06,
      "loss": 0.77795494,
      "num_input_tokens_seen": 27308025,
      "step": 1291,
      "time_per_iteration": 2.686770439147949
    },
    {
      "auxiliary_loss_clip": 0.01194009,
      "auxiliary_loss_mlp": 0.00765336,
      "balance_loss_clip": 1.0609138,
      "balance_loss_mlp": 1.00055504,
      "epoch": 0.15535381470570553,
      "flos": 18478769806080.0,
      "grad_norm": 1.9698188184998782,
      "language_loss": 0.76867735,
      "learning_rate": 3.837538855811998e-06,
      "loss": 0.78827089,
      "num_input_tokens_seen": 27326200,
      "step": 1292,
      "time_per_iteration": 2.5315773487091064
    },
    {
      "auxiliary_loss_clip": 0.01185678,
      "auxiliary_loss_mlp": 0.00765734,
      "balance_loss_clip": 1.05873632,
      "balance_loss_mlp": 1.00055814,
      "epoch": 0.1554740575963446,
      "flos": 13918150759680.0,
      "grad_norm": 2.1328593868534513,
      "language_loss": 0.71259063,
      "learning_rate": 3.837231182680606e-06,
      "loss": 0.73210478,
      "num_input_tokens_seen": 27344165,
      "step": 1293,
      "time_per_iteration": 2.5581231117248535
    },
    {
      "auxiliary_loss_clip": 0.01166291,
      "auxiliary_loss_mlp": 0.01039555,
      "balance_loss_clip": 1.05699801,
      "balance_loss_mlp": 1.02842665,
      "epoch": 0.1555943004869837,
      "flos": 20847078161280.0,
      "grad_norm": 1.621692920263699,
      "language_loss": 0.76323509,
      "learning_rate": 3.836923230842706e-06,
      "loss": 0.78529358,
      "num_input_tokens_seen": 27363280,
      "step": 1294,
      "time_per_iteration": 2.635871648788452
    },
    {
      "auxiliary_loss_clip": 0.01169794,
      "auxiliary_loss_mlp": 0.01046143,
      "balance_loss_clip": 1.0566783,
      "balance_loss_mlp": 1.03371525,
      "epoch": 0.1557145433776228,
      "flos": 22085888371200.0,
      "grad_norm": 2.2052160662845184,
      "language_loss": 0.80639386,
      "learning_rate": 3.836615000345011e-06,
      "loss": 0.8285532,
      "num_input_tokens_seen": 27381460,
      "step": 1295,
      "time_per_iteration": 2.5942468643188477
    },
    {
      "auxiliary_loss_clip": 0.01183665,
      "auxiliary_loss_mlp": 0.01043787,
      "balance_loss_clip": 1.0585196,
      "balance_loss_mlp": 1.03208661,
      "epoch": 0.1558347862682619,
      "flos": 19791987039360.0,
      "grad_norm": 2.340057170302178,
      "language_loss": 0.78597736,
      "learning_rate": 3.836306491234282e-06,
      "loss": 0.80825186,
      "num_input_tokens_seen": 27399310,
      "step": 1296,
      "time_per_iteration": 2.5711283683776855
    },
    {
      "auxiliary_loss_clip": 0.01186102,
      "auxiliary_loss_mlp": 0.01038431,
      "balance_loss_clip": 1.06356955,
      "balance_loss_mlp": 1.02646828,
      "epoch": 0.15595502915890097,
      "flos": 17237086508160.0,
      "grad_norm": 5.049260996907364,
      "language_loss": 0.75779676,
      "learning_rate": 3.835997703557317e-06,
      "loss": 0.78004211,
      "num_input_tokens_seen": 27416050,
      "step": 1297,
      "time_per_iteration": 2.56089448928833
    },
    {
      "auxiliary_loss_clip": 0.01181045,
      "auxiliary_loss_mlp": 0.01043936,
      "balance_loss_clip": 1.05778599,
      "balance_loss_mlp": 1.03267026,
      "epoch": 0.15607527204954008,
      "flos": 19719519350400.0,
      "grad_norm": 1.8138002145661785,
      "language_loss": 0.80120277,
      "learning_rate": 3.83568863736096e-06,
      "loss": 0.82345259,
      "num_input_tokens_seen": 27434920,
      "step": 1298,
      "time_per_iteration": 2.5677785873413086
    },
    {
      "auxiliary_loss_clip": 0.01149343,
      "auxiliary_loss_mlp": 0.01039796,
      "balance_loss_clip": 1.04861152,
      "balance_loss_mlp": 1.02765453,
      "epoch": 0.15619551494017916,
      "flos": 18515650095360.0,
      "grad_norm": 3.9475836455556843,
      "language_loss": 0.8941161,
      "learning_rate": 3.8353792926920975e-06,
      "loss": 0.91600752,
      "num_input_tokens_seen": 27453570,
      "step": 1299,
      "time_per_iteration": 2.611579418182373
    },
    {
      "auxiliary_loss_clip": 0.01177553,
      "auxiliary_loss_mlp": 0.0104489,
      "balance_loss_clip": 1.06162345,
      "balance_loss_mlp": 1.03229499,
      "epoch": 0.15631575783081825,
      "flos": 19902125116800.0,
      "grad_norm": 2.6657906435223158,
      "language_loss": 0.8207894,
      "learning_rate": 3.835069669597655e-06,
      "loss": 0.84301376,
      "num_input_tokens_seen": 27471960,
      "step": 1300,
      "time_per_iteration": 2.552718162536621
    },
    {
      "auxiliary_loss_clip": 0.01205249,
      "auxiliary_loss_mlp": 0.01044918,
      "balance_loss_clip": 1.06067634,
      "balance_loss_mlp": 1.03144157,
      "epoch": 0.15643600072145733,
      "flos": 20777663128320.0,
      "grad_norm": 1.9955358161486234,
      "language_loss": 0.80036342,
      "learning_rate": 3.834759768124603e-06,
      "loss": 0.82286507,
      "num_input_tokens_seen": 27490835,
      "step": 1301,
      "time_per_iteration": 2.5593230724334717
    },
    {
      "auxiliary_loss_clip": 0.01204777,
      "auxiliary_loss_mlp": 0.01043168,
      "balance_loss_clip": 1.06870341,
      "balance_loss_mlp": 1.0308001,
      "epoch": 0.15655624361209644,
      "flos": 18546389159040.0,
      "grad_norm": 2.730714698227011,
      "language_loss": 0.76533818,
      "learning_rate": 3.834449588319953e-06,
      "loss": 0.7878176,
      "num_input_tokens_seen": 27508870,
      "step": 1302,
      "time_per_iteration": 2.506943464279175
    },
    {
      "auxiliary_loss_clip": 0.01195912,
      "auxiliary_loss_mlp": 0.01041667,
      "balance_loss_clip": 1.0653528,
      "balance_loss_mlp": 1.03022873,
      "epoch": 0.15667648650273552,
      "flos": 25229544727680.0,
      "grad_norm": 2.1556524471701732,
      "language_loss": 0.85358071,
      "learning_rate": 3.834139130230758e-06,
      "loss": 0.87595648,
      "num_input_tokens_seen": 27528175,
      "step": 1303,
      "time_per_iteration": 2.601762056350708
    },
    {
      "auxiliary_loss_clip": 0.01140743,
      "auxiliary_loss_mlp": 0.01036295,
      "balance_loss_clip": 1.05260348,
      "balance_loss_mlp": 1.02284884,
      "epoch": 0.1567967293933746,
      "flos": 24827093769600.0,
      "grad_norm": 1.6199832980470321,
      "language_loss": 0.8119961,
      "learning_rate": 3.833828393904117e-06,
      "loss": 0.83376652,
      "num_input_tokens_seen": 27548455,
      "step": 1304,
      "time_per_iteration": 2.667160987854004
    },
    {
      "auxiliary_loss_clip": 0.01155104,
      "auxiliary_loss_mlp": 0.01048889,
      "balance_loss_clip": 1.05753636,
      "balance_loss_mlp": 1.0366044,
      "epoch": 0.15691697228401372,
      "flos": 19164555244800.0,
      "grad_norm": 2.46149046856209,
      "language_loss": 0.77398372,
      "learning_rate": 3.833517379387165e-06,
      "loss": 0.79602361,
      "num_input_tokens_seen": 27564910,
      "step": 1305,
      "time_per_iteration": 2.6556203365325928
    },
    {
      "auxiliary_loss_clip": 0.01198115,
      "auxiliary_loss_mlp": 0.01038002,
      "balance_loss_clip": 1.0610373,
      "balance_loss_mlp": 1.0257833,
      "epoch": 0.1570372151746528,
      "flos": 24790931752320.0,
      "grad_norm": 2.0046218235838884,
      "language_loss": 0.89007664,
      "learning_rate": 3.833206086727085e-06,
      "loss": 0.91243786,
      "num_input_tokens_seen": 27584260,
      "step": 1306,
      "time_per_iteration": 2.5646166801452637
    },
    {
      "auxiliary_loss_clip": 0.0116209,
      "auxiliary_loss_mlp": 0.01035714,
      "balance_loss_clip": 1.05551457,
      "balance_loss_mlp": 1.02345979,
      "epoch": 0.15715745806529188,
      "flos": 24863650836480.0,
      "grad_norm": 2.3366486925776173,
      "language_loss": 0.70915645,
      "learning_rate": 3.8328945159710994e-06,
      "loss": 0.73113447,
      "num_input_tokens_seen": 27604440,
      "step": 1307,
      "time_per_iteration": 2.672318696975708
    },
    {
      "auxiliary_loss_clip": 0.01185953,
      "auxiliary_loss_mlp": 0.01035823,
      "balance_loss_clip": 1.06423473,
      "balance_loss_mlp": 1.0235858,
      "epoch": 0.157277700955931,
      "flos": 21872148491520.0,
      "grad_norm": 2.1321278722931614,
      "language_loss": 0.89258134,
      "learning_rate": 3.832582667166473e-06,
      "loss": 0.91479909,
      "num_input_tokens_seen": 27624250,
      "step": 1308,
      "time_per_iteration": 2.562042236328125
    },
    {
      "auxiliary_loss_clip": 0.01194804,
      "auxiliary_loss_mlp": 0.0103896,
      "balance_loss_clip": 1.06121469,
      "balance_loss_mlp": 1.02643144,
      "epoch": 0.15739794384657008,
      "flos": 24533344344960.0,
      "grad_norm": 1.6776629582472713,
      "language_loss": 0.81323647,
      "learning_rate": 3.8322705403605125e-06,
      "loss": 0.83557415,
      "num_input_tokens_seen": 27644595,
      "step": 1309,
      "time_per_iteration": 2.568938732147217
    },
    {
      "auxiliary_loss_clip": 0.0115697,
      "auxiliary_loss_mlp": 0.0103195,
      "balance_loss_clip": 1.05215025,
      "balance_loss_mlp": 1.02007139,
      "epoch": 0.15751818673720916,
      "flos": 17745329998080.0,
      "grad_norm": 1.9757473255244484,
      "language_loss": 0.80999476,
      "learning_rate": 3.831958135600568e-06,
      "loss": 0.83188391,
      "num_input_tokens_seen": 27662145,
      "step": 1310,
      "time_per_iteration": 3.2811052799224854
    },
    {
      "auxiliary_loss_clip": 0.01162405,
      "auxiliary_loss_mlp": 0.01036343,
      "balance_loss_clip": 1.05520141,
      "balance_loss_mlp": 1.02466011,
      "epoch": 0.15763842962784824,
      "flos": 17858520731520.0,
      "grad_norm": 2.14747346988324,
      "language_loss": 0.79194427,
      "learning_rate": 3.831645452934032e-06,
      "loss": 0.81393182,
      "num_input_tokens_seen": 27680575,
      "step": 1311,
      "time_per_iteration": 2.582979440689087
    },
    {
      "auxiliary_loss_clip": 0.0121869,
      "auxiliary_loss_mlp": 0.01039067,
      "balance_loss_clip": 1.06748128,
      "balance_loss_mlp": 1.02669883,
      "epoch": 0.15775867251848735,
      "flos": 26980908059520.0,
      "grad_norm": 1.9441941336647295,
      "language_loss": 0.79953277,
      "learning_rate": 3.831332492408336e-06,
      "loss": 0.82211035,
      "num_input_tokens_seen": 27701985,
      "step": 1312,
      "time_per_iteration": 3.243582010269165
    },
    {
      "auxiliary_loss_clip": 0.01194256,
      "auxiliary_loss_mlp": 0.0103586,
      "balance_loss_clip": 1.0578959,
      "balance_loss_mlp": 1.02392113,
      "epoch": 0.15787891540912644,
      "flos": 19240398812160.0,
      "grad_norm": 2.200364735316712,
      "language_loss": 0.69073057,
      "learning_rate": 3.831019254070957e-06,
      "loss": 0.71303171,
      "num_input_tokens_seen": 27719770,
      "step": 1313,
      "time_per_iteration": 3.2778193950653076
    },
    {
      "auxiliary_loss_clip": 0.01166921,
      "auxiliary_loss_mlp": 0.01039061,
      "balance_loss_clip": 1.05970418,
      "balance_loss_mlp": 1.02711582,
      "epoch": 0.15799915829976552,
      "flos": 27271102037760.0,
      "grad_norm": 2.474902415479631,
      "language_loss": 0.95546114,
      "learning_rate": 3.8307057379694135e-06,
      "loss": 0.97752106,
      "num_input_tokens_seen": 27739105,
      "step": 1314,
      "time_per_iteration": 2.6603939533233643
    },
    {
      "auxiliary_loss_clip": 0.01160081,
      "auxiliary_loss_mlp": 0.00765409,
      "balance_loss_clip": 1.05766296,
      "balance_loss_mlp": 1.00039184,
      "epoch": 0.15811940119040463,
      "flos": 20405520270720.0,
      "grad_norm": 2.0824001558701157,
      "language_loss": 0.82629573,
      "learning_rate": 3.830391944151264e-06,
      "loss": 0.84555066,
      "num_input_tokens_seen": 27754985,
      "step": 1315,
      "time_per_iteration": 3.295638084411621
    },
    {
      "auxiliary_loss_clip": 0.0121399,
      "auxiliary_loss_mlp": 0.01031424,
      "balance_loss_clip": 1.06401062,
      "balance_loss_mlp": 1.02022457,
      "epoch": 0.1582396440810437,
      "flos": 32599347661440.0,
      "grad_norm": 2.278239680618027,
      "language_loss": 0.67367256,
      "learning_rate": 3.830077872664114e-06,
      "loss": 0.69612664,
      "num_input_tokens_seen": 27776110,
      "step": 1316,
      "time_per_iteration": 2.6514344215393066
    },
    {
      "auxiliary_loss_clip": 0.0121603,
      "auxiliary_loss_mlp": 0.01035886,
      "balance_loss_clip": 1.06232142,
      "balance_loss_mlp": 1.02477551,
      "epoch": 0.1583598869716828,
      "flos": 33800559310080.0,
      "grad_norm": 1.952637861978402,
      "language_loss": 0.73358428,
      "learning_rate": 3.829763523555604e-06,
      "loss": 0.75610346,
      "num_input_tokens_seen": 27796510,
      "step": 1317,
      "time_per_iteration": 2.611593008041382
    },
    {
      "auxiliary_loss_clip": 0.01189442,
      "auxiliary_loss_mlp": 0.01032784,
      "balance_loss_clip": 1.06361818,
      "balance_loss_mlp": 1.02153075,
      "epoch": 0.15848012986232188,
      "flos": 24681332378880.0,
      "grad_norm": 2.5907492273745656,
      "language_loss": 0.78014475,
      "learning_rate": 3.829448896873423e-06,
      "loss": 0.80236697,
      "num_input_tokens_seen": 27815610,
      "step": 1318,
      "time_per_iteration": 2.644508123397827
    },
    {
      "auxiliary_loss_clip": 0.01174328,
      "auxiliary_loss_mlp": 0.01041625,
      "balance_loss_clip": 1.06287467,
      "balance_loss_mlp": 1.03050232,
      "epoch": 0.158600372752961,
      "flos": 22602068766720.0,
      "grad_norm": 1.7813321493006586,
      "language_loss": 0.79168063,
      "learning_rate": 3.829133992665299e-06,
      "loss": 0.81384015,
      "num_input_tokens_seen": 27834735,
      "step": 1319,
      "time_per_iteration": 2.592498540878296
    },
    {
      "auxiliary_loss_clip": 0.01160304,
      "auxiliary_loss_mlp": 0.0104001,
      "balance_loss_clip": 1.05644846,
      "balance_loss_mlp": 1.02892327,
      "epoch": 0.15872061564360007,
      "flos": 27927944092800.0,
      "grad_norm": 2.626333814052586,
      "language_loss": 0.88961279,
      "learning_rate": 3.828818810979002e-06,
      "loss": 0.91161597,
      "num_input_tokens_seen": 27853065,
      "step": 1320,
      "time_per_iteration": 2.674118757247925
    },
    {
      "auxiliary_loss_clip": 0.0120674,
      "auxiliary_loss_mlp": 0.01041775,
      "balance_loss_clip": 1.06324494,
      "balance_loss_mlp": 1.02944279,
      "epoch": 0.15884085853423915,
      "flos": 23696805525120.0,
      "grad_norm": 2.082290903697534,
      "language_loss": 0.80796432,
      "learning_rate": 3.8285033518623454e-06,
      "loss": 0.83044952,
      "num_input_tokens_seen": 27873315,
      "step": 1321,
      "time_per_iteration": 2.5716638565063477
    },
    {
      "auxiliary_loss_clip": 0.01188842,
      "auxiliary_loss_mlp": 0.01044177,
      "balance_loss_clip": 1.0611732,
      "balance_loss_mlp": 1.03179145,
      "epoch": 0.15896110142487826,
      "flos": 23112359331840.0,
      "grad_norm": 2.3232694496666864,
      "language_loss": 0.81109142,
      "learning_rate": 3.8281876153631845e-06,
      "loss": 0.83342159,
      "num_input_tokens_seen": 27890070,
      "step": 1322,
      "time_per_iteration": 2.5865743160247803
    },
    {
      "auxiliary_loss_clip": 0.01150587,
      "auxiliary_loss_mlp": 0.01039537,
      "balance_loss_clip": 1.05369973,
      "balance_loss_mlp": 1.02770543,
      "epoch": 0.15908134431551735,
      "flos": 14685238632960.0,
      "grad_norm": 2.0014097435056017,
      "language_loss": 0.64314032,
      "learning_rate": 3.827871601529416e-06,
      "loss": 0.66504157,
      "num_input_tokens_seen": 27908590,
      "step": 1323,
      "time_per_iteration": 2.6737823486328125
    },
    {
      "auxiliary_loss_clip": 0.01200321,
      "auxiliary_loss_mlp": 0.0103725,
      "balance_loss_clip": 1.06414545,
      "balance_loss_mlp": 1.02532339,
      "epoch": 0.15920158720615643,
      "flos": 20193611984640.0,
      "grad_norm": 1.8526555463801495,
      "language_loss": 0.80574834,
      "learning_rate": 3.827555310408979e-06,
      "loss": 0.82812405,
      "num_input_tokens_seen": 27927985,
      "step": 1324,
      "time_per_iteration": 2.594107151031494
    },
    {
      "auxiliary_loss_clip": 0.01218077,
      "auxiliary_loss_mlp": 0.01037867,
      "balance_loss_clip": 1.06504083,
      "balance_loss_mlp": 1.0264827,
      "epoch": 0.1593218300967955,
      "flos": 24826626892800.0,
      "grad_norm": 1.664653785184928,
      "language_loss": 0.83011448,
      "learning_rate": 3.827238742049854e-06,
      "loss": 0.85267395,
      "num_input_tokens_seen": 27948280,
      "step": 1325,
      "time_per_iteration": 2.558812379837036
    },
    {
      "auxiliary_loss_clip": 0.01198367,
      "auxiliary_loss_mlp": 0.00765317,
      "balance_loss_clip": 1.05946231,
      "balance_loss_mlp": 1.00043344,
      "epoch": 0.15944207298743462,
      "flos": 28328707111680.0,
      "grad_norm": 2.0114754427881416,
      "language_loss": 0.52091444,
      "learning_rate": 3.826921896500066e-06,
      "loss": 0.5405513,
      "num_input_tokens_seen": 27969565,
      "step": 1326,
      "time_per_iteration": 2.6109704971313477
    },
    {
      "auxiliary_loss_clip": 0.01190843,
      "auxiliary_loss_mlp": 0.01038424,
      "balance_loss_clip": 1.05846882,
      "balance_loss_mlp": 1.02608562,
      "epoch": 0.1595623158780737,
      "flos": 22964838174720.0,
      "grad_norm": 1.9913569438699064,
      "language_loss": 0.78393197,
      "learning_rate": 3.826604773807678e-06,
      "loss": 0.80622458,
      "num_input_tokens_seen": 27987540,
      "step": 1327,
      "time_per_iteration": 2.5647764205932617
    },
    {
      "auxiliary_loss_clip": 0.01198643,
      "auxiliary_loss_mlp": 0.01038261,
      "balance_loss_clip": 1.06533623,
      "balance_loss_mlp": 1.02560699,
      "epoch": 0.1596825587687128,
      "flos": 19710540950400.0,
      "grad_norm": 2.629621231196323,
      "language_loss": 0.73201835,
      "learning_rate": 3.826287374020798e-06,
      "loss": 0.75438738,
      "num_input_tokens_seen": 28002345,
      "step": 1328,
      "time_per_iteration": 2.499680995941162
    },
    {
      "auxiliary_loss_clip": 0.01172804,
      "auxiliary_loss_mlp": 0.01038907,
      "balance_loss_clip": 1.06229138,
      "balance_loss_mlp": 1.02725422,
      "epoch": 0.1598028016593519,
      "flos": 22637727993600.0,
      "grad_norm": 2.3428608062532144,
      "language_loss": 0.81745893,
      "learning_rate": 3.825969697187575e-06,
      "loss": 0.83957607,
      "num_input_tokens_seen": 28021675,
      "step": 1329,
      "time_per_iteration": 2.589390754699707
    },
    {
      "auxiliary_loss_clip": 0.01188271,
      "auxiliary_loss_mlp": 0.01033205,
      "balance_loss_clip": 1.06220663,
      "balance_loss_mlp": 1.02258348,
      "epoch": 0.15992304454999098,
      "flos": 20482908122880.0,
      "grad_norm": 2.0387171206680637,
      "language_loss": 0.69746882,
      "learning_rate": 3.8256517433562015e-06,
      "loss": 0.71968359,
      "num_input_tokens_seen": 28039615,
      "step": 1330,
      "time_per_iteration": 2.5462403297424316
    },
    {
      "auxiliary_loss_clip": 0.0121215,
      "auxiliary_loss_mlp": 0.01036518,
      "balance_loss_clip": 1.06359911,
      "balance_loss_mlp": 1.02449012,
      "epoch": 0.16004328744063007,
      "flos": 17676094533120.0,
      "grad_norm": 2.452096967253773,
      "language_loss": 0.91346931,
      "learning_rate": 3.82533351257491e-06,
      "loss": 0.935956,
      "num_input_tokens_seen": 28057565,
      "step": 1331,
      "time_per_iteration": 2.4882216453552246
    },
    {
      "auxiliary_loss_clip": 0.01214119,
      "auxiliary_loss_mlp": 0.00764753,
      "balance_loss_clip": 1.06248665,
      "balance_loss_mlp": 1.00039673,
      "epoch": 0.16016353033126918,
      "flos": 24098717779200.0,
      "grad_norm": 1.7980183038007762,
      "language_loss": 0.88411021,
      "learning_rate": 3.825015004891975e-06,
      "loss": 0.90389895,
      "num_input_tokens_seen": 28076305,
      "step": 1332,
      "time_per_iteration": 2.5121028423309326
    },
    {
      "auxiliary_loss_clip": 0.0119762,
      "auxiliary_loss_mlp": 0.01038231,
      "balance_loss_clip": 1.05958271,
      "balance_loss_mlp": 1.02715015,
      "epoch": 0.16028377322190826,
      "flos": 27634841112960.0,
      "grad_norm": 2.3035765591055744,
      "language_loss": 0.75687414,
      "learning_rate": 3.824696220355716e-06,
      "loss": 0.77923268,
      "num_input_tokens_seen": 28097895,
      "step": 1333,
      "time_per_iteration": 2.5807383060455322
    },
    {
      "auxiliary_loss_clip": 0.01205624,
      "auxiliary_loss_mlp": 0.01038476,
      "balance_loss_clip": 1.06105554,
      "balance_loss_mlp": 1.02663875,
      "epoch": 0.16040401611254734,
      "flos": 20961202648320.0,
      "grad_norm": 1.6172145247593024,
      "language_loss": 0.78861117,
      "learning_rate": 3.824377159014491e-06,
      "loss": 0.8110522,
      "num_input_tokens_seen": 28118790,
      "step": 1334,
      "time_per_iteration": 2.5488593578338623
    },
    {
      "auxiliary_loss_clip": 0.01187651,
      "auxiliary_loss_mlp": 0.01042913,
      "balance_loss_clip": 1.0593605,
      "balance_loss_mlp": 1.03159428,
      "epoch": 0.16052425900318643,
      "flos": 21247051080960.0,
      "grad_norm": 2.2353920049737277,
      "language_loss": 0.84701157,
      "learning_rate": 3.824057820916702e-06,
      "loss": 0.86931717,
      "num_input_tokens_seen": 28135995,
      "step": 1335,
      "time_per_iteration": 2.538578987121582
    },
    {
      "auxiliary_loss_clip": 0.01198464,
      "auxiliary_loss_mlp": 0.00765471,
      "balance_loss_clip": 1.06299162,
      "balance_loss_mlp": 1.00050378,
      "epoch": 0.16064450189382554,
      "flos": 15524004096000.0,
      "grad_norm": 2.2519340051277186,
      "language_loss": 0.72170156,
      "learning_rate": 3.8237382061107904e-06,
      "loss": 0.74134099,
      "num_input_tokens_seen": 28152715,
      "step": 1336,
      "time_per_iteration": 3.2298238277435303
    },
    {
      "auxiliary_loss_clip": 0.01188125,
      "auxiliary_loss_mlp": 0.01033804,
      "balance_loss_clip": 1.05877888,
      "balance_loss_mlp": 1.0219965,
      "epoch": 0.16076474478446462,
      "flos": 21178497974400.0,
      "grad_norm": 1.8223891966943613,
      "language_loss": 0.78903246,
      "learning_rate": 3.823418314645243e-06,
      "loss": 0.81125176,
      "num_input_tokens_seen": 28171590,
      "step": 1337,
      "time_per_iteration": 3.3107943534851074
    },
    {
      "auxiliary_loss_clip": 0.0114832,
      "auxiliary_loss_mlp": 0.01046269,
      "balance_loss_clip": 1.05324793,
      "balance_loss_mlp": 1.03456891,
      "epoch": 0.1608849876751037,
      "flos": 18366476912640.0,
      "grad_norm": 2.670631379720977,
      "language_loss": 0.75177592,
      "learning_rate": 3.823098146568588e-06,
      "loss": 0.77372181,
      "num_input_tokens_seen": 28191295,
      "step": 1338,
      "time_per_iteration": 2.639279842376709
    },
    {
      "auxiliary_loss_clip": 0.01215169,
      "auxiliary_loss_mlp": 0.01035895,
      "balance_loss_clip": 1.06480777,
      "balance_loss_mlp": 1.02446842,
      "epoch": 0.1610052305657428,
      "flos": 29497024880640.0,
      "grad_norm": 2.0612820724962204,
      "language_loss": 0.71498734,
      "learning_rate": 3.822777701929394e-06,
      "loss": 0.73749793,
      "num_input_tokens_seen": 28213120,
      "step": 1339,
      "time_per_iteration": 2.60864520072937
    },
    {
      "auxiliary_loss_clip": 0.01211163,
      "auxiliary_loss_mlp": 0.01040282,
      "balance_loss_clip": 1.0640409,
      "balance_loss_mlp": 1.02898705,
      "epoch": 0.1611254734563819,
      "flos": 26797871329920.0,
      "grad_norm": 1.949065430597509,
      "language_loss": 0.73535919,
      "learning_rate": 3.8224569807762714e-06,
      "loss": 0.75787365,
      "num_input_tokens_seen": 28232440,
      "step": 1340,
      "time_per_iteration": 3.2597463130950928
    },
    {
      "auxiliary_loss_clip": 0.01110593,
      "auxiliary_loss_mlp": 0.01032771,
      "balance_loss_clip": 1.04982269,
      "balance_loss_mlp": 1.02071261,
      "epoch": 0.16124571634702098,
      "flos": 22419570741120.0,
      "grad_norm": 1.9275636694934901,
      "language_loss": 0.76198518,
      "learning_rate": 3.822135983157873e-06,
      "loss": 0.78341877,
      "num_input_tokens_seen": 28251715,
      "step": 1341,
      "time_per_iteration": 3.3938488960266113
    },
    {
      "auxiliary_loss_clip": 0.01183162,
      "auxiliary_loss_mlp": 0.00764168,
      "balance_loss_clip": 1.0566721,
      "balance_loss_mlp": 1.00048971,
      "epoch": 0.16136595923766006,
      "flos": 10999116103680.0,
      "grad_norm": 2.2182944442777495,
      "language_loss": 0.84304243,
      "learning_rate": 3.821814709122896e-06,
      "loss": 0.86251569,
      "num_input_tokens_seen": 28269765,
      "step": 1342,
      "time_per_iteration": 2.5603530406951904
    },
    {
      "auxiliary_loss_clip": 0.01199556,
      "auxiliary_loss_mlp": 0.010349,
      "balance_loss_clip": 1.06242633,
      "balance_loss_mlp": 1.0250591,
      "epoch": 0.16148620212829917,
      "flos": 21214983214080.0,
      "grad_norm": 2.0795979768050725,
      "language_loss": 0.84850407,
      "learning_rate": 3.821493158720076e-06,
      "loss": 0.87084866,
      "num_input_tokens_seen": 28288870,
      "step": 1343,
      "time_per_iteration": 2.546074628829956
    },
    {
      "auxiliary_loss_clip": 0.01174423,
      "auxiliary_loss_mlp": 0.01040171,
      "balance_loss_clip": 1.05921388,
      "balance_loss_mlp": 1.02805936,
      "epoch": 0.16160644501893826,
      "flos": 16758468760320.0,
      "grad_norm": 3.1342269347691265,
      "language_loss": 0.73307067,
      "learning_rate": 3.821171331998191e-06,
      "loss": 0.7552166,
      "num_input_tokens_seen": 28305400,
      "step": 1344,
      "time_per_iteration": 2.553805351257324
    },
    {
      "auxiliary_loss_clip": 0.01125045,
      "auxiliary_loss_mlp": 0.01002131,
      "balance_loss_clip": 1.04930234,
      "balance_loss_mlp": 0.99983072,
      "epoch": 0.16172668790957734,
      "flos": 64444967308800.0,
      "grad_norm": 0.8554587235758561,
      "language_loss": 0.54491007,
      "learning_rate": 3.820849229006064e-06,
      "loss": 0.56618178,
      "num_input_tokens_seen": 28373150,
      "step": 1345,
      "time_per_iteration": 3.2714414596557617
    },
    {
      "auxiliary_loss_clip": 0.01199498,
      "auxiliary_loss_mlp": 0.01047789,
      "balance_loss_clip": 1.06034815,
      "balance_loss_mlp": 1.03638625,
      "epoch": 0.16184693080021645,
      "flos": 23257689759360.0,
      "grad_norm": 1.9052777302689876,
      "language_loss": 0.70793056,
      "learning_rate": 3.8205268497925564e-06,
      "loss": 0.73040348,
      "num_input_tokens_seen": 28393620,
      "step": 1346,
      "time_per_iteration": 2.5408647060394287
    },
    {
      "auxiliary_loss_clip": 0.01181503,
      "auxiliary_loss_mlp": 0.01038413,
      "balance_loss_clip": 1.06021833,
      "balance_loss_mlp": 1.02726734,
      "epoch": 0.16196717369085553,
      "flos": 17451113696640.0,
      "grad_norm": 2.1113428062179067,
      "language_loss": 0.78387243,
      "learning_rate": 3.8202041944065725e-06,
      "loss": 0.80607164,
      "num_input_tokens_seen": 28409440,
      "step": 1347,
      "time_per_iteration": 2.5444588661193848
    },
    {
      "auxiliary_loss_clip": 0.01138249,
      "auxiliary_loss_mlp": 0.01044016,
      "balance_loss_clip": 1.0516758,
      "balance_loss_mlp": 1.03264952,
      "epoch": 0.16208741658149461,
      "flos": 23873377806720.0,
      "grad_norm": 2.1462614976410888,
      "language_loss": 0.73952186,
      "learning_rate": 3.819881262897061e-06,
      "loss": 0.76134455,
      "num_input_tokens_seen": 28427575,
      "step": 1348,
      "time_per_iteration": 2.627826452255249
    },
    {
      "auxiliary_loss_clip": 0.01173463,
      "auxiliary_loss_mlp": 0.01035154,
      "balance_loss_clip": 1.05970418,
      "balance_loss_mlp": 1.02236247,
      "epoch": 0.1622076594721337,
      "flos": 25884806584320.0,
      "grad_norm": 7.370232018482667,
      "language_loss": 0.73585594,
      "learning_rate": 3.819558055313008e-06,
      "loss": 0.75794208,
      "num_input_tokens_seen": 28448260,
      "step": 1349,
      "time_per_iteration": 2.6450541019439697
    },
    {
      "auxiliary_loss_clip": 0.01200085,
      "auxiliary_loss_mlp": 0.01054852,
      "balance_loss_clip": 1.06097174,
      "balance_loss_mlp": 1.04357529,
      "epoch": 0.1623279023627728,
      "flos": 21539759011200.0,
      "grad_norm": 2.069179288135169,
      "language_loss": 0.77343524,
      "learning_rate": 3.819234571703444e-06,
      "loss": 0.79598463,
      "num_input_tokens_seen": 28467085,
      "step": 1350,
      "time_per_iteration": 2.5270514488220215
    },
    {
      "auxiliary_loss_clip": 0.01169246,
      "auxiliary_loss_mlp": 0.01038906,
      "balance_loss_clip": 1.05801797,
      "balance_loss_mlp": 1.02581692,
      "epoch": 0.1624481452534119,
      "flos": 22085421494400.0,
      "grad_norm": 1.7148074046590782,
      "language_loss": 0.8525306,
      "learning_rate": 3.8189108121174435e-06,
      "loss": 0.87461209,
      "num_input_tokens_seen": 28486850,
      "step": 1351,
      "time_per_iteration": 2.6074275970458984
    },
    {
      "auxiliary_loss_clip": 0.01164733,
      "auxiliary_loss_mlp": 0.01044844,
      "balance_loss_clip": 1.05954361,
      "balance_loss_mlp": 1.03418112,
      "epoch": 0.16256838814405097,
      "flos": 27087490690560.0,
      "grad_norm": 2.530021375176404,
      "language_loss": 0.83712399,
      "learning_rate": 3.818586776604118e-06,
      "loss": 0.85921979,
      "num_input_tokens_seen": 28507490,
      "step": 1352,
      "time_per_iteration": 2.6319925785064697
    },
    {
      "auxiliary_loss_clip": 0.01130762,
      "auxiliary_loss_mlp": 0.01044206,
      "balance_loss_clip": 1.0479548,
      "balance_loss_mlp": 1.03232646,
      "epoch": 0.16268863103469008,
      "flos": 20120354196480.0,
      "grad_norm": 2.034800169677207,
      "language_loss": 0.6133458,
      "learning_rate": 3.818262465212625e-06,
      "loss": 0.63509548,
      "num_input_tokens_seen": 28527615,
      "step": 1353,
      "time_per_iteration": 2.789869785308838
    },
    {
      "auxiliary_loss_clip": 0.01175561,
      "auxiliary_loss_mlp": 0.01035426,
      "balance_loss_clip": 1.05916369,
      "balance_loss_mlp": 1.02269471,
      "epoch": 0.16280887392532917,
      "flos": 18332792933760.0,
      "grad_norm": 2.058859350106934,
      "language_loss": 0.77159828,
      "learning_rate": 3.817937877992161e-06,
      "loss": 0.79370809,
      "num_input_tokens_seen": 28544910,
      "step": 1354,
      "time_per_iteration": 2.729691743850708
    },
    {
      "auxiliary_loss_clip": 0.01136441,
      "auxiliary_loss_mlp": 0.01048558,
      "balance_loss_clip": 1.05528128,
      "balance_loss_mlp": 1.03632748,
      "epoch": 0.16292911681596825,
      "flos": 11874330892800.0,
      "grad_norm": 2.6494113262882655,
      "language_loss": 0.85420865,
      "learning_rate": 3.817613014991967e-06,
      "loss": 0.87605864,
      "num_input_tokens_seen": 28561050,
      "step": 1355,
      "time_per_iteration": 2.629944086074829
    },
    {
      "auxiliary_loss_clip": 0.01179675,
      "auxiliary_loss_mlp": 0.01038128,
      "balance_loss_clip": 1.05788064,
      "balance_loss_mlp": 1.02617121,
      "epoch": 0.16304935970660733,
      "flos": 26103466627200.0,
      "grad_norm": 2.6027443196041276,
      "language_loss": 0.76632971,
      "learning_rate": 3.817287876261323e-06,
      "loss": 0.7885077,
      "num_input_tokens_seen": 28581385,
      "step": 1356,
      "time_per_iteration": 2.6000521183013916
    },
    {
      "auxiliary_loss_clip": 0.01160856,
      "auxiliary_loss_mlp": 0.0103185,
      "balance_loss_clip": 1.05911696,
      "balance_loss_mlp": 1.01954722,
      "epoch": 0.16316960259724644,
      "flos": 29351945848320.0,
      "grad_norm": 1.9185960860684905,
      "language_loss": 0.80108565,
      "learning_rate": 3.816962461849553e-06,
      "loss": 0.82301277,
      "num_input_tokens_seen": 28603255,
      "step": 1357,
      "time_per_iteration": 2.6758785247802734
    },
    {
      "auxiliary_loss_clip": 0.01195918,
      "auxiliary_loss_mlp": 0.01042703,
      "balance_loss_clip": 1.05884254,
      "balance_loss_mlp": 1.03159308,
      "epoch": 0.16328984548788553,
      "flos": 20886759711360.0,
      "grad_norm": 2.2739593155998628,
      "language_loss": 0.84453392,
      "learning_rate": 3.8166367718060235e-06,
      "loss": 0.86692011,
      "num_input_tokens_seen": 28623145,
      "step": 1358,
      "time_per_iteration": 2.5265235900878906
    },
    {
      "auxiliary_loss_clip": 0.01166905,
      "auxiliary_loss_mlp": 0.01039706,
      "balance_loss_clip": 1.05370843,
      "balance_loss_mlp": 1.02863717,
      "epoch": 0.1634100883785246,
      "flos": 18041090584320.0,
      "grad_norm": 2.922306141867973,
      "language_loss": 0.76434886,
      "learning_rate": 3.816310806180139e-06,
      "loss": 0.78641498,
      "num_input_tokens_seen": 28641555,
      "step": 1359,
      "time_per_iteration": 2.538111686706543
    },
    {
      "auxiliary_loss_clip": 0.01182646,
      "auxiliary_loss_mlp": 0.01043283,
      "balance_loss_clip": 1.06138301,
      "balance_loss_mlp": 1.03168368,
      "epoch": 0.16353033126916372,
      "flos": 24572128055040.0,
      "grad_norm": 2.308644228792325,
      "language_loss": 0.80968416,
      "learning_rate": 3.81598456502135e-06,
      "loss": 0.83194339,
      "num_input_tokens_seen": 28661575,
      "step": 1360,
      "time_per_iteration": 2.5795791149139404
    },
    {
      "auxiliary_loss_clip": 0.01198807,
      "auxiliary_loss_mlp": 0.01033514,
      "balance_loss_clip": 1.06304502,
      "balance_loss_mlp": 1.02230287,
      "epoch": 0.1636505741598028,
      "flos": 19892895321600.0,
      "grad_norm": 2.1215410798561725,
      "language_loss": 0.87190276,
      "learning_rate": 3.8156580483791455e-06,
      "loss": 0.89422596,
      "num_input_tokens_seen": 28676765,
      "step": 1361,
      "time_per_iteration": 2.519425392150879
    },
    {
      "auxiliary_loss_clip": 0.0118842,
      "auxiliary_loss_mlp": 0.01045443,
      "balance_loss_clip": 1.06057954,
      "balance_loss_mlp": 1.03303885,
      "epoch": 0.16377081705044189,
      "flos": 28402611344640.0,
      "grad_norm": 2.057794412713308,
      "language_loss": 0.77427852,
      "learning_rate": 3.815331256303059e-06,
      "loss": 0.79661715,
      "num_input_tokens_seen": 28696795,
      "step": 1362,
      "time_per_iteration": 3.4982388019561768
    },
    {
      "auxiliary_loss_clip": 0.01133373,
      "auxiliary_loss_mlp": 0.0104535,
      "balance_loss_clip": 1.04847717,
      "balance_loss_mlp": 1.0336796,
      "epoch": 0.163891059941081,
      "flos": 21908059113600.0,
      "grad_norm": 2.2389670499305616,
      "language_loss": 0.77284187,
      "learning_rate": 3.815004188842665e-06,
      "loss": 0.7946291,
      "num_input_tokens_seen": 28714835,
      "step": 1363,
      "time_per_iteration": 3.4714009761810303
    },
    {
      "auxiliary_loss_clip": 0.01149419,
      "auxiliary_loss_mlp": 0.0103468,
      "balance_loss_clip": 1.04960465,
      "balance_loss_mlp": 1.02324796,
      "epoch": 0.16401130283172008,
      "flos": 26797619934720.0,
      "grad_norm": 1.6224268254131684,
      "language_loss": 0.7955572,
      "learning_rate": 3.814676846047578e-06,
      "loss": 0.81739825,
      "num_input_tokens_seen": 28735710,
      "step": 1364,
      "time_per_iteration": 2.620753288269043
    },
    {
      "auxiliary_loss_clip": 0.01153545,
      "auxiliary_loss_mlp": 0.01054524,
      "balance_loss_clip": 1.05324388,
      "balance_loss_mlp": 1.04345536,
      "epoch": 0.16413154572235916,
      "flos": 32997417160320.0,
      "grad_norm": 1.697171208943534,
      "language_loss": 0.69445318,
      "learning_rate": 3.8143492279674565e-06,
      "loss": 0.71653384,
      "num_input_tokens_seen": 28758405,
      "step": 1365,
      "time_per_iteration": 3.613297700881958
    },
    {
      "auxiliary_loss_clip": 0.01099474,
      "auxiliary_loss_mlp": 0.01003429,
      "balance_loss_clip": 1.03738427,
      "balance_loss_mlp": 1.00110483,
      "epoch": 0.16425178861299825,
      "flos": 40113622074240.0,
      "grad_norm": 0.8451925185922223,
      "language_loss": 0.58416939,
      "learning_rate": 3.8140213346519997e-06,
      "loss": 0.60519838,
      "num_input_tokens_seen": 28809000,
      "step": 1366,
      "time_per_iteration": 3.675506353378296
    },
    {
      "auxiliary_loss_clip": 0.01212319,
      "auxiliary_loss_mlp": 0.01037876,
      "balance_loss_clip": 1.06560576,
      "balance_loss_mlp": 1.02621722,
      "epoch": 0.16437203150363736,
      "flos": 25447486498560.0,
      "grad_norm": 1.7058058910728895,
      "language_loss": 0.76757586,
      "learning_rate": 3.813693166150948e-06,
      "loss": 0.79007775,
      "num_input_tokens_seen": 28829210,
      "step": 1367,
      "time_per_iteration": 2.5437140464782715
    },
    {
      "auxiliary_loss_clip": 0.01154984,
      "auxiliary_loss_mlp": 0.01043933,
      "balance_loss_clip": 1.0544889,
      "balance_loss_mlp": 1.03155291,
      "epoch": 0.16449227439427644,
      "flos": 23476888506240.0,
      "grad_norm": 3.8999561952318893,
      "language_loss": 0.8561126,
      "learning_rate": 3.813364722514086e-06,
      "loss": 0.87810183,
      "num_input_tokens_seen": 28847545,
      "step": 1368,
      "time_per_iteration": 2.589738607406616
    },
    {
      "auxiliary_loss_clip": 0.01137478,
      "auxiliary_loss_mlp": 0.01034161,
      "balance_loss_clip": 1.05121255,
      "balance_loss_mlp": 1.023265,
      "epoch": 0.16461251728491552,
      "flos": 13545217802880.0,
      "grad_norm": 2.2183712213810702,
      "language_loss": 0.80856794,
      "learning_rate": 3.8130360037912368e-06,
      "loss": 0.83028436,
      "num_input_tokens_seen": 28863990,
      "step": 1369,
      "time_per_iteration": 2.616884708404541
    },
    {
      "auxiliary_loss_clip": 0.01118569,
      "auxiliary_loss_mlp": 0.01032447,
      "balance_loss_clip": 1.04689217,
      "balance_loss_mlp": 1.02116919,
      "epoch": 0.16473276017555463,
      "flos": 23003298662400.0,
      "grad_norm": 2.580169461035191,
      "language_loss": 0.81711239,
      "learning_rate": 3.812707010032268e-06,
      "loss": 0.83862257,
      "num_input_tokens_seen": 28883045,
      "step": 1370,
      "time_per_iteration": 2.659796953201294
    },
    {
      "auxiliary_loss_clip": 0.01188175,
      "auxiliary_loss_mlp": 0.01040466,
      "balance_loss_clip": 1.06144476,
      "balance_loss_mlp": 1.02900982,
      "epoch": 0.16485300306619372,
      "flos": 24790680357120.0,
      "grad_norm": 1.6925137714772727,
      "language_loss": 0.79293716,
      "learning_rate": 3.8123777412870863e-06,
      "loss": 0.81522357,
      "num_input_tokens_seen": 28902545,
      "step": 1371,
      "time_per_iteration": 2.6310269832611084
    },
    {
      "auxiliary_loss_clip": 0.012092,
      "auxiliary_loss_mlp": 0.01037804,
      "balance_loss_clip": 1.06212199,
      "balance_loss_mlp": 1.02618074,
      "epoch": 0.1649732459568328,
      "flos": 21106497162240.0,
      "grad_norm": 1.8414370452626916,
      "language_loss": 0.77952099,
      "learning_rate": 3.812048197605643e-06,
      "loss": 0.80199099,
      "num_input_tokens_seen": 28921440,
      "step": 1372,
      "time_per_iteration": 2.494055986404419
    },
    {
      "auxiliary_loss_clip": 0.01213179,
      "auxiliary_loss_mlp": 0.01032306,
      "balance_loss_clip": 1.06033289,
      "balance_loss_mlp": 1.02126145,
      "epoch": 0.16509348884747188,
      "flos": 20266726118400.0,
      "grad_norm": 1.9264879344463604,
      "language_loss": 0.81264222,
      "learning_rate": 3.8117183790379277e-06,
      "loss": 0.83509707,
      "num_input_tokens_seen": 28939890,
      "step": 1373,
      "time_per_iteration": 2.4992353916168213
    },
    {
      "auxiliary_loss_clip": 0.01223429,
      "auxiliary_loss_mlp": 0.0102877,
      "balance_loss_clip": 1.06200361,
      "balance_loss_mlp": 1.01743901,
      "epoch": 0.165213731738111,
      "flos": 11035493602560.0,
      "grad_norm": 3.8857366751453735,
      "language_loss": 0.93888319,
      "learning_rate": 3.811388285633976e-06,
      "loss": 0.9614051,
      "num_input_tokens_seen": 28955875,
      "step": 1374,
      "time_per_iteration": 2.477545738220215
    },
    {
      "auxiliary_loss_clip": 0.0114414,
      "auxiliary_loss_mlp": 0.01039135,
      "balance_loss_clip": 1.05607843,
      "balance_loss_mlp": 1.02683854,
      "epoch": 0.16533397462875007,
      "flos": 29972051268480.0,
      "grad_norm": 2.1865701324664206,
      "language_loss": 0.6242668,
      "learning_rate": 3.811057917443861e-06,
      "loss": 0.64609957,
      "num_input_tokens_seen": 28975140,
      "step": 1375,
      "time_per_iteration": 2.7019596099853516
    },
    {
      "auxiliary_loss_clip": 0.01093577,
      "auxiliary_loss_mlp": 0.01002221,
      "balance_loss_clip": 1.05843079,
      "balance_loss_mlp": 0.999753,
      "epoch": 0.16545421751938916,
      "flos": 65556763027200.0,
      "grad_norm": 0.8644737925679137,
      "language_loss": 0.68265188,
      "learning_rate": 3.8107272745177e-06,
      "loss": 0.70360994,
      "num_input_tokens_seen": 29047470,
      "step": 1376,
      "time_per_iteration": 3.3748061656951904
    },
    {
      "auxiliary_loss_clip": 0.01198165,
      "auxiliary_loss_mlp": 0.01040488,
      "balance_loss_clip": 1.06160641,
      "balance_loss_mlp": 1.02841234,
      "epoch": 0.16557446041002827,
      "flos": 22492361652480.0,
      "grad_norm": 1.7474827901863335,
      "language_loss": 0.78901744,
      "learning_rate": 3.8103963569056513e-06,
      "loss": 0.81140399,
      "num_input_tokens_seen": 29066605,
      "step": 1377,
      "time_per_iteration": 2.5183510780334473
    },
    {
      "auxiliary_loss_clip": 0.01134328,
      "auxiliary_loss_mlp": 0.01036113,
      "balance_loss_clip": 1.05645013,
      "balance_loss_mlp": 1.02456784,
      "epoch": 0.16569470330066735,
      "flos": 24602723464320.0,
      "grad_norm": 1.6492039710021158,
      "language_loss": 0.88227069,
      "learning_rate": 3.8100651646579146e-06,
      "loss": 0.90397513,
      "num_input_tokens_seen": 29085815,
      "step": 1378,
      "time_per_iteration": 2.6872599124908447
    },
    {
      "auxiliary_loss_clip": 0.01141391,
      "auxiliary_loss_mlp": 0.01035571,
      "balance_loss_clip": 1.05785942,
      "balance_loss_mlp": 1.02394271,
      "epoch": 0.16581494619130643,
      "flos": 15006207588480.0,
      "grad_norm": 2.5091608159707137,
      "language_loss": 0.92482388,
      "learning_rate": 3.8097336978247317e-06,
      "loss": 0.94659352,
      "num_input_tokens_seen": 29102520,
      "step": 1379,
      "time_per_iteration": 2.582056999206543
    },
    {
      "auxiliary_loss_clip": 0.01191372,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 1.06074619,
      "balance_loss_mlp": 1.01943469,
      "epoch": 0.16593518908194552,
      "flos": 17420338719360.0,
      "grad_norm": 2.715472041468203,
      "language_loss": 0.88826847,
      "learning_rate": 3.8094019564563854e-06,
      "loss": 0.91050053,
      "num_input_tokens_seen": 29119450,
      "step": 1380,
      "time_per_iteration": 2.5045852661132812
    },
    {
      "auxiliary_loss_clip": 0.01145611,
      "auxiliary_loss_mlp": 0.01040843,
      "balance_loss_clip": 1.05225861,
      "balance_loss_mlp": 1.02885628,
      "epoch": 0.16605543197258463,
      "flos": 20412631163520.0,
      "grad_norm": 2.1785146485414573,
      "language_loss": 0.75309759,
      "learning_rate": 3.809069940603201e-06,
      "loss": 0.77496207,
      "num_input_tokens_seen": 29137405,
      "step": 1381,
      "time_per_iteration": 2.563157796859741
    },
    {
      "auxiliary_loss_clip": 0.01103932,
      "auxiliary_loss_mlp": 0.01036119,
      "balance_loss_clip": 1.04959822,
      "balance_loss_mlp": 1.02384603,
      "epoch": 0.1661756748632237,
      "flos": 14209745368320.0,
      "grad_norm": 3.2489556755997433,
      "language_loss": 0.78102803,
      "learning_rate": 3.8087376503155452e-06,
      "loss": 0.80242854,
      "num_input_tokens_seen": 29154890,
      "step": 1382,
      "time_per_iteration": 2.8879895210266113
    },
    {
      "auxiliary_loss_clip": 0.01080391,
      "auxiliary_loss_mlp": 0.01005525,
      "balance_loss_clip": 1.03575873,
      "balance_loss_mlp": 1.00324821,
      "epoch": 0.1662959177538628,
      "flos": 66080877350400.0,
      "grad_norm": 0.8998888196713615,
      "language_loss": 0.56232882,
      "learning_rate": 3.808405085643826e-06,
      "loss": 0.58318794,
      "num_input_tokens_seen": 29219770,
      "step": 1383,
      "time_per_iteration": 3.7439382076263428
    },
    {
      "auxiliary_loss_clip": 0.0119748,
      "auxiliary_loss_mlp": 0.01034898,
      "balance_loss_clip": 1.06099367,
      "balance_loss_mlp": 1.02359653,
      "epoch": 0.1664161606445019,
      "flos": 20740567357440.0,
      "grad_norm": 3.057825197406683,
      "language_loss": 0.8855902,
      "learning_rate": 3.8080722466384925e-06,
      "loss": 0.90791398,
      "num_input_tokens_seen": 29237620,
      "step": 1384,
      "time_per_iteration": 2.9080958366394043
    },
    {
      "auxiliary_loss_clip": 0.01166963,
      "auxiliary_loss_mlp": 0.01036407,
      "balance_loss_clip": 1.05396962,
      "balance_loss_mlp": 1.02437305,
      "epoch": 0.166536403535141,
      "flos": 25260930236160.0,
      "grad_norm": 1.999562267048828,
      "language_loss": 0.71195412,
      "learning_rate": 3.8077391333500376e-06,
      "loss": 0.73398787,
      "num_input_tokens_seen": 29256760,
      "step": 1385,
      "time_per_iteration": 2.633784055709839
    },
    {
      "auxiliary_loss_clip": 0.01189127,
      "auxiliary_loss_mlp": 0.01040581,
      "balance_loss_clip": 1.0608592,
      "balance_loss_mlp": 1.02981043,
      "epoch": 0.16665664642578007,
      "flos": 25447450584960.0,
      "grad_norm": 1.6651246003671454,
      "language_loss": 0.76499802,
      "learning_rate": 3.8074057458289934e-06,
      "loss": 0.78729504,
      "num_input_tokens_seen": 29277450,
      "step": 1386,
      "time_per_iteration": 2.5567233562469482
    },
    {
      "auxiliary_loss_clip": 0.01200357,
      "auxiliary_loss_mlp": 0.01040675,
      "balance_loss_clip": 1.061342,
      "balance_loss_mlp": 1.02781188,
      "epoch": 0.16677688931641918,
      "flos": 22200767043840.0,
      "grad_norm": 2.382641640546947,
      "language_loss": 0.82294559,
      "learning_rate": 3.807072084125934e-06,
      "loss": 0.84535587,
      "num_input_tokens_seen": 29299300,
      "step": 1387,
      "time_per_iteration": 3.2923052310943604
    },
    {
      "auxiliary_loss_clip": 0.01169662,
      "auxiliary_loss_mlp": 0.01035382,
      "balance_loss_clip": 1.05347395,
      "balance_loss_mlp": 1.02335978,
      "epoch": 0.16689713220705826,
      "flos": 16945958776320.0,
      "grad_norm": 2.859687731401405,
      "language_loss": 0.80697,
      "learning_rate": 3.806738148291477e-06,
      "loss": 0.82902044,
      "num_input_tokens_seen": 29316125,
      "step": 1388,
      "time_per_iteration": 2.541929006576538
    },
    {
      "auxiliary_loss_clip": 0.01188586,
      "auxiliary_loss_mlp": 0.01042563,
      "balance_loss_clip": 1.05838394,
      "balance_loss_mlp": 1.02931857,
      "epoch": 0.16701737509769735,
      "flos": 36244423923840.0,
      "grad_norm": 1.9616460174532204,
      "language_loss": 0.71032286,
      "learning_rate": 3.8064039383762793e-06,
      "loss": 0.73263437,
      "num_input_tokens_seen": 29338490,
      "step": 1389,
      "time_per_iteration": 3.384282112121582
    },
    {
      "auxiliary_loss_clip": 0.01197387,
      "auxiliary_loss_mlp": 0.00765616,
      "balance_loss_clip": 1.0622164,
      "balance_loss_mlp": 1.00068474,
      "epoch": 0.16713761798833643,
      "flos": 23258659426560.0,
      "grad_norm": 2.153642442408102,
      "language_loss": 0.77412015,
      "learning_rate": 3.8060694544310396e-06,
      "loss": 0.79375017,
      "num_input_tokens_seen": 29357000,
      "step": 1390,
      "time_per_iteration": 2.558734893798828
    },
    {
      "auxiliary_loss_clip": 0.01124546,
      "auxiliary_loss_mlp": 0.01040054,
      "balance_loss_clip": 1.04808617,
      "balance_loss_mlp": 1.02720881,
      "epoch": 0.16725786087897554,
      "flos": 25302515207040.0,
      "grad_norm": 1.7171126245917072,
      "language_loss": 0.78611958,
      "learning_rate": 3.8057346965065006e-06,
      "loss": 0.8077656,
      "num_input_tokens_seen": 29378230,
      "step": 1391,
      "time_per_iteration": 3.450310468673706
    },
    {
      "auxiliary_loss_clip": 0.01195432,
      "auxiliary_loss_mlp": 0.010348,
      "balance_loss_clip": 1.06211376,
      "balance_loss_mlp": 1.02364206,
      "epoch": 0.16737810376961462,
      "flos": 31831541516160.0,
      "grad_norm": 2.7162959377212745,
      "language_loss": 0.84746778,
      "learning_rate": 3.805399664653443e-06,
      "loss": 0.86977011,
      "num_input_tokens_seen": 29400370,
      "step": 1392,
      "time_per_iteration": 2.6319353580474854
    },
    {
      "auxiliary_loss_clip": 0.01229662,
      "auxiliary_loss_mlp": 0.00765837,
      "balance_loss_clip": 1.06483793,
      "balance_loss_mlp": 1.00055599,
      "epoch": 0.1674983466602537,
      "flos": 27961843553280.0,
      "grad_norm": 2.4567049617619854,
      "language_loss": 0.74684131,
      "learning_rate": 3.805064358922692e-06,
      "loss": 0.76679635,
      "num_input_tokens_seen": 29418660,
      "step": 1393,
      "time_per_iteration": 2.4861271381378174
    },
    {
      "auxiliary_loss_clip": 0.01165943,
      "auxiliary_loss_mlp": 0.01039761,
      "balance_loss_clip": 1.05585241,
      "balance_loss_mlp": 1.02669525,
      "epoch": 0.16761858955089282,
      "flos": 21762656858880.0,
      "grad_norm": 1.7763907747926737,
      "language_loss": 0.81287086,
      "learning_rate": 3.8047287793651136e-06,
      "loss": 0.83492792,
      "num_input_tokens_seen": 29440105,
      "step": 1394,
      "time_per_iteration": 2.6196420192718506
    },
    {
      "auxiliary_loss_clip": 0.01138408,
      "auxiliary_loss_mlp": 0.01037791,
      "balance_loss_clip": 1.05441475,
      "balance_loss_mlp": 1.02653134,
      "epoch": 0.1677388324415319,
      "flos": 23805507058560.0,
      "grad_norm": 1.9094429032249765,
      "language_loss": 0.88597822,
      "learning_rate": 3.8043929260316137e-06,
      "loss": 0.90774024,
      "num_input_tokens_seen": 29458260,
      "step": 1395,
      "time_per_iteration": 2.6308863162994385
    },
    {
      "auxiliary_loss_clip": 0.01161021,
      "auxiliary_loss_mlp": 0.01037861,
      "balance_loss_clip": 1.05503798,
      "balance_loss_mlp": 1.02579689,
      "epoch": 0.16785907533217098,
      "flos": 20558859431040.0,
      "grad_norm": 2.0661672077860227,
      "language_loss": 0.83257341,
      "learning_rate": 3.8040567989731417e-06,
      "loss": 0.85456222,
      "num_input_tokens_seen": 29476205,
      "step": 1396,
      "time_per_iteration": 2.584303617477417
    },
    {
      "auxiliary_loss_clip": 0.01163183,
      "auxiliary_loss_mlp": 0.01034369,
      "balance_loss_clip": 1.06076622,
      "balance_loss_mlp": 1.02396798,
      "epoch": 0.16797931822281006,
      "flos": 15669657745920.0,
      "grad_norm": 2.0588214071306963,
      "language_loss": 0.79652989,
      "learning_rate": 3.8037203982406876e-06,
      "loss": 0.81850541,
      "num_input_tokens_seen": 29494370,
      "step": 1397,
      "time_per_iteration": 2.564234495162964
    },
    {
      "auxiliary_loss_clip": 0.01163723,
      "auxiliary_loss_mlp": 0.01033629,
      "balance_loss_clip": 1.05885601,
      "balance_loss_mlp": 1.02142811,
      "epoch": 0.16809956111344918,
      "flos": 16541101607040.0,
      "grad_norm": 2.111617611440656,
      "language_loss": 0.7308445,
      "learning_rate": 3.8033837238852835e-06,
      "loss": 0.75281799,
      "num_input_tokens_seen": 29511070,
      "step": 1398,
      "time_per_iteration": 2.578726291656494
    },
    {
      "auxiliary_loss_clip": 0.01169603,
      "auxiliary_loss_mlp": 0.01037965,
      "balance_loss_clip": 1.05704522,
      "balance_loss_mlp": 1.0263238,
      "epoch": 0.16821980400408826,
      "flos": 23258084808960.0,
      "grad_norm": 1.8555104886142608,
      "language_loss": 0.69373977,
      "learning_rate": 3.8030467759580017e-06,
      "loss": 0.71581542,
      "num_input_tokens_seen": 29531990,
      "step": 1399,
      "time_per_iteration": 2.568063497543335
    },
    {
      "auxiliary_loss_clip": 0.01210046,
      "auxiliary_loss_mlp": 0.01040516,
      "balance_loss_clip": 1.06276906,
      "balance_loss_mlp": 1.02785575,
      "epoch": 0.16834004689472734,
      "flos": 20774754126720.0,
      "grad_norm": 2.0774045128402268,
      "language_loss": 0.86841464,
      "learning_rate": 3.802709554509958e-06,
      "loss": 0.89092022,
      "num_input_tokens_seen": 29549790,
      "step": 1400,
      "time_per_iteration": 2.4970812797546387
    },
    {
      "auxiliary_loss_clip": 0.0117909,
      "auxiliary_loss_mlp": 0.01042375,
      "balance_loss_clip": 1.05485475,
      "balance_loss_mlp": 1.03039992,
      "epoch": 0.16846028978536645,
      "flos": 26687302289280.0,
      "grad_norm": 1.7569377352040358,
      "language_loss": 0.79119444,
      "learning_rate": 3.8023720595923083e-06,
      "loss": 0.81340909,
      "num_input_tokens_seen": 29569045,
      "step": 1401,
      "time_per_iteration": 2.5818450450897217
    },
    {
      "auxiliary_loss_clip": 0.01146616,
      "auxiliary_loss_mlp": 0.01037273,
      "balance_loss_clip": 1.05346715,
      "balance_loss_mlp": 1.02554858,
      "epoch": 0.16858053267600553,
      "flos": 18843298980480.0,
      "grad_norm": 2.700462633185866,
      "language_loss": 0.87599903,
      "learning_rate": 3.80203429125625e-06,
      "loss": 0.89783788,
      "num_input_tokens_seen": 29587220,
      "step": 1402,
      "time_per_iteration": 2.6314780712127686
    },
    {
      "auxiliary_loss_clip": 0.01149256,
      "auxiliary_loss_mlp": 0.00765008,
      "balance_loss_clip": 1.05324388,
      "balance_loss_mlp": 1.00064862,
      "epoch": 0.16870077556664462,
      "flos": 27744548227200.0,
      "grad_norm": 2.36310771453595,
      "language_loss": 0.70175999,
      "learning_rate": 3.8016962495530225e-06,
      "loss": 0.72090268,
      "num_input_tokens_seen": 29606410,
      "step": 1403,
      "time_per_iteration": 2.6358044147491455
    },
    {
      "auxiliary_loss_clip": 0.01201391,
      "auxiliary_loss_mlp": 0.01035683,
      "balance_loss_clip": 1.05674171,
      "balance_loss_mlp": 1.023875,
      "epoch": 0.1688210184572837,
      "flos": 13730768484480.0,
      "grad_norm": 2.9949529944218773,
      "language_loss": 0.76509017,
      "learning_rate": 3.8013579345339063e-06,
      "loss": 0.78746092,
      "num_input_tokens_seen": 29621275,
      "step": 1404,
      "time_per_iteration": 2.454556703567505
    },
    {
      "auxiliary_loss_clip": 0.01199375,
      "auxiliary_loss_mlp": 0.0103623,
      "balance_loss_clip": 1.06200218,
      "balance_loss_mlp": 1.02427948,
      "epoch": 0.1689412613479228,
      "flos": 26468785900800.0,
      "grad_norm": 2.1950198284468994,
      "language_loss": 0.69703358,
      "learning_rate": 3.801019346250224e-06,
      "loss": 0.71938962,
      "num_input_tokens_seen": 29641420,
      "step": 1405,
      "time_per_iteration": 2.54946231842041
    },
    {
      "auxiliary_loss_clip": 0.01208166,
      "auxiliary_loss_mlp": 0.01033513,
      "balance_loss_clip": 1.05919409,
      "balance_loss_mlp": 1.0223248,
      "epoch": 0.1690615042385619,
      "flos": 21138852337920.0,
      "grad_norm": 2.531175060545712,
      "language_loss": 0.83972931,
      "learning_rate": 3.8006804847533395e-06,
      "loss": 0.86214602,
      "num_input_tokens_seen": 29660935,
      "step": 1406,
      "time_per_iteration": 2.5134143829345703
    },
    {
      "auxiliary_loss_clip": 0.01193497,
      "auxiliary_loss_mlp": 0.0103695,
      "balance_loss_clip": 1.05400848,
      "balance_loss_mlp": 1.02390254,
      "epoch": 0.16918174712920098,
      "flos": 20849340718080.0,
      "grad_norm": 2.2722340606103386,
      "language_loss": 0.86053926,
      "learning_rate": 3.8003413500946556e-06,
      "loss": 0.88284373,
      "num_input_tokens_seen": 29681045,
      "step": 1407,
      "time_per_iteration": 2.5093016624450684
    },
    {
      "auxiliary_loss_clip": 0.01215693,
      "auxiliary_loss_mlp": 0.01044581,
      "balance_loss_clip": 1.06256592,
      "balance_loss_mlp": 1.03242159,
      "epoch": 0.1693019900198401,
      "flos": 16983270028800.0,
      "grad_norm": 2.678985915572635,
      "language_loss": 0.83525592,
      "learning_rate": 3.8000019423256216e-06,
      "loss": 0.85785866,
      "num_input_tokens_seen": 29698810,
      "step": 1408,
      "time_per_iteration": 2.4584240913391113
    },
    {
      "auxiliary_loss_clip": 0.01123573,
      "auxiliary_loss_mlp": 0.01036257,
      "balance_loss_clip": 1.05203128,
      "balance_loss_mlp": 1.02509928,
      "epoch": 0.16942223291047917,
      "flos": 26796901662720.0,
      "grad_norm": 1.8248037587803005,
      "language_loss": 0.87962711,
      "learning_rate": 3.7996622614977234e-06,
      "loss": 0.90122539,
      "num_input_tokens_seen": 29720000,
      "step": 1409,
      "time_per_iteration": 2.6627197265625
    },
    {
      "auxiliary_loss_clip": 0.01177568,
      "auxiliary_loss_mlp": 0.01043941,
      "balance_loss_clip": 1.0575316,
      "balance_loss_mlp": 1.03157353,
      "epoch": 0.16954247580111825,
      "flos": 18583700411520.0,
      "grad_norm": 2.330016081706269,
      "language_loss": 0.79345071,
      "learning_rate": 3.799322307662492e-06,
      "loss": 0.81566584,
      "num_input_tokens_seen": 29737820,
      "step": 1410,
      "time_per_iteration": 2.5541558265686035
    },
    {
      "auxiliary_loss_clip": 0.01123532,
      "auxiliary_loss_mlp": 0.01041143,
      "balance_loss_clip": 1.04911411,
      "balance_loss_mlp": 1.02793467,
      "epoch": 0.16966271869175734,
      "flos": 13983651210240.0,
      "grad_norm": 2.3290144629781726,
      "language_loss": 0.84072089,
      "learning_rate": 3.798982080871496e-06,
      "loss": 0.86236763,
      "num_input_tokens_seen": 29752960,
      "step": 1411,
      "time_per_iteration": 2.902001142501831
    },
    {
      "auxiliary_loss_clip": 0.01157596,
      "auxiliary_loss_mlp": 0.01044018,
      "balance_loss_clip": 1.05206347,
      "balance_loss_mlp": 1.03086936,
      "epoch": 0.16978296158239645,
      "flos": 37487328284160.0,
      "grad_norm": 1.868420112407675,
      "language_loss": 0.67890859,
      "learning_rate": 3.798641581176349e-06,
      "loss": 0.70092469,
      "num_input_tokens_seen": 29775240,
      "step": 1412,
      "time_per_iteration": 3.042672634124756
    },
    {
      "auxiliary_loss_clip": 0.01157072,
      "auxiliary_loss_mlp": 0.01041417,
      "balance_loss_clip": 1.04989493,
      "balance_loss_mlp": 1.02897143,
      "epoch": 0.16990320447303553,
      "flos": 28328958506880.0,
      "grad_norm": 2.240041108437384,
      "language_loss": 0.74582219,
      "learning_rate": 3.7983008086287044e-06,
      "loss": 0.76780707,
      "num_input_tokens_seen": 29796560,
      "step": 1413,
      "time_per_iteration": 3.4316537380218506
    },
    {
      "auxiliary_loss_clip": 0.01203383,
      "auxiliary_loss_mlp": 0.01038071,
      "balance_loss_clip": 1.06062186,
      "balance_loss_mlp": 1.02508271,
      "epoch": 0.1700234473636746,
      "flos": 20188189031040.0,
      "grad_norm": 2.1397379371413017,
      "language_loss": 0.79565334,
      "learning_rate": 3.797959763280257e-06,
      "loss": 0.81806791,
      "num_input_tokens_seen": 29815245,
      "step": 1414,
      "time_per_iteration": 3.7177298069000244
    },
    {
      "auxiliary_loss_clip": 0.01131429,
      "auxiliary_loss_mlp": 0.01039333,
      "balance_loss_clip": 1.05002511,
      "balance_loss_mlp": 1.02669692,
      "epoch": 0.17014369025431372,
      "flos": 24858658846080.0,
      "grad_norm": 1.8758532572122817,
      "language_loss": 0.7890141,
      "learning_rate": 3.797618445182743e-06,
      "loss": 0.81072176,
      "num_input_tokens_seen": 29836640,
      "step": 1415,
      "time_per_iteration": 2.694269895553589
    },
    {
      "auxiliary_loss_clip": 0.0116626,
      "auxiliary_loss_mlp": 0.0104258,
      "balance_loss_clip": 1.05759668,
      "balance_loss_mlp": 1.03129661,
      "epoch": 0.1702639331449528,
      "flos": 16467233287680.0,
      "grad_norm": 2.197586641445347,
      "language_loss": 0.84712571,
      "learning_rate": 3.79727685438794e-06,
      "loss": 0.86921412,
      "num_input_tokens_seen": 29850830,
      "step": 1416,
      "time_per_iteration": 3.3104615211486816
    },
    {
      "auxiliary_loss_clip": 0.011247,
      "auxiliary_loss_mlp": 0.01002049,
      "balance_loss_clip": 1.05602241,
      "balance_loss_mlp": 0.99945033,
      "epoch": 0.1703841760355919,
      "flos": 52508870979840.0,
      "grad_norm": 0.8346193097083773,
      "language_loss": 0.61640269,
      "learning_rate": 3.796934990947667e-06,
      "loss": 0.63767016,
      "num_input_tokens_seen": 29912515,
      "step": 1417,
      "time_per_iteration": 3.9361679553985596
    },
    {
      "auxiliary_loss_clip": 0.01117731,
      "auxiliary_loss_mlp": 0.01004029,
      "balance_loss_clip": 1.03908861,
      "balance_loss_mlp": 1.00166881,
      "epoch": 0.170504418926231,
      "flos": 49370637576960.0,
      "grad_norm": 0.8764906698840612,
      "language_loss": 0.6245054,
      "learning_rate": 3.7965928549137854e-06,
      "loss": 0.64572299,
      "num_input_tokens_seen": 29969330,
      "step": 1418,
      "time_per_iteration": 3.033372402191162
    },
    {
      "auxiliary_loss_clip": 0.01214546,
      "auxiliary_loss_mlp": 0.01045148,
      "balance_loss_clip": 1.06298769,
      "balance_loss_mlp": 1.03071141,
      "epoch": 0.17062466181687008,
      "flos": 25849219184640.0,
      "grad_norm": 2.461780066296096,
      "language_loss": 0.77732491,
      "learning_rate": 3.7962504463381953e-06,
      "loss": 0.79992187,
      "num_input_tokens_seen": 29990820,
      "step": 1419,
      "time_per_iteration": 2.5438380241394043
    },
    {
      "auxiliary_loss_clip": 0.01193024,
      "auxiliary_loss_mlp": 0.01048507,
      "balance_loss_clip": 1.06093621,
      "balance_loss_mlp": 1.03492892,
      "epoch": 0.17074490470750917,
      "flos": 20960412549120.0,
      "grad_norm": 2.229513837667959,
      "language_loss": 0.78812194,
      "learning_rate": 3.7959077652728412e-06,
      "loss": 0.81053722,
      "num_input_tokens_seen": 30009275,
      "step": 1420,
      "time_per_iteration": 2.534440279006958
    },
    {
      "auxiliary_loss_clip": 0.01086559,
      "auxiliary_loss_mlp": 0.01040431,
      "balance_loss_clip": 1.03862524,
      "balance_loss_mlp": 1.02805662,
      "epoch": 0.17086514759814825,
      "flos": 20959766104320.0,
      "grad_norm": 2.182509431012162,
      "language_loss": 0.77337074,
      "learning_rate": 3.795564811769707e-06,
      "loss": 0.7946406,
      "num_input_tokens_seen": 30027630,
      "step": 1421,
      "time_per_iteration": 2.9759345054626465
    },
    {
      "auxiliary_loss_clip": 0.01198344,
      "auxiliary_loss_mlp": 0.01035454,
      "balance_loss_clip": 1.05929363,
      "balance_loss_mlp": 1.02246058,
      "epoch": 0.17098539048878736,
      "flos": 28474073452800.0,
      "grad_norm": 2.0292290989850805,
      "language_loss": 0.77966034,
      "learning_rate": 3.795221585880818e-06,
      "loss": 0.80199838,
      "num_input_tokens_seen": 30048310,
      "step": 1422,
      "time_per_iteration": 2.8062140941619873
    },
    {
      "auxiliary_loss_clip": 0.01167762,
      "auxiliary_loss_mlp": 0.01037062,
      "balance_loss_clip": 1.05810761,
      "balance_loss_mlp": 1.02501011,
      "epoch": 0.17110563337942644,
      "flos": 16290014561280.0,
      "grad_norm": 3.4559272447113503,
      "language_loss": 0.91499496,
      "learning_rate": 3.794878087658242e-06,
      "loss": 0.93704319,
      "num_input_tokens_seen": 30066080,
      "step": 1423,
      "time_per_iteration": 2.5624938011169434
    },
    {
      "auxiliary_loss_clip": 0.01157446,
      "auxiliary_loss_mlp": 0.01043278,
      "balance_loss_clip": 1.05398512,
      "balance_loss_mlp": 1.03102374,
      "epoch": 0.17122587627006552,
      "flos": 29674207693440.0,
      "grad_norm": 1.7590895145794434,
      "language_loss": 0.7861979,
      "learning_rate": 3.7945343171540873e-06,
      "loss": 0.80820525,
      "num_input_tokens_seen": 30086955,
      "step": 1424,
      "time_per_iteration": 2.646812915802002
    },
    {
      "auxiliary_loss_clip": 0.01143512,
      "auxiliary_loss_mlp": 0.01041746,
      "balance_loss_clip": 1.05267668,
      "balance_loss_mlp": 1.02917552,
      "epoch": 0.17134611916070464,
      "flos": 25338389915520.0,
      "grad_norm": 2.169714477552452,
      "language_loss": 0.78804892,
      "learning_rate": 3.7941902744205033e-06,
      "loss": 0.80990148,
      "num_input_tokens_seen": 30107990,
      "step": 1425,
      "time_per_iteration": 2.657409906387329
    },
    {
      "auxiliary_loss_clip": 0.01201438,
      "auxiliary_loss_mlp": 0.01039074,
      "balance_loss_clip": 1.06115758,
      "balance_loss_mlp": 1.02520406,
      "epoch": 0.17146636205134372,
      "flos": 13953845900160.0,
      "grad_norm": 1.898607769667348,
      "language_loss": 0.83073485,
      "learning_rate": 3.7938459595096817e-06,
      "loss": 0.85314,
      "num_input_tokens_seen": 30126535,
      "step": 1426,
      "time_per_iteration": 2.484649419784546
    },
    {
      "auxiliary_loss_clip": 0.01216666,
      "auxiliary_loss_mlp": 0.01042707,
      "balance_loss_clip": 1.06223106,
      "balance_loss_mlp": 1.02863407,
      "epoch": 0.1715866049419828,
      "flos": 23915214172800.0,
      "grad_norm": 2.0223392228143053,
      "language_loss": 0.86296314,
      "learning_rate": 3.7935013724738545e-06,
      "loss": 0.88555688,
      "num_input_tokens_seen": 30147035,
      "step": 1427,
      "time_per_iteration": 2.514620065689087
    },
    {
      "auxiliary_loss_clip": 0.01181842,
      "auxiliary_loss_mlp": 0.01040399,
      "balance_loss_clip": 1.05330634,
      "balance_loss_mlp": 1.02906191,
      "epoch": 0.17170684783262188,
      "flos": 22709369669760.0,
      "grad_norm": 1.9791233447936663,
      "language_loss": 0.78002512,
      "learning_rate": 3.7931565133652945e-06,
      "loss": 0.80224752,
      "num_input_tokens_seen": 30167110,
      "step": 1428,
      "time_per_iteration": 2.529268741607666
    },
    {
      "auxiliary_loss_clip": 0.01185179,
      "auxiliary_loss_mlp": 0.01045352,
      "balance_loss_clip": 1.05578208,
      "balance_loss_mlp": 1.0322392,
      "epoch": 0.171827090723261,
      "flos": 26613290315520.0,
      "grad_norm": 2.1521226316056756,
      "language_loss": 0.68183887,
      "learning_rate": 3.792811382236317e-06,
      "loss": 0.70414424,
      "num_input_tokens_seen": 30185620,
      "step": 1429,
      "time_per_iteration": 2.593625783920288
    },
    {
      "auxiliary_loss_clip": 0.01150437,
      "auxiliary_loss_mlp": 0.01046717,
      "balance_loss_clip": 1.05718946,
      "balance_loss_mlp": 1.03338969,
      "epoch": 0.17194733361390008,
      "flos": 28148507556480.0,
      "grad_norm": 1.9628374120575163,
      "language_loss": 0.78224123,
      "learning_rate": 3.792465979139279e-06,
      "loss": 0.80421269,
      "num_input_tokens_seen": 30208225,
      "step": 1430,
      "time_per_iteration": 2.6679186820983887
    },
    {
      "auxiliary_loss_clip": 0.01097561,
      "auxiliary_loss_mlp": 0.01006179,
      "balance_loss_clip": 1.03407145,
      "balance_loss_mlp": 1.00350916,
      "epoch": 0.17206757650453916,
      "flos": 65530689753600.0,
      "grad_norm": 0.9239335340639511,
      "language_loss": 0.65671265,
      "learning_rate": 3.792120304126576e-06,
      "loss": 0.67775011,
      "num_input_tokens_seen": 30271600,
      "step": 1431,
      "time_per_iteration": 3.1951286792755127
    },
    {
      "auxiliary_loss_clip": 0.01162715,
      "auxiliary_loss_mlp": 0.0103837,
      "balance_loss_clip": 1.0556612,
      "balance_loss_mlp": 1.02709889,
      "epoch": 0.17218781939517827,
      "flos": 22273486128000.0,
      "grad_norm": 1.9622002085763552,
      "language_loss": 0.83709264,
      "learning_rate": 3.791774357250649e-06,
      "loss": 0.85910344,
      "num_input_tokens_seen": 30290430,
      "step": 1432,
      "time_per_iteration": 2.5914671421051025
    },
    {
      "auxiliary_loss_clip": 0.01193055,
      "auxiliary_loss_mlp": 0.00765719,
      "balance_loss_clip": 1.05971992,
      "balance_loss_mlp": 1.00073409,
      "epoch": 0.17230806228581735,
      "flos": 14137313592960.0,
      "grad_norm": 3.1205189591827764,
      "language_loss": 0.7905556,
      "learning_rate": 3.7914281385639757e-06,
      "loss": 0.81014335,
      "num_input_tokens_seen": 30308305,
      "step": 1433,
      "time_per_iteration": 2.484269857406616
    },
    {
      "auxiliary_loss_clip": 0.01195074,
      "auxiliary_loss_mlp": 0.0104269,
      "balance_loss_clip": 1.05574954,
      "balance_loss_mlp": 1.03160942,
      "epoch": 0.17242830517645644,
      "flos": 20704836303360.0,
      "grad_norm": 1.9165778678068324,
      "language_loss": 0.79417396,
      "learning_rate": 3.7910816481190784e-06,
      "loss": 0.81655157,
      "num_input_tokens_seen": 30328120,
      "step": 1434,
      "time_per_iteration": 2.5041604042053223
    },
    {
      "auxiliary_loss_clip": 0.01154357,
      "auxiliary_loss_mlp": 0.01040182,
      "balance_loss_clip": 1.05288291,
      "balance_loss_mlp": 1.02783799,
      "epoch": 0.17254854806709552,
      "flos": 30774582887040.0,
      "grad_norm": 2.1678525824522663,
      "language_loss": 0.74720967,
      "learning_rate": 3.7907348859685193e-06,
      "loss": 0.76915509,
      "num_input_tokens_seen": 30349825,
      "step": 1435,
      "time_per_iteration": 2.671630382537842
    },
    {
      "auxiliary_loss_clip": 0.0118496,
      "auxiliary_loss_mlp": 0.01038108,
      "balance_loss_clip": 1.06060326,
      "balance_loss_mlp": 1.02595472,
      "epoch": 0.17266879095773463,
      "flos": 26614726859520.0,
      "grad_norm": 3.0962641071784116,
      "language_loss": 0.80690128,
      "learning_rate": 3.790387852164902e-06,
      "loss": 0.82913196,
      "num_input_tokens_seen": 30370555,
      "step": 1436,
      "time_per_iteration": 2.619539737701416
    },
    {
      "auxiliary_loss_clip": 0.0116841,
      "auxiliary_loss_mlp": 0.01040655,
      "balance_loss_clip": 1.05560875,
      "balance_loss_mlp": 1.02891874,
      "epoch": 0.1727890338483737,
      "flos": 20266295155200.0,
      "grad_norm": 1.8261583290548944,
      "language_loss": 0.76582754,
      "learning_rate": 3.7900405467608707e-06,
      "loss": 0.78791815,
      "num_input_tokens_seen": 30390100,
      "step": 1437,
      "time_per_iteration": 2.570369243621826
    },
    {
      "auxiliary_loss_clip": 0.01152639,
      "auxiliary_loss_mlp": 0.00766647,
      "balance_loss_clip": 1.05232763,
      "balance_loss_mlp": 1.00057161,
      "epoch": 0.1729092767390128,
      "flos": 18179812909440.0,
      "grad_norm": 3.188324915925543,
      "language_loss": 0.79134041,
      "learning_rate": 3.7896929698091114e-06,
      "loss": 0.81053329,
      "num_input_tokens_seen": 30402915,
      "step": 1438,
      "time_per_iteration": 3.3672475814819336
    },
    {
      "auxiliary_loss_clip": 0.01213688,
      "auxiliary_loss_mlp": 0.01048793,
      "balance_loss_clip": 1.06354678,
      "balance_loss_mlp": 1.03575766,
      "epoch": 0.1730295196296519,
      "flos": 26759518583040.0,
      "grad_norm": 2.6991281718631965,
      "language_loss": 0.68220013,
      "learning_rate": 3.7893451213623518e-06,
      "loss": 0.70482492,
      "num_input_tokens_seen": 30420145,
      "step": 1439,
      "time_per_iteration": 2.538511276245117
    },
    {
      "auxiliary_loss_clip": 0.01231962,
      "auxiliary_loss_mlp": 0.01046448,
      "balance_loss_clip": 1.06719768,
      "balance_loss_mlp": 1.03391957,
      "epoch": 0.173149762520291,
      "flos": 23842531002240.0,
      "grad_norm": 2.1664284874854673,
      "language_loss": 0.82086539,
      "learning_rate": 3.7889970014733606e-06,
      "loss": 0.84364951,
      "num_input_tokens_seen": 30439250,
      "step": 1440,
      "time_per_iteration": 2.521919012069702
    },
    {
      "auxiliary_loss_clip": 0.01193939,
      "auxiliary_loss_mlp": 0.00765834,
      "balance_loss_clip": 1.06016755,
      "balance_loss_mlp": 1.00075769,
      "epoch": 0.17327000541093007,
      "flos": 23368186972800.0,
      "grad_norm": 1.7419056376644,
      "language_loss": 0.77768469,
      "learning_rate": 3.7886486101949463e-06,
      "loss": 0.79728246,
      "num_input_tokens_seen": 30460430,
      "step": 1441,
      "time_per_iteration": 3.9006237983703613
    },
    {
      "auxiliary_loss_clip": 0.01135918,
      "auxiliary_loss_mlp": 0.01032332,
      "balance_loss_clip": 1.05508709,
      "balance_loss_mlp": 1.02040529,
      "epoch": 0.17339024830156918,
      "flos": 18221290139520.0,
      "grad_norm": 3.1806868278912903,
      "language_loss": 0.87742996,
      "learning_rate": 3.7882999475799594e-06,
      "loss": 0.89911246,
      "num_input_tokens_seen": 30478465,
      "step": 1442,
      "time_per_iteration": 3.3290467262268066
    },
    {
      "auxiliary_loss_clip": 0.01161433,
      "auxiliary_loss_mlp": 0.01044842,
      "balance_loss_clip": 1.0561192,
      "balance_loss_mlp": 1.03429246,
      "epoch": 0.17351049119220827,
      "flos": 23332024955520.0,
      "grad_norm": 1.8873433285601238,
      "language_loss": 0.8178395,
      "learning_rate": 3.787951013681293e-06,
      "loss": 0.83990228,
      "num_input_tokens_seen": 30496510,
      "step": 1443,
      "time_per_iteration": 3.3537538051605225
    },
    {
      "auxiliary_loss_clip": 0.0120357,
      "auxiliary_loss_mlp": 0.01046542,
      "balance_loss_clip": 1.06153631,
      "balance_loss_mlp": 1.03361988,
      "epoch": 0.17363073408284735,
      "flos": 23803495896960.0,
      "grad_norm": 2.0385706692729766,
      "language_loss": 0.77763975,
      "learning_rate": 3.787601808551879e-06,
      "loss": 0.80014086,
      "num_input_tokens_seen": 30516325,
      "step": 1444,
      "time_per_iteration": 2.5389909744262695
    },
    {
      "auxiliary_loss_clip": 0.01178354,
      "auxiliary_loss_mlp": 0.01044456,
      "balance_loss_clip": 1.05700552,
      "balance_loss_mlp": 1.03211832,
      "epoch": 0.17375097697348643,
      "flos": 18515290959360.0,
      "grad_norm": 2.5615691516207284,
      "language_loss": 0.8413136,
      "learning_rate": 3.7872523322446926e-06,
      "loss": 0.86354172,
      "num_input_tokens_seen": 30535210,
      "step": 1445,
      "time_per_iteration": 2.5025150775909424
    },
    {
      "auxiliary_loss_clip": 0.011491,
      "auxiliary_loss_mlp": 0.01034817,
      "balance_loss_clip": 1.05107868,
      "balance_loss_mlp": 1.02176988,
      "epoch": 0.17387121986412554,
      "flos": 38877897456000.0,
      "grad_norm": 1.807312148492903,
      "language_loss": 0.60452211,
      "learning_rate": 3.7869025848127478e-06,
      "loss": 0.62636125,
      "num_input_tokens_seen": 30559405,
      "step": 1446,
      "time_per_iteration": 2.751358985900879
    },
    {
      "auxiliary_loss_clip": 0.0119354,
      "auxiliary_loss_mlp": 0.01044286,
      "balance_loss_clip": 1.06165087,
      "balance_loss_mlp": 1.03248405,
      "epoch": 0.17399146275476463,
      "flos": 20375714960640.0,
      "grad_norm": 2.5329634446441016,
      "language_loss": 0.81005669,
      "learning_rate": 3.786552566309102e-06,
      "loss": 0.83243495,
      "num_input_tokens_seen": 30577615,
      "step": 1447,
      "time_per_iteration": 2.509758472442627
    },
    {
      "auxiliary_loss_clip": 0.01197088,
      "auxiliary_loss_mlp": 0.01044363,
      "balance_loss_clip": 1.06114352,
      "balance_loss_mlp": 1.03229904,
      "epoch": 0.1741117056454037,
      "flos": 19164339763200.0,
      "grad_norm": 2.4055695112523394,
      "language_loss": 0.86261433,
      "learning_rate": 3.7862022767868517e-06,
      "loss": 0.88502884,
      "num_input_tokens_seen": 30595205,
      "step": 1448,
      "time_per_iteration": 2.5215635299682617
    },
    {
      "auxiliary_loss_clip": 0.01182366,
      "auxiliary_loss_mlp": 0.01043567,
      "balance_loss_clip": 1.0615474,
      "balance_loss_mlp": 1.0316695,
      "epoch": 0.17423194853604282,
      "flos": 25374300537600.0,
      "grad_norm": 2.2068751822289996,
      "language_loss": 0.84672964,
      "learning_rate": 3.7858517162991367e-06,
      "loss": 0.86898899,
      "num_input_tokens_seen": 30615280,
      "step": 1449,
      "time_per_iteration": 2.574272632598877
    },
    {
      "auxiliary_loss_clip": 0.01116011,
      "auxiliary_loss_mlp": 0.01039748,
      "balance_loss_clip": 1.0508455,
      "balance_loss_mlp": 1.02847099,
      "epoch": 0.1743521914266819,
      "flos": 25191874339200.0,
      "grad_norm": 2.435134963934152,
      "language_loss": 0.60355079,
      "learning_rate": 3.7855008848991363e-06,
      "loss": 0.62510836,
      "num_input_tokens_seen": 30633485,
      "step": 1450,
      "time_per_iteration": 2.831441879272461
    },
    {
      "auxiliary_loss_clip": 0.01154977,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.05490828,
      "balance_loss_mlp": 1.01838648,
      "epoch": 0.17447243431732098,
      "flos": 25666577504640.0,
      "grad_norm": 1.958467288265599,
      "language_loss": 0.77618849,
      "learning_rate": 3.7851497826400714e-06,
      "loss": 0.79803604,
      "num_input_tokens_seen": 30653625,
      "step": 1451,
      "time_per_iteration": 2.986922025680542
    },
    {
      "auxiliary_loss_clip": 0.01197127,
      "auxiliary_loss_mlp": 0.0104647,
      "balance_loss_clip": 1.05864716,
      "balance_loss_mlp": 1.03380966,
      "epoch": 0.17459267720796007,
      "flos": 36281950657920.0,
      "grad_norm": 1.8939669164802537,
      "language_loss": 0.75908434,
      "learning_rate": 3.7847984095752034e-06,
      "loss": 0.78152031,
      "num_input_tokens_seen": 30677080,
      "step": 1452,
      "time_per_iteration": 2.6750648021698
    },
    {
      "auxiliary_loss_clip": 0.01122555,
      "auxiliary_loss_mlp": 0.010406,
      "balance_loss_clip": 1.04713082,
      "balance_loss_mlp": 1.02920985,
      "epoch": 0.17471292009859918,
      "flos": 20011113959040.0,
      "grad_norm": 2.0349989138778337,
      "language_loss": 0.80153239,
      "learning_rate": 3.784446765757836e-06,
      "loss": 0.82316399,
      "num_input_tokens_seen": 30695725,
      "step": 1453,
      "time_per_iteration": 2.6654858589172363
    },
    {
      "auxiliary_loss_clip": 0.01179692,
      "auxiliary_loss_mlp": 0.01045063,
      "balance_loss_clip": 1.06293178,
      "balance_loss_mlp": 1.03211081,
      "epoch": 0.17483316298923826,
      "flos": 27819242559360.0,
      "grad_norm": 2.329253952022514,
      "language_loss": 0.78058386,
      "learning_rate": 3.7840948512413133e-06,
      "loss": 0.80283141,
      "num_input_tokens_seen": 30713310,
      "step": 1454,
      "time_per_iteration": 2.6208550930023193
    },
    {
      "auxiliary_loss_clip": 0.01149352,
      "auxiliary_loss_mlp": 0.01040079,
      "balance_loss_clip": 1.05049264,
      "balance_loss_mlp": 1.02723455,
      "epoch": 0.17495340587987734,
      "flos": 44017934791680.0,
      "grad_norm": 2.1472316527053485,
      "language_loss": 0.78705537,
      "learning_rate": 3.7837426660790196e-06,
      "loss": 0.80894971,
      "num_input_tokens_seen": 30734725,
      "step": 1455,
      "time_per_iteration": 2.7768852710723877
    },
    {
      "auxiliary_loss_clip": 0.01152176,
      "auxiliary_loss_mlp": 0.01043173,
      "balance_loss_clip": 1.05508828,
      "balance_loss_mlp": 1.03162718,
      "epoch": 0.17507364877051645,
      "flos": 20885825957760.0,
      "grad_norm": 2.165181859774156,
      "language_loss": 0.81613469,
      "learning_rate": 3.783390210324382e-06,
      "loss": 0.83808815,
      "num_input_tokens_seen": 30754450,
      "step": 1456,
      "time_per_iteration": 2.6289520263671875
    },
    {
      "auxiliary_loss_clip": 0.01158636,
      "auxiliary_loss_mlp": 0.01041463,
      "balance_loss_clip": 1.05545759,
      "balance_loss_mlp": 1.02943528,
      "epoch": 0.17519389166115554,
      "flos": 24717602136960.0,
      "grad_norm": 8.535661357863779,
      "language_loss": 0.72544348,
      "learning_rate": 3.7830374840308676e-06,
      "loss": 0.74744451,
      "num_input_tokens_seen": 30774605,
      "step": 1457,
      "time_per_iteration": 2.6509594917297363
    },
    {
      "auxiliary_loss_clip": 0.01202318,
      "auxiliary_loss_mlp": 0.00765865,
      "balance_loss_clip": 1.06652641,
      "balance_loss_mlp": 1.00069058,
      "epoch": 0.17531413455179462,
      "flos": 23798144770560.0,
      "grad_norm": 3.4041239599731203,
      "language_loss": 0.8247717,
      "learning_rate": 3.7826844872519842e-06,
      "loss": 0.84445351,
      "num_input_tokens_seen": 30792460,
      "step": 1458,
      "time_per_iteration": 2.569262981414795
    },
    {
      "auxiliary_loss_clip": 0.01156514,
      "auxiliary_loss_mlp": 0.01045729,
      "balance_loss_clip": 1.05438328,
      "balance_loss_mlp": 1.03484499,
      "epoch": 0.1754343774424337,
      "flos": 24572379450240.0,
      "grad_norm": 2.315334956847813,
      "language_loss": 0.72737128,
      "learning_rate": 3.782331220041282e-06,
      "loss": 0.7493937,
      "num_input_tokens_seen": 30812525,
      "step": 1459,
      "time_per_iteration": 2.6894497871398926
    },
    {
      "auxiliary_loss_clip": 0.01194096,
      "auxiliary_loss_mlp": 0.00765698,
      "balance_loss_clip": 1.05826235,
      "balance_loss_mlp": 1.00076365,
      "epoch": 0.17555462033307281,
      "flos": 18114599767680.0,
      "grad_norm": 2.250088157520169,
      "language_loss": 0.82589996,
      "learning_rate": 3.7819776824523504e-06,
      "loss": 0.84549785,
      "num_input_tokens_seen": 30830390,
      "step": 1460,
      "time_per_iteration": 2.5459249019622803
    },
    {
      "auxiliary_loss_clip": 0.01141914,
      "auxiliary_loss_mlp": 0.01041056,
      "balance_loss_clip": 1.05872464,
      "balance_loss_mlp": 1.02839053,
      "epoch": 0.1756748632237119,
      "flos": 28366018364160.0,
      "grad_norm": 3.055990391631219,
      "language_loss": 0.8390376,
      "learning_rate": 3.7816238745388213e-06,
      "loss": 0.86086726,
      "num_input_tokens_seen": 30849935,
      "step": 1461,
      "time_per_iteration": 2.691166639328003
    },
    {
      "auxiliary_loss_clip": 0.01182584,
      "auxiliary_loss_mlp": 0.0104329,
      "balance_loss_clip": 1.05875337,
      "balance_loss_mlp": 1.03276348,
      "epoch": 0.17579510611435098,
      "flos": 25732939881600.0,
      "grad_norm": 1.9888346555082583,
      "language_loss": 0.87068415,
      "learning_rate": 3.781269796354367e-06,
      "loss": 0.89294285,
      "num_input_tokens_seen": 30869555,
      "step": 1462,
      "time_per_iteration": 2.5969204902648926
    },
    {
      "auxiliary_loss_clip": 0.0119136,
      "auxiliary_loss_mlp": 0.01045986,
      "balance_loss_clip": 1.06323862,
      "balance_loss_mlp": 1.03414273,
      "epoch": 0.1759153490049901,
      "flos": 18588081870720.0,
      "grad_norm": 2.0758638739925868,
      "language_loss": 0.8604179,
      "learning_rate": 3.7809154479527006e-06,
      "loss": 0.8827914,
      "num_input_tokens_seen": 30888760,
      "step": 1463,
      "time_per_iteration": 2.517840623855591
    },
    {
      "auxiliary_loss_clip": 0.01172613,
      "auxiliary_loss_mlp": 0.00765748,
      "balance_loss_clip": 1.05934763,
      "balance_loss_mlp": 1.00067174,
      "epoch": 0.17603559189562917,
      "flos": 18619323724800.0,
      "grad_norm": 2.126368115499326,
      "language_loss": 0.83973598,
      "learning_rate": 3.780560829387577e-06,
      "loss": 0.85911953,
      "num_input_tokens_seen": 30907260,
      "step": 1464,
      "time_per_iteration": 3.412911891937256
    },
    {
      "auxiliary_loss_clip": 0.01108744,
      "auxiliary_loss_mlp": 0.01008813,
      "balance_loss_clip": 1.03868246,
      "balance_loss_mlp": 1.00640476,
      "epoch": 0.17615583478626826,
      "flos": 60530775373440.0,
      "grad_norm": 0.8559954308393418,
      "language_loss": 0.57906997,
      "learning_rate": 3.7802059407127915e-06,
      "loss": 0.60024559,
      "num_input_tokens_seen": 30965810,
      "step": 1465,
      "time_per_iteration": 3.471820116043091
    },
    {
      "auxiliary_loss_clip": 0.01189839,
      "auxiliary_loss_mlp": 0.01043645,
      "balance_loss_clip": 1.05826628,
      "balance_loss_mlp": 1.031986,
      "epoch": 0.17627607767690734,
      "flos": 23616221362560.0,
      "grad_norm": 2.118159526078746,
      "language_loss": 0.86318219,
      "learning_rate": 3.7798507819821797e-06,
      "loss": 0.88551706,
      "num_input_tokens_seen": 30982935,
      "step": 1466,
      "time_per_iteration": 3.3321852684020996
    },
    {
      "auxiliary_loss_clip": 0.01168989,
      "auxiliary_loss_mlp": 0.0103678,
      "balance_loss_clip": 1.05459654,
      "balance_loss_mlp": 1.02551448,
      "epoch": 0.17639632056754645,
      "flos": 17639070589440.0,
      "grad_norm": 3.3044740141713533,
      "language_loss": 0.78977442,
      "learning_rate": 3.7794953532496197e-06,
      "loss": 0.81183207,
      "num_input_tokens_seen": 30998840,
      "step": 1467,
      "time_per_iteration": 2.5097341537475586
    },
    {
      "auxiliary_loss_clip": 0.01104259,
      "auxiliary_loss_mlp": 0.01004729,
      "balance_loss_clip": 1.03820872,
      "balance_loss_mlp": 1.00246418,
      "epoch": 0.17651656345818553,
      "flos": 57932604910080.0,
      "grad_norm": 0.8650497480274542,
      "language_loss": 0.57905185,
      "learning_rate": 3.7791396545690295e-06,
      "loss": 0.60014176,
      "num_input_tokens_seen": 31060075,
      "step": 1468,
      "time_per_iteration": 4.650123596191406
    },
    {
      "auxiliary_loss_clip": 0.01215444,
      "auxiliary_loss_mlp": 0.01044969,
      "balance_loss_clip": 1.06303728,
      "balance_loss_mlp": 1.03302383,
      "epoch": 0.17663680634882462,
      "flos": 22929502170240.0,
      "grad_norm": 1.9908938890188537,
      "language_loss": 0.80767787,
      "learning_rate": 3.7787836859943685e-06,
      "loss": 0.83028197,
      "num_input_tokens_seen": 31078800,
      "step": 1469,
      "time_per_iteration": 2.5779545307159424
    },
    {
      "auxiliary_loss_clip": 0.01184111,
      "auxiliary_loss_mlp": 0.01037891,
      "balance_loss_clip": 1.05891597,
      "balance_loss_mlp": 1.02583289,
      "epoch": 0.17675704923946373,
      "flos": 22637979388800.0,
      "grad_norm": 2.5963935646138623,
      "language_loss": 0.78982639,
      "learning_rate": 3.7784274475796363e-06,
      "loss": 0.81204641,
      "num_input_tokens_seen": 31097430,
      "step": 1470,
      "time_per_iteration": 2.5915279388427734
    },
    {
      "auxiliary_loss_clip": 0.0119232,
      "auxiliary_loss_mlp": 0.01037879,
      "balance_loss_clip": 1.0589695,
      "balance_loss_mlp": 1.0250107,
      "epoch": 0.1768772921301028,
      "flos": 27126525795840.0,
      "grad_norm": 2.1907209759583233,
      "language_loss": 0.76400781,
      "learning_rate": 3.7780709393788745e-06,
      "loss": 0.78630984,
      "num_input_tokens_seen": 31117905,
      "step": 1471,
      "time_per_iteration": 2.6519393920898438
    },
    {
      "auxiliary_loss_clip": 0.0114922,
      "auxiliary_loss_mlp": 0.01040592,
      "balance_loss_clip": 1.04958165,
      "balance_loss_mlp": 1.02823031,
      "epoch": 0.1769975350207419,
      "flos": 19172133014400.0,
      "grad_norm": 2.584548405031066,
      "language_loss": 0.75047749,
      "learning_rate": 3.777714161446165e-06,
      "loss": 0.77237558,
      "num_input_tokens_seen": 31137610,
      "step": 1472,
      "time_per_iteration": 2.6205551624298096
    },
    {
      "auxiliary_loss_clip": 0.01173211,
      "auxiliary_loss_mlp": 0.0103906,
      "balance_loss_clip": 1.05568206,
      "balance_loss_mlp": 1.02744913,
      "epoch": 0.177117777911381,
      "flos": 36134932291200.0,
      "grad_norm": 2.4070214143634603,
      "language_loss": 0.694897,
      "learning_rate": 3.7773571138356304e-06,
      "loss": 0.71701968,
      "num_input_tokens_seen": 31157780,
      "step": 1473,
      "time_per_iteration": 2.681887626647949
    },
    {
      "auxiliary_loss_clip": 0.01168843,
      "auxiliary_loss_mlp": 0.01037725,
      "balance_loss_clip": 1.05439305,
      "balance_loss_mlp": 1.02634692,
      "epoch": 0.17723802080202009,
      "flos": 22090593052800.0,
      "grad_norm": 2.4956864502486713,
      "language_loss": 0.88801181,
      "learning_rate": 3.776999796601435e-06,
      "loss": 0.91007745,
      "num_input_tokens_seen": 31176540,
      "step": 1474,
      "time_per_iteration": 2.537156581878662
    },
    {
      "auxiliary_loss_clip": 0.01150842,
      "auxiliary_loss_mlp": 0.01041138,
      "balance_loss_clip": 1.05337238,
      "balance_loss_mlp": 1.02908611,
      "epoch": 0.17735826369265917,
      "flos": 30222671437440.0,
      "grad_norm": 2.226111381729052,
      "language_loss": 0.72508109,
      "learning_rate": 3.776642209797783e-06,
      "loss": 0.74700087,
      "num_input_tokens_seen": 31198370,
      "step": 1475,
      "time_per_iteration": 2.6783111095428467
    },
    {
      "auxiliary_loss_clip": 0.01176209,
      "auxiliary_loss_mlp": 0.01049161,
      "balance_loss_clip": 1.06188011,
      "balance_loss_mlp": 1.03664398,
      "epoch": 0.17747850658329825,
      "flos": 21397588980480.0,
      "grad_norm": 2.0531396862990268,
      "language_loss": 0.77820432,
      "learning_rate": 3.7762843534789205e-06,
      "loss": 0.80045801,
      "num_input_tokens_seen": 31217120,
      "step": 1476,
      "time_per_iteration": 2.600242853164673
    },
    {
      "auxiliary_loss_clip": 0.01194084,
      "auxiliary_loss_mlp": 0.01037047,
      "balance_loss_clip": 1.06118035,
      "balance_loss_mlp": 1.02513194,
      "epoch": 0.17759874947393736,
      "flos": 16983341856000.0,
      "grad_norm": 2.4300461251144005,
      "language_loss": 0.88490802,
      "learning_rate": 3.7759262276991343e-06,
      "loss": 0.90721929,
      "num_input_tokens_seen": 31234730,
      "step": 1477,
      "time_per_iteration": 2.551907777786255
    },
    {
      "auxiliary_loss_clip": 0.01138677,
      "auxiliary_loss_mlp": 0.01038743,
      "balance_loss_clip": 1.05164218,
      "balance_loss_mlp": 1.02665567,
      "epoch": 0.17771899236457644,
      "flos": 11546107390080.0,
      "grad_norm": 2.244195575975624,
      "language_loss": 0.80377764,
      "learning_rate": 3.7755678325127506e-06,
      "loss": 0.82555181,
      "num_input_tokens_seen": 31252410,
      "step": 1478,
      "time_per_iteration": 2.6347224712371826
    },
    {
      "auxiliary_loss_clip": 0.01185365,
      "auxiliary_loss_mlp": 0.01038485,
      "balance_loss_clip": 1.05505109,
      "balance_loss_mlp": 1.0263195,
      "epoch": 0.17783923525521553,
      "flos": 18807747494400.0,
      "grad_norm": 1.900072784903336,
      "language_loss": 0.75710112,
      "learning_rate": 3.7752091679741393e-06,
      "loss": 0.77933961,
      "num_input_tokens_seen": 31270200,
      "step": 1479,
      "time_per_iteration": 2.6083297729492188
    },
    {
      "auxiliary_loss_clip": 0.01191725,
      "auxiliary_loss_mlp": 0.01043891,
      "balance_loss_clip": 1.05670083,
      "balance_loss_mlp": 1.03097486,
      "epoch": 0.17795947814585464,
      "flos": 30408365773440.0,
      "grad_norm": 2.4483057756732696,
      "language_loss": 0.77643657,
      "learning_rate": 3.774850234137708e-06,
      "loss": 0.79879272,
      "num_input_tokens_seen": 31287495,
      "step": 1480,
      "time_per_iteration": 2.6008267402648926
    },
    {
      "auxiliary_loss_clip": 0.01173386,
      "auxiliary_loss_mlp": 0.01043223,
      "balance_loss_clip": 1.06121457,
      "balance_loss_mlp": 1.03128386,
      "epoch": 0.17807972103649372,
      "flos": 24389055411840.0,
      "grad_norm": 2.338282640897755,
      "language_loss": 0.83111012,
      "learning_rate": 3.7744910310579076e-06,
      "loss": 0.85327619,
      "num_input_tokens_seen": 31306420,
      "step": 1481,
      "time_per_iteration": 2.649675130844116
    },
    {
      "auxiliary_loss_clip": 0.01176774,
      "auxiliary_loss_mlp": 0.0103761,
      "balance_loss_clip": 1.06166017,
      "balance_loss_mlp": 1.02626097,
      "epoch": 0.1781999639271328,
      "flos": 20301559332480.0,
      "grad_norm": 2.059657698126154,
      "language_loss": 0.84958482,
      "learning_rate": 3.774131558789229e-06,
      "loss": 0.87172866,
      "num_input_tokens_seen": 31325750,
      "step": 1482,
      "time_per_iteration": 2.6603498458862305
    },
    {
      "auxiliary_loss_clip": 0.01229358,
      "auxiliary_loss_mlp": 0.01048798,
      "balance_loss_clip": 1.0684247,
      "balance_loss_mlp": 1.03737223,
      "epoch": 0.1783202068177719,
      "flos": 15924479806080.0,
      "grad_norm": 3.5639809674957235,
      "language_loss": 0.70100522,
      "learning_rate": 3.773771817386203e-06,
      "loss": 0.72378683,
      "num_input_tokens_seen": 31343080,
      "step": 1483,
      "time_per_iteration": 2.467280387878418
    },
    {
      "auxiliary_loss_clip": 0.01182337,
      "auxiliary_loss_mlp": 0.01032042,
      "balance_loss_clip": 1.06074286,
      "balance_loss_mlp": 1.02100945,
      "epoch": 0.178440449708411,
      "flos": 20631758083200.0,
      "grad_norm": 1.5729361653539713,
      "language_loss": 0.79271024,
      "learning_rate": 3.773411806903403e-06,
      "loss": 0.81485403,
      "num_input_tokens_seen": 31362160,
      "step": 1484,
      "time_per_iteration": 2.5954058170318604
    },
    {
      "auxiliary_loss_clip": 0.01199411,
      "auxiliary_loss_mlp": 0.0104197,
      "balance_loss_clip": 1.06627917,
      "balance_loss_mlp": 1.03083634,
      "epoch": 0.17856069259905008,
      "flos": 21686059105920.0,
      "grad_norm": 2.528792599810528,
      "language_loss": 0.94754827,
      "learning_rate": 3.7730515273954415e-06,
      "loss": 0.96996218,
      "num_input_tokens_seen": 31380770,
      "step": 1485,
      "time_per_iteration": 2.5735528469085693
    },
    {
      "auxiliary_loss_clip": 0.0112897,
      "auxiliary_loss_mlp": 0.0104247,
      "balance_loss_clip": 1.05582881,
      "balance_loss_mlp": 1.03142524,
      "epoch": 0.17868093548968916,
      "flos": 26572962320640.0,
      "grad_norm": 2.094631801345556,
      "language_loss": 0.85084867,
      "learning_rate": 3.772690978916973e-06,
      "loss": 0.87256312,
      "num_input_tokens_seen": 31400525,
      "step": 1486,
      "time_per_iteration": 2.7832295894622803
    },
    {
      "auxiliary_loss_clip": 0.01121615,
      "auxiliary_loss_mlp": 0.01043784,
      "balance_loss_clip": 1.04844248,
      "balance_loss_mlp": 1.03192234,
      "epoch": 0.17880117838032827,
      "flos": 18581006891520.0,
      "grad_norm": 2.361445885545015,
      "language_loss": 0.86759913,
      "learning_rate": 3.772330161522693e-06,
      "loss": 0.88925314,
      "num_input_tokens_seen": 31418435,
      "step": 1487,
      "time_per_iteration": 2.8851799964904785
    },
    {
      "auxiliary_loss_clip": 0.01141063,
      "auxiliary_loss_mlp": 0.01047691,
      "balance_loss_clip": 1.06028581,
      "balance_loss_mlp": 1.03565669,
      "epoch": 0.17892142127096736,
      "flos": 26541217676160.0,
      "grad_norm": 2.19703291315551,
      "language_loss": 0.79995692,
      "learning_rate": 3.7719690752673365e-06,
      "loss": 0.82184446,
      "num_input_tokens_seen": 31439230,
      "step": 1488,
      "time_per_iteration": 2.7063002586364746
    },
    {
      "auxiliary_loss_clip": 0.01220842,
      "auxiliary_loss_mlp": 0.0076542,
      "balance_loss_clip": 1.06762028,
      "balance_loss_mlp": 1.00072086,
      "epoch": 0.17904166416160644,
      "flos": 23872623621120.0,
      "grad_norm": 2.392679211083732,
      "language_loss": 0.78019786,
      "learning_rate": 3.7716077202056796e-06,
      "loss": 0.80006039,
      "num_input_tokens_seen": 31457705,
      "step": 1489,
      "time_per_iteration": 2.5293631553649902
    },
    {
      "auxiliary_loss_clip": 0.01190886,
      "auxiliary_loss_mlp": 0.010373,
      "balance_loss_clip": 1.05894554,
      "balance_loss_mlp": 1.02566516,
      "epoch": 0.17916190705224552,
      "flos": 19134426712320.0,
      "grad_norm": 2.4508336896528573,
      "language_loss": 0.93876618,
      "learning_rate": 3.7712460963925404e-06,
      "loss": 0.96104807,
      "num_input_tokens_seen": 31473645,
      "step": 1490,
      "time_per_iteration": 3.2583019733428955
    },
    {
      "auxiliary_loss_clip": 0.01194376,
      "auxiliary_loss_mlp": 0.01038337,
      "balance_loss_clip": 1.05967391,
      "balance_loss_mlp": 1.02705359,
      "epoch": 0.17928214994288463,
      "flos": 25152120961920.0,
      "grad_norm": 1.90907872242967,
      "language_loss": 0.75674105,
      "learning_rate": 3.7708842038827775e-06,
      "loss": 0.77906823,
      "num_input_tokens_seen": 31492605,
      "step": 1491,
      "time_per_iteration": 2.5399816036224365
    },
    {
      "auxiliary_loss_clip": 0.01151545,
      "auxiliary_loss_mlp": 0.01038714,
      "balance_loss_clip": 1.05183911,
      "balance_loss_mlp": 1.02679908,
      "epoch": 0.17940239283352372,
      "flos": 22384629786240.0,
      "grad_norm": 1.8215274437100184,
      "language_loss": 0.85766071,
      "learning_rate": 3.770522042731288e-06,
      "loss": 0.87956327,
      "num_input_tokens_seen": 31514500,
      "step": 1492,
      "time_per_iteration": 3.5495948791503906
    },
    {
      "auxiliary_loss_clip": 0.01157911,
      "auxiliary_loss_mlp": 0.01033951,
      "balance_loss_clip": 1.05564237,
      "balance_loss_mlp": 1.02155364,
      "epoch": 0.1795226357241628,
      "flos": 23178685795200.0,
      "grad_norm": 2.168962027919063,
      "language_loss": 0.87533933,
      "learning_rate": 3.7701596129930122e-06,
      "loss": 0.89725792,
      "num_input_tokens_seen": 31533225,
      "step": 1493,
      "time_per_iteration": 2.7004363536834717
    },
    {
      "auxiliary_loss_clip": 0.01214327,
      "auxiliary_loss_mlp": 0.0104077,
      "balance_loss_clip": 1.0647974,
      "balance_loss_mlp": 1.02775204,
      "epoch": 0.1796428786148019,
      "flos": 22090413484800.0,
      "grad_norm": 2.102859642208792,
      "language_loss": 0.73459363,
      "learning_rate": 3.7697969147229315e-06,
      "loss": 0.75714457,
      "num_input_tokens_seen": 31551385,
      "step": 1494,
      "time_per_iteration": 4.004640817642212
    },
    {
      "auxiliary_loss_clip": 0.01185024,
      "auxiliary_loss_mlp": 0.01039547,
      "balance_loss_clip": 1.05817473,
      "balance_loss_mlp": 1.02835321,
      "epoch": 0.179763121505441,
      "flos": 21324618501120.0,
      "grad_norm": 2.0655550929489017,
      "language_loss": 0.85134804,
      "learning_rate": 3.7694339479760647e-06,
      "loss": 0.87359375,
      "num_input_tokens_seen": 31570415,
      "step": 1495,
      "time_per_iteration": 2.5632569789886475
    },
    {
      "auxiliary_loss_clip": 0.01078719,
      "auxiliary_loss_mlp": 0.01008687,
      "balance_loss_clip": 1.05262101,
      "balance_loss_mlp": 1.0063622,
      "epoch": 0.17988336439608008,
      "flos": 68161864815360.0,
      "grad_norm": 0.7791987309305327,
      "language_loss": 0.57292753,
      "learning_rate": 3.769070712807476e-06,
      "loss": 0.59380156,
      "num_input_tokens_seen": 31632445,
      "step": 1496,
      "time_per_iteration": 3.286642551422119
    },
    {
      "auxiliary_loss_clip": 0.01163383,
      "auxiliary_loss_mlp": 0.01038439,
      "balance_loss_clip": 1.05679178,
      "balance_loss_mlp": 1.02610731,
      "epoch": 0.18000360728671919,
      "flos": 21945047143680.0,
      "grad_norm": 1.7928878132945911,
      "language_loss": 0.78962213,
      "learning_rate": 3.768707209272266e-06,
      "loss": 0.81164032,
      "num_input_tokens_seen": 31652575,
      "step": 1497,
      "time_per_iteration": 2.59309983253479
    },
    {
      "auxiliary_loss_clip": 0.0114335,
      "auxiliary_loss_mlp": 0.0103819,
      "balance_loss_clip": 1.05617738,
      "balance_loss_mlp": 1.02489209,
      "epoch": 0.18012385017735827,
      "flos": 18986330937600.0,
      "grad_norm": 2.1906907844893095,
      "language_loss": 0.76215404,
      "learning_rate": 3.768343437425579e-06,
      "loss": 0.7839694,
      "num_input_tokens_seen": 31671145,
      "step": 1498,
      "time_per_iteration": 2.6102139949798584
    },
    {
      "auxiliary_loss_clip": 0.01227693,
      "auxiliary_loss_mlp": 0.01047179,
      "balance_loss_clip": 1.06447518,
      "balance_loss_mlp": 1.03370833,
      "epoch": 0.18024409306799735,
      "flos": 19748103598080.0,
      "grad_norm": 2.3949214243492523,
      "language_loss": 0.86044466,
      "learning_rate": 3.7679793973225987e-06,
      "loss": 0.88319337,
      "num_input_tokens_seen": 31686955,
      "step": 1499,
      "time_per_iteration": 2.4560978412628174
    },
    {
      "auxiliary_loss_clip": 0.0107977,
      "auxiliary_loss_mlp": 0.01004188,
      "balance_loss_clip": 1.03598607,
      "balance_loss_mlp": 1.00211382,
      "epoch": 0.18036433595863643,
      "flos": 67227183060480.0,
      "grad_norm": 0.8511808421576808,
      "language_loss": 0.61586511,
      "learning_rate": 3.767615089018549e-06,
      "loss": 0.63670468,
      "num_input_tokens_seen": 31749300,
      "step": 1500,
      "time_per_iteration": 3.1701366901397705
    },
    {
      "auxiliary_loss_clip": 0.01165498,
      "auxiliary_loss_mlp": 0.00765941,
      "balance_loss_clip": 1.05737519,
      "balance_loss_mlp": 1.00051188,
      "epoch": 0.18048457884927555,
      "flos": 18181464935040.0,
      "grad_norm": 2.066042579643156,
      "language_loss": 0.85946625,
      "learning_rate": 3.7672505125686966e-06,
      "loss": 0.8787806,
      "num_input_tokens_seen": 31765665,
      "step": 1501,
      "time_per_iteration": 2.5355076789855957
    },
    {
      "auxiliary_loss_clip": 0.01154186,
      "auxiliary_loss_mlp": 0.01043497,
      "balance_loss_clip": 1.05823219,
      "balance_loss_mlp": 1.03145647,
      "epoch": 0.18060482173991463,
      "flos": 15813767111040.0,
      "grad_norm": 3.0317821699957808,
      "language_loss": 0.84569192,
      "learning_rate": 3.7668856680283455e-06,
      "loss": 0.86766875,
      "num_input_tokens_seen": 31782690,
      "step": 1502,
      "time_per_iteration": 2.578212022781372
    },
    {
      "auxiliary_loss_clip": 0.01209369,
      "auxiliary_loss_mlp": 0.01040167,
      "balance_loss_clip": 1.06477928,
      "balance_loss_mlp": 1.02760267,
      "epoch": 0.1807250646305537,
      "flos": 18587399512320.0,
      "grad_norm": 2.110413764818164,
      "language_loss": 0.82217336,
      "learning_rate": 3.7665205554528437e-06,
      "loss": 0.84466875,
      "num_input_tokens_seen": 31802045,
      "step": 1503,
      "time_per_iteration": 2.4986982345581055
    },
    {
      "auxiliary_loss_clip": 0.01183415,
      "auxiliary_loss_mlp": 0.01047224,
      "balance_loss_clip": 1.0624547,
      "balance_loss_mlp": 1.03474307,
      "epoch": 0.18084530752119282,
      "flos": 23149131880320.0,
      "grad_norm": 2.9518692138074276,
      "language_loss": 0.74131709,
      "learning_rate": 3.7661551748975782e-06,
      "loss": 0.76362348,
      "num_input_tokens_seen": 31820220,
      "step": 1504,
      "time_per_iteration": 2.5870912075042725
    },
    {
      "auxiliary_loss_clip": 0.01113728,
      "auxiliary_loss_mlp": 0.01003794,
      "balance_loss_clip": 1.04271054,
      "balance_loss_mlp": 1.00173128,
      "epoch": 0.1809655504118319,
      "flos": 59803153568640.0,
      "grad_norm": 0.8138730227538095,
      "language_loss": 0.6044066,
      "learning_rate": 3.7657895264179772e-06,
      "loss": 0.62558186,
      "num_input_tokens_seen": 31876195,
      "step": 1505,
      "time_per_iteration": 3.112074136734009
    },
    {
      "auxiliary_loss_clip": 0.0121519,
      "auxiliary_loss_mlp": 0.00765597,
      "balance_loss_clip": 1.06037784,
      "balance_loss_mlp": 1.00058675,
      "epoch": 0.181085793302471,
      "flos": 44201941188480.0,
      "grad_norm": 1.7801385215493688,
      "language_loss": 0.74380195,
      "learning_rate": 3.765423610069509e-06,
      "loss": 0.76360977,
      "num_input_tokens_seen": 31901585,
      "step": 1506,
      "time_per_iteration": 2.688555955886841
    },
    {
      "auxiliary_loss_clip": 0.0121248,
      "auxiliary_loss_mlp": 0.01043198,
      "balance_loss_clip": 1.06551278,
      "balance_loss_mlp": 1.03054357,
      "epoch": 0.18120603619311007,
      "flos": 34898384638080.0,
      "grad_norm": 1.8774062052405867,
      "language_loss": 0.725999,
      "learning_rate": 3.765057425907683e-06,
      "loss": 0.74855578,
      "num_input_tokens_seen": 31923045,
      "step": 1507,
      "time_per_iteration": 2.6096272468566895
    },
    {
      "auxiliary_loss_clip": 0.01119497,
      "auxiliary_loss_mlp": 0.01047557,
      "balance_loss_clip": 1.05004573,
      "balance_loss_mlp": 1.03531432,
      "epoch": 0.18132627908374918,
      "flos": 21506757390720.0,
      "grad_norm": 2.120419033649588,
      "language_loss": 0.78574634,
      "learning_rate": 3.764690973988048e-06,
      "loss": 0.80741692,
      "num_input_tokens_seen": 31943385,
      "step": 1508,
      "time_per_iteration": 2.661842107772827
    },
    {
      "auxiliary_loss_clip": 0.01176735,
      "auxiliary_loss_mlp": 0.01035834,
      "balance_loss_clip": 1.05374813,
      "balance_loss_mlp": 1.02439046,
      "epoch": 0.18144652197438826,
      "flos": 29057693633280.0,
      "grad_norm": 2.1996423290275233,
      "language_loss": 0.73770142,
      "learning_rate": 3.7643242543661967e-06,
      "loss": 0.75982714,
      "num_input_tokens_seen": 31966045,
      "step": 1509,
      "time_per_iteration": 2.6136250495910645
    },
    {
      "auxiliary_loss_clip": 0.01068617,
      "auxiliary_loss_mlp": 0.01006246,
      "balance_loss_clip": 1.0369854,
      "balance_loss_mlp": 1.00393343,
      "epoch": 0.18156676486502735,
      "flos": 68675064382080.0,
      "grad_norm": 0.8161151963094327,
      "language_loss": 0.60527921,
      "learning_rate": 3.7639572670977573e-06,
      "loss": 0.62602782,
      "num_input_tokens_seen": 32021540,
      "step": 1510,
      "time_per_iteration": 3.127746105194092
    },
    {
      "auxiliary_loss_clip": 0.01160333,
      "auxiliary_loss_mlp": 0.01041734,
      "balance_loss_clip": 1.05824876,
      "balance_loss_mlp": 1.0294435,
      "epoch": 0.18168700775566646,
      "flos": 26471515334400.0,
      "grad_norm": 1.6175140454643904,
      "language_loss": 0.76634347,
      "learning_rate": 3.7635900122384042e-06,
      "loss": 0.78836417,
      "num_input_tokens_seen": 32044535,
      "step": 1511,
      "time_per_iteration": 2.647618293762207
    },
    {
      "auxiliary_loss_clip": 0.01188195,
      "auxiliary_loss_mlp": 0.01045152,
      "balance_loss_clip": 1.05919111,
      "balance_loss_mlp": 1.03150845,
      "epoch": 0.18180725064630554,
      "flos": 15005668884480.0,
      "grad_norm": 2.381376732470191,
      "language_loss": 0.86873817,
      "learning_rate": 3.7632224898438477e-06,
      "loss": 0.89107162,
      "num_input_tokens_seen": 32061010,
      "step": 1512,
      "time_per_iteration": 2.5259251594543457
    },
    {
      "auxiliary_loss_clip": 0.01190268,
      "auxiliary_loss_mlp": 0.00765618,
      "balance_loss_clip": 1.06167006,
      "balance_loss_mlp": 1.00059593,
      "epoch": 0.18192749353694462,
      "flos": 19682387665920.0,
      "grad_norm": 1.5910260834798784,
      "language_loss": 0.79290617,
      "learning_rate": 3.762854699969842e-06,
      "loss": 0.81246507,
      "num_input_tokens_seen": 32081520,
      "step": 1513,
      "time_per_iteration": 2.560279607772827
    },
    {
      "auxiliary_loss_clip": 0.01165674,
      "auxiliary_loss_mlp": 0.01037901,
      "balance_loss_clip": 1.05325603,
      "balance_loss_mlp": 1.02583647,
      "epoch": 0.1820477364275837,
      "flos": 20702717400960.0,
      "grad_norm": 1.8612669093897296,
      "language_loss": 0.733446,
      "learning_rate": 3.762486642672179e-06,
      "loss": 0.75548178,
      "num_input_tokens_seen": 32098460,
      "step": 1514,
      "time_per_iteration": 2.5511844158172607
    },
    {
      "auxiliary_loss_clip": 0.01180764,
      "auxiliary_loss_mlp": 0.01044371,
      "balance_loss_clip": 1.05943823,
      "balance_loss_mlp": 1.03194356,
      "epoch": 0.18216797931822282,
      "flos": 17128708197120.0,
      "grad_norm": 1.8743918241180801,
      "language_loss": 0.86890888,
      "learning_rate": 3.7621183180066946e-06,
      "loss": 0.89116025,
      "num_input_tokens_seen": 32116420,
      "step": 1515,
      "time_per_iteration": 3.224025249481201
    },
    {
      "auxiliary_loss_clip": 0.0114941,
      "auxiliary_loss_mlp": 0.01044515,
      "balance_loss_clip": 1.05015278,
      "balance_loss_mlp": 1.03205192,
      "epoch": 0.1822882222088619,
      "flos": 29242561956480.0,
      "grad_norm": 1.7452174820239181,
      "language_loss": 0.74033684,
      "learning_rate": 3.7617497260292625e-06,
      "loss": 0.76227611,
      "num_input_tokens_seen": 32138475,
      "step": 1516,
      "time_per_iteration": 2.657198905944824
    },
    {
      "auxiliary_loss_clip": 0.01195635,
      "auxiliary_loss_mlp": 0.01041557,
      "balance_loss_clip": 1.06309915,
      "balance_loss_mlp": 1.02974951,
      "epoch": 0.18240846509950098,
      "flos": 17702739446400.0,
      "grad_norm": 2.74745454536021,
      "language_loss": 0.78772724,
      "learning_rate": 3.7613808667957967e-06,
      "loss": 0.81009912,
      "num_input_tokens_seen": 32151165,
      "step": 1517,
      "time_per_iteration": 2.480591058731079
    },
    {
      "auxiliary_loss_clip": 0.01212703,
      "auxiliary_loss_mlp": 0.01045803,
      "balance_loss_clip": 1.0615232,
      "balance_loss_mlp": 1.03313696,
      "epoch": 0.1825287079901401,
      "flos": 14790025584000.0,
      "grad_norm": 1.8336883544063771,
      "language_loss": 0.91015005,
      "learning_rate": 3.7610117403622547e-06,
      "loss": 0.93273509,
      "num_input_tokens_seen": 32167725,
      "step": 1518,
      "time_per_iteration": 3.5683586597442627
    },
    {
      "auxiliary_loss_clip": 0.01191312,
      "auxiliary_loss_mlp": 0.01032393,
      "balance_loss_clip": 1.06037402,
      "balance_loss_mlp": 1.020877,
      "epoch": 0.18264895088077918,
      "flos": 21946232292480.0,
      "grad_norm": 1.9741642948453857,
      "language_loss": 0.89767224,
      "learning_rate": 3.7606423467846313e-06,
      "loss": 0.91990924,
      "num_input_tokens_seen": 32187330,
      "step": 1519,
      "time_per_iteration": 2.507725954055786
    },
    {
      "auxiliary_loss_clip": 0.01124313,
      "auxiliary_loss_mlp": 0.01044527,
      "balance_loss_clip": 1.05281329,
      "balance_loss_mlp": 1.03215301,
      "epoch": 0.18276919377141826,
      "flos": 20886759711360.0,
      "grad_norm": 1.544877730438656,
      "language_loss": 0.8006838,
      "learning_rate": 3.760272686118964e-06,
      "loss": 0.8223722,
      "num_input_tokens_seen": 32205550,
      "step": 1520,
      "time_per_iteration": 3.386478900909424
    },
    {
      "auxiliary_loss_clip": 0.01192177,
      "auxiliary_loss_mlp": 0.00766234,
      "balance_loss_clip": 1.06143486,
      "balance_loss_mlp": 1.00053954,
      "epoch": 0.18288943666205737,
      "flos": 21469877101440.0,
      "grad_norm": 2.170965603810338,
      "language_loss": 0.92496192,
      "learning_rate": 3.7599027584213297e-06,
      "loss": 0.94454598,
      "num_input_tokens_seen": 32224430,
      "step": 1521,
      "time_per_iteration": 2.521584987640381
    },
    {
      "auxiliary_loss_clip": 0.01183498,
      "auxiliary_loss_mlp": 0.00766133,
      "balance_loss_clip": 1.06228602,
      "balance_loss_mlp": 1.00050712,
      "epoch": 0.18300967955269645,
      "flos": 21539363961600.0,
      "grad_norm": 2.0125601015034476,
      "language_loss": 0.78176874,
      "learning_rate": 3.7595325637478465e-06,
      "loss": 0.80126506,
      "num_input_tokens_seen": 32242455,
      "step": 1522,
      "time_per_iteration": 2.571197509765625
    },
    {
      "auxiliary_loss_clip": 0.01176612,
      "auxiliary_loss_mlp": 0.01041506,
      "balance_loss_clip": 1.0607028,
      "balance_loss_mlp": 1.0289712,
      "epoch": 0.18312992244333554,
      "flos": 28876237102080.0,
      "grad_norm": 2.0618245016932764,
      "language_loss": 0.81518006,
      "learning_rate": 3.7591621021546723e-06,
      "loss": 0.83736122,
      "num_input_tokens_seen": 32264450,
      "step": 1523,
      "time_per_iteration": 2.6159214973449707
    },
    {
      "auxiliary_loss_clip": 0.01214109,
      "auxiliary_loss_mlp": 0.01043455,
      "balance_loss_clip": 1.0615766,
      "balance_loss_mlp": 1.03050876,
      "epoch": 0.18325016533397462,
      "flos": 20120102801280.0,
      "grad_norm": 1.7926123007600057,
      "language_loss": 0.81495041,
      "learning_rate": 3.7587913736980062e-06,
      "loss": 0.83752596,
      "num_input_tokens_seen": 32284090,
      "step": 1524,
      "time_per_iteration": 2.5231006145477295
    },
    {
      "auxiliary_loss_clip": 0.01112423,
      "auxiliary_loss_mlp": 0.01035816,
      "balance_loss_clip": 1.04909921,
      "balance_loss_mlp": 1.02441931,
      "epoch": 0.18337040822461373,
      "flos": 23329187781120.0,
      "grad_norm": 1.6456896077757779,
      "language_loss": 0.84106874,
      "learning_rate": 3.7584203784340865e-06,
      "loss": 0.86255109,
      "num_input_tokens_seen": 32303260,
      "step": 1525,
      "time_per_iteration": 2.9540398120880127
    },
    {
      "auxiliary_loss_clip": 0.01203436,
      "auxiliary_loss_mlp": 0.01039819,
      "balance_loss_clip": 1.06177974,
      "balance_loss_mlp": 1.02664042,
      "epoch": 0.1834906511152528,
      "flos": 25009555881600.0,
      "grad_norm": 2.01223171845927,
      "language_loss": 0.86131042,
      "learning_rate": 3.7580491164191938e-06,
      "loss": 0.88374299,
      "num_input_tokens_seen": 32321570,
      "step": 1526,
      "time_per_iteration": 2.825200319290161
    },
    {
      "auxiliary_loss_clip": 0.01087333,
      "auxiliary_loss_mlp": 0.01001687,
      "balance_loss_clip": 1.03195858,
      "balance_loss_mlp": 0.99976778,
      "epoch": 0.1836108940058919,
      "flos": 67251493589760.0,
      "grad_norm": 0.7440982573554715,
      "language_loss": 0.61265671,
      "learning_rate": 3.757677587709648e-06,
      "loss": 0.63354689,
      "num_input_tokens_seen": 32384835,
      "step": 1527,
      "time_per_iteration": 3.2783122062683105
    },
    {
      "auxiliary_loss_clip": 0.01213025,
      "auxiliary_loss_mlp": 0.01038797,
      "balance_loss_clip": 1.06601596,
      "balance_loss_mlp": 1.02665532,
      "epoch": 0.183731136896531,
      "flos": 25738721971200.0,
      "grad_norm": 3.1016229147990213,
      "language_loss": 0.75485712,
      "learning_rate": 3.7573057923618095e-06,
      "loss": 0.77737534,
      "num_input_tokens_seen": 32404930,
      "step": 1528,
      "time_per_iteration": 2.5280845165252686
    },
    {
      "auxiliary_loss_clip": 0.01192285,
      "auxiliary_loss_mlp": 0.01040286,
      "balance_loss_clip": 1.05714846,
      "balance_loss_mlp": 1.02697623,
      "epoch": 0.1838513797871701,
      "flos": 20449403712000.0,
      "grad_norm": 7.621880367269744,
      "language_loss": 0.74343538,
      "learning_rate": 3.7569337304320793e-06,
      "loss": 0.76576114,
      "num_input_tokens_seen": 32424515,
      "step": 1529,
      "time_per_iteration": 2.5365633964538574
    },
    {
      "auxiliary_loss_clip": 0.01083803,
      "auxiliary_loss_mlp": 0.01002475,
      "balance_loss_clip": 1.03597903,
      "balance_loss_mlp": 1.00060391,
      "epoch": 0.18397162267780917,
      "flos": 68565141786240.0,
      "grad_norm": 0.8498016601288025,
      "language_loss": 0.64445859,
      "learning_rate": 3.756561401976899e-06,
      "loss": 0.66532135,
      "num_input_tokens_seen": 32484220,
      "step": 1530,
      "time_per_iteration": 3.054025888442993
    },
    {
      "auxiliary_loss_clip": 0.0121614,
      "auxiliary_loss_mlp": 0.00765567,
      "balance_loss_clip": 1.06645191,
      "balance_loss_mlp": 1.0005765,
      "epoch": 0.18409186556844825,
      "flos": 31941104976000.0,
      "grad_norm": 2.020403072581596,
      "language_loss": 0.82505351,
      "learning_rate": 3.7561888070527514e-06,
      "loss": 0.84487057,
      "num_input_tokens_seen": 32506260,
      "step": 1531,
      "time_per_iteration": 2.6196694374084473
    },
    {
      "auxiliary_loss_clip": 0.01194957,
      "auxiliary_loss_mlp": 0.01032916,
      "balance_loss_clip": 1.06098056,
      "balance_loss_mlp": 1.02127576,
      "epoch": 0.18421210845908736,
      "flos": 20120533764480.0,
      "grad_norm": 2.0364441341527146,
      "language_loss": 0.7984975,
      "learning_rate": 3.7558159457161577e-06,
      "loss": 0.82077622,
      "num_input_tokens_seen": 32524225,
      "step": 1532,
      "time_per_iteration": 2.5534443855285645
    },
    {
      "auxiliary_loss_clip": 0.01165222,
      "auxiliary_loss_mlp": 0.01046608,
      "balance_loss_clip": 1.06117487,
      "balance_loss_mlp": 1.03509283,
      "epoch": 0.18433235134972645,
      "flos": 23110491824640.0,
      "grad_norm": 2.656208634324104,
      "language_loss": 0.78116781,
      "learning_rate": 3.755442818023681e-06,
      "loss": 0.80328608,
      "num_input_tokens_seen": 32543850,
      "step": 1533,
      "time_per_iteration": 2.601785659790039
    },
    {
      "auxiliary_loss_clip": 0.01150669,
      "auxiliary_loss_mlp": 0.01041416,
      "balance_loss_clip": 1.05250347,
      "balance_loss_mlp": 1.0297811,
      "epoch": 0.18445259424036553,
      "flos": 18291351617280.0,
      "grad_norm": 2.7108298522968504,
      "language_loss": 0.7633931,
      "learning_rate": 3.7550694240319246e-06,
      "loss": 0.78531396,
      "num_input_tokens_seen": 32561725,
      "step": 1534,
      "time_per_iteration": 2.5696043968200684
    },
    {
      "auxiliary_loss_clip": 0.01196199,
      "auxiliary_loss_mlp": 0.01041351,
      "balance_loss_clip": 1.06529021,
      "balance_loss_mlp": 1.02951312,
      "epoch": 0.18457283713100464,
      "flos": 21324079797120.0,
      "grad_norm": 2.5379912996131484,
      "language_loss": 0.76584113,
      "learning_rate": 3.7546957637975326e-06,
      "loss": 0.78821659,
      "num_input_tokens_seen": 32579135,
      "step": 1535,
      "time_per_iteration": 2.548673152923584
    },
    {
      "auxiliary_loss_clip": 0.01179439,
      "auxiliary_loss_mlp": 0.01034009,
      "balance_loss_clip": 1.05944562,
      "balance_loss_mlp": 1.02314854,
      "epoch": 0.18469308002164372,
      "flos": 20375679047040.0,
      "grad_norm": 1.6043028637859391,
      "language_loss": 0.74159861,
      "learning_rate": 3.7543218373771873e-06,
      "loss": 0.76373303,
      "num_input_tokens_seen": 32598460,
      "step": 1536,
      "time_per_iteration": 2.5329697132110596
    },
    {
      "auxiliary_loss_clip": 0.01181562,
      "auxiliary_loss_mlp": 0.01038532,
      "balance_loss_clip": 1.06622612,
      "balance_loss_mlp": 1.02761829,
      "epoch": 0.1848133229122828,
      "flos": 26435892021120.0,
      "grad_norm": 1.3813649331964792,
      "language_loss": 0.78223878,
      "learning_rate": 3.753947644827615e-06,
      "loss": 0.80443966,
      "num_input_tokens_seen": 32621920,
      "step": 1537,
      "time_per_iteration": 2.6440892219543457
    },
    {
      "auxiliary_loss_clip": 0.01062201,
      "auxiliary_loss_mlp": 0.01002072,
      "balance_loss_clip": 1.030339,
      "balance_loss_mlp": 1.00009334,
      "epoch": 0.1849335658029219,
      "flos": 70547447612160.0,
      "grad_norm": 1.0389736096606235,
      "language_loss": 0.57182372,
      "learning_rate": 3.753573186205579e-06,
      "loss": 0.59246647,
      "num_input_tokens_seen": 32690040,
      "step": 1538,
      "time_per_iteration": 3.345210075378418
    },
    {
      "auxiliary_loss_clip": 0.01202651,
      "auxiliary_loss_mlp": 0.01040023,
      "balance_loss_clip": 1.06126475,
      "balance_loss_mlp": 1.02849507,
      "epoch": 0.185053808693561,
      "flos": 17384140788480.0,
      "grad_norm": 2.2034173232609024,
      "language_loss": 0.77800852,
      "learning_rate": 3.753198461567885e-06,
      "loss": 0.8004353,
      "num_input_tokens_seen": 32707285,
      "step": 1539,
      "time_per_iteration": 2.476626396179199
    },
    {
      "auxiliary_loss_clip": 0.01131947,
      "auxiliary_loss_mlp": 0.01029757,
      "balance_loss_clip": 1.05316913,
      "balance_loss_mlp": 1.01889682,
      "epoch": 0.18517405158420008,
      "flos": 28986159697920.0,
      "grad_norm": 1.7778397390655662,
      "language_loss": 0.92549455,
      "learning_rate": 3.7528234709713783e-06,
      "loss": 0.94711155,
      "num_input_tokens_seen": 32730030,
      "step": 1540,
      "time_per_iteration": 3.5669314861297607
    },
    {
      "auxiliary_loss_clip": 0.01116044,
      "auxiliary_loss_mlp": 0.01035574,
      "balance_loss_clip": 1.04861522,
      "balance_loss_mlp": 1.02340877,
      "epoch": 0.18529429447483917,
      "flos": 26794962328320.0,
      "grad_norm": 1.8894215114453208,
      "language_loss": 0.84512496,
      "learning_rate": 3.7524482144729447e-06,
      "loss": 0.86664116,
      "num_input_tokens_seen": 32749485,
      "step": 1541,
      "time_per_iteration": 2.805020809173584
    },
    {
      "auxiliary_loss_clip": 0.0115098,
      "auxiliary_loss_mlp": 0.01036561,
      "balance_loss_clip": 1.0529263,
      "balance_loss_mlp": 1.02495575,
      "epoch": 0.18541453736547828,
      "flos": 13581595301760.0,
      "grad_norm": 2.1411925849882696,
      "language_loss": 0.83437049,
      "learning_rate": 3.7520726921295106e-06,
      "loss": 0.85624588,
      "num_input_tokens_seen": 32766205,
      "step": 1542,
      "time_per_iteration": 3.0852975845336914
    },
    {
      "auxiliary_loss_clip": 0.01109615,
      "auxiliary_loss_mlp": 0.01034127,
      "balance_loss_clip": 1.05186355,
      "balance_loss_mlp": 1.02242088,
      "epoch": 0.18553478025611736,
      "flos": 24025424077440.0,
      "grad_norm": 1.8518033377711849,
      "language_loss": 0.72589737,
      "learning_rate": 3.751696903998042e-06,
      "loss": 0.74733478,
      "num_input_tokens_seen": 32784840,
      "step": 1543,
      "time_per_iteration": 2.7335867881774902
    },
    {
      "auxiliary_loss_clip": 0.01213507,
      "auxiliary_loss_mlp": 0.01043055,
      "balance_loss_clip": 1.06562817,
      "balance_loss_mlp": 1.03127718,
      "epoch": 0.18565502314675644,
      "flos": 25885165720320.0,
      "grad_norm": 2.6509740008923504,
      "language_loss": 0.70189887,
      "learning_rate": 3.7513208501355456e-06,
      "loss": 0.72446454,
      "num_input_tokens_seen": 32805945,
      "step": 1544,
      "time_per_iteration": 3.6290531158447266
    },
    {
      "auxiliary_loss_clip": 0.01160804,
      "auxiliary_loss_mlp": 0.01037717,
      "balance_loss_clip": 1.05590129,
      "balance_loss_mlp": 1.02580833,
      "epoch": 0.18577526603739553,
      "flos": 19610063631360.0,
      "grad_norm": 2.0831634032517483,
      "language_loss": 0.83666921,
      "learning_rate": 3.750944530599069e-06,
      "loss": 0.85865438,
      "num_input_tokens_seen": 32825515,
      "step": 1545,
      "time_per_iteration": 3.3706493377685547
    },
    {
      "auxiliary_loss_clip": 0.01194306,
      "auxiliary_loss_mlp": 0.01044041,
      "balance_loss_clip": 1.0606904,
      "balance_loss_mlp": 1.03167915,
      "epoch": 0.18589550892803464,
      "flos": 18474891137280.0,
      "grad_norm": 2.6688471716902127,
      "language_loss": 0.8112449,
      "learning_rate": 3.7505679454456992e-06,
      "loss": 0.83362842,
      "num_input_tokens_seen": 32842125,
      "step": 1546,
      "time_per_iteration": 2.501757860183716
    },
    {
      "auxiliary_loss_clip": 0.01200732,
      "auxiliary_loss_mlp": 0.01038147,
      "balance_loss_clip": 1.06433439,
      "balance_loss_mlp": 1.02625561,
      "epoch": 0.18601575181867372,
      "flos": 23549966726400.0,
      "grad_norm": 2.251644142322038,
      "language_loss": 0.69787717,
      "learning_rate": 3.750191094732564e-06,
      "loss": 0.72026598,
      "num_input_tokens_seen": 32862990,
      "step": 1547,
      "time_per_iteration": 2.571782112121582
    },
    {
      "auxiliary_loss_clip": 0.01169866,
      "auxiliary_loss_mlp": 0.01040839,
      "balance_loss_clip": 1.06191289,
      "balance_loss_mlp": 1.02878726,
      "epoch": 0.1861359947093128,
      "flos": 26360192108160.0,
      "grad_norm": 2.0476959112009627,
      "language_loss": 0.75689256,
      "learning_rate": 3.7498139785168313e-06,
      "loss": 0.77899963,
      "num_input_tokens_seen": 32883595,
      "step": 1548,
      "time_per_iteration": 2.6489315032958984
    },
    {
      "auxiliary_loss_clip": 0.01202742,
      "auxiliary_loss_mlp": 0.01037135,
      "balance_loss_clip": 1.06343913,
      "balance_loss_mlp": 1.02571452,
      "epoch": 0.1862562375999519,
      "flos": 23331198942720.0,
      "grad_norm": 1.6454322333690816,
      "language_loss": 0.7717247,
      "learning_rate": 3.749436596855709e-06,
      "loss": 0.79412353,
      "num_input_tokens_seen": 32902895,
      "step": 1549,
      "time_per_iteration": 2.5372540950775146
    },
    {
      "auxiliary_loss_clip": 0.0115717,
      "auxiliary_loss_mlp": 0.00765809,
      "balance_loss_clip": 1.05253339,
      "balance_loss_mlp": 1.0006094,
      "epoch": 0.186376480490591,
      "flos": 16648222942080.0,
      "grad_norm": 1.8881414475025435,
      "language_loss": 0.90650147,
      "learning_rate": 3.749058949806446e-06,
      "loss": 0.9257313,
      "num_input_tokens_seen": 32919620,
      "step": 1550,
      "time_per_iteration": 2.5446083545684814
    },
    {
      "auxiliary_loss_clip": 0.01171582,
      "auxiliary_loss_mlp": 0.01032803,
      "balance_loss_clip": 1.05461895,
      "balance_loss_mlp": 1.02147782,
      "epoch": 0.18649672338123008,
      "flos": 21468656039040.0,
      "grad_norm": 1.712053009200074,
      "language_loss": 0.84466255,
      "learning_rate": 3.748681037426331e-06,
      "loss": 0.86670637,
      "num_input_tokens_seen": 32938830,
      "step": 1551,
      "time_per_iteration": 2.557217836380005
    },
    {
      "auxiliary_loss_clip": 0.01187292,
      "auxiliary_loss_mlp": 0.01032973,
      "balance_loss_clip": 1.06062102,
      "balance_loss_mlp": 1.02207136,
      "epoch": 0.1866169662718692,
      "flos": 12312728386560.0,
      "grad_norm": 2.2179061899930916,
      "language_loss": 0.91466683,
      "learning_rate": 3.7483028597726936e-06,
      "loss": 0.9368695,
      "num_input_tokens_seen": 32955600,
      "step": 1552,
      "time_per_iteration": 2.52274227142334
    },
    {
      "auxiliary_loss_clip": 0.01198051,
      "auxiliary_loss_mlp": 0.01032666,
      "balance_loss_clip": 1.06649268,
      "balance_loss_mlp": 1.02066755,
      "epoch": 0.18673720916250827,
      "flos": 23581280407680.0,
      "grad_norm": 1.8012870144244082,
      "language_loss": 0.62331009,
      "learning_rate": 3.7479244169029017e-06,
      "loss": 0.64561725,
      "num_input_tokens_seen": 32975390,
      "step": 1553,
      "time_per_iteration": 2.57204270362854
    },
    {
      "auxiliary_loss_clip": 0.01160664,
      "auxiliary_loss_mlp": 0.01032063,
      "balance_loss_clip": 1.0604701,
      "balance_loss_mlp": 1.0201478,
      "epoch": 0.18685745205314735,
      "flos": 19718370115200.0,
      "grad_norm": 2.654873972809411,
      "language_loss": 0.73370039,
      "learning_rate": 3.7475457088743658e-06,
      "loss": 0.75562763,
      "num_input_tokens_seen": 32992640,
      "step": 1554,
      "time_per_iteration": 2.604414939880371
    },
    {
      "auxiliary_loss_clip": 0.01175638,
      "auxiliary_loss_mlp": 0.01035983,
      "balance_loss_clip": 1.05858731,
      "balance_loss_mlp": 1.02396643,
      "epoch": 0.18697769494378644,
      "flos": 34204123589760.0,
      "grad_norm": 1.7719059729466975,
      "language_loss": 0.74713182,
      "learning_rate": 3.7471667357445348e-06,
      "loss": 0.76924807,
      "num_input_tokens_seen": 33012470,
      "step": 1555,
      "time_per_iteration": 2.634336471557617
    },
    {
      "auxiliary_loss_clip": 0.01202601,
      "auxiliary_loss_mlp": 0.01036384,
      "balance_loss_clip": 1.05919194,
      "balance_loss_mlp": 1.02458191,
      "epoch": 0.18709793783442555,
      "flos": 34241327101440.0,
      "grad_norm": 1.8069172735021315,
      "language_loss": 0.7227543,
      "learning_rate": 3.7467874975709e-06,
      "loss": 0.74514413,
      "num_input_tokens_seen": 33033275,
      "step": 1556,
      "time_per_iteration": 2.599592685699463
    },
    {
      "auxiliary_loss_clip": 0.01196693,
      "auxiliary_loss_mlp": 0.01039062,
      "balance_loss_clip": 1.06553602,
      "balance_loss_mlp": 1.02823222,
      "epoch": 0.18721818072506463,
      "flos": 40734550529280.0,
      "grad_norm": 1.9872203474736396,
      "language_loss": 0.78310478,
      "learning_rate": 3.7464079944109904e-06,
      "loss": 0.80546224,
      "num_input_tokens_seen": 33055135,
      "step": 1557,
      "time_per_iteration": 2.693528413772583
    },
    {
      "auxiliary_loss_clip": 0.01155948,
      "auxiliary_loss_mlp": 0.01037502,
      "balance_loss_clip": 1.05687237,
      "balance_loss_mlp": 1.02605748,
      "epoch": 0.18733842361570371,
      "flos": 22157386392960.0,
      "grad_norm": 1.9053100331969772,
      "language_loss": 0.77612484,
      "learning_rate": 3.746028226322376e-06,
      "loss": 0.79805934,
      "num_input_tokens_seen": 33071015,
      "step": 1558,
      "time_per_iteration": 2.625192642211914
    },
    {
      "auxiliary_loss_clip": 0.01178464,
      "auxiliary_loss_mlp": 0.01035775,
      "balance_loss_clip": 1.05820453,
      "balance_loss_mlp": 1.02423596,
      "epoch": 0.18745866650634282,
      "flos": 18914940656640.0,
      "grad_norm": 1.9410018946654377,
      "language_loss": 0.75252891,
      "learning_rate": 3.745648193362669e-06,
      "loss": 0.77467132,
      "num_input_tokens_seen": 33090370,
      "step": 1559,
      "time_per_iteration": 2.5726397037506104
    },
    {
      "auxiliary_loss_clip": 0.0119327,
      "auxiliary_loss_mlp": 0.0103672,
      "balance_loss_clip": 1.06262553,
      "balance_loss_mlp": 1.02555633,
      "epoch": 0.1875789093969819,
      "flos": 19314626267520.0,
      "grad_norm": 2.070685778609955,
      "language_loss": 0.71926463,
      "learning_rate": 3.745267895589518e-06,
      "loss": 0.74156463,
      "num_input_tokens_seen": 33108910,
      "step": 1560,
      "time_per_iteration": 2.522773504257202
    },
    {
      "auxiliary_loss_clip": 0.01162906,
      "auxiliary_loss_mlp": 0.01031659,
      "balance_loss_clip": 1.06012392,
      "balance_loss_mlp": 1.02021456,
      "epoch": 0.187699152287621,
      "flos": 17018965169280.0,
      "grad_norm": 2.2827595486932832,
      "language_loss": 0.81958115,
      "learning_rate": 3.7448873330606154e-06,
      "loss": 0.84152675,
      "num_input_tokens_seen": 33126680,
      "step": 1561,
      "time_per_iteration": 2.5548717975616455
    },
    {
      "auxiliary_loss_clip": 0.01163553,
      "auxiliary_loss_mlp": 0.01039352,
      "balance_loss_clip": 1.06278241,
      "balance_loss_mlp": 1.0272646,
      "epoch": 0.18781939517826007,
      "flos": 22346384780160.0,
      "grad_norm": 2.038657077811118,
      "language_loss": 0.87466693,
      "learning_rate": 3.7445065058336914e-06,
      "loss": 0.89669603,
      "num_input_tokens_seen": 33145550,
      "step": 1562,
      "time_per_iteration": 2.6198549270629883
    },
    {
      "auxiliary_loss_clip": 0.01205246,
      "auxiliary_loss_mlp": 0.01033384,
      "balance_loss_clip": 1.06349134,
      "balance_loss_mlp": 1.02263117,
      "epoch": 0.18793963806889918,
      "flos": 14611478054400.0,
      "grad_norm": 1.8730276564220993,
      "language_loss": 0.86469358,
      "learning_rate": 3.7441254139665176e-06,
      "loss": 0.88707983,
      "num_input_tokens_seen": 33161735,
      "step": 1563,
      "time_per_iteration": 2.462156295776367
    },
    {
      "auxiliary_loss_clip": 0.01209535,
      "auxiliary_loss_mlp": 0.01040724,
      "balance_loss_clip": 1.06729841,
      "balance_loss_mlp": 1.0295186,
      "epoch": 0.18805988095953827,
      "flos": 17457075354240.0,
      "grad_norm": 3.221152108795195,
      "language_loss": 0.82387131,
      "learning_rate": 3.743744057516905e-06,
      "loss": 0.84637392,
      "num_input_tokens_seen": 33179795,
      "step": 1564,
      "time_per_iteration": 2.4737796783447266
    },
    {
      "auxiliary_loss_clip": 0.01192123,
      "auxiliary_loss_mlp": 0.01036901,
      "balance_loss_clip": 1.06387866,
      "balance_loss_mlp": 1.02558815,
      "epoch": 0.18818012385017735,
      "flos": 15043877976960.0,
      "grad_norm": 4.2919596674166955,
      "language_loss": 0.88505733,
      "learning_rate": 3.743362436542706e-06,
      "loss": 0.90734756,
      "num_input_tokens_seen": 33194485,
      "step": 1565,
      "time_per_iteration": 2.4802544116973877
    },
    {
      "auxiliary_loss_clip": 0.01182411,
      "auxiliary_loss_mlp": 0.01028579,
      "balance_loss_clip": 1.05223811,
      "balance_loss_mlp": 1.01804662,
      "epoch": 0.18830036674081646,
      "flos": 47551975136640.0,
      "grad_norm": 1.7913411451998487,
      "language_loss": 0.76380813,
      "learning_rate": 3.7429805511018115e-06,
      "loss": 0.785918,
      "num_input_tokens_seen": 33216145,
      "step": 1566,
      "time_per_iteration": 3.5419135093688965
    },
    {
      "auxiliary_loss_clip": 0.01172485,
      "auxiliary_loss_mlp": 0.01036886,
      "balance_loss_clip": 1.05979681,
      "balance_loss_mlp": 1.02447677,
      "epoch": 0.18842060963145554,
      "flos": 30044626698240.0,
      "grad_norm": 1.7778222250066278,
      "language_loss": 0.78077126,
      "learning_rate": 3.7425984012521524e-06,
      "loss": 0.80286497,
      "num_input_tokens_seen": 33236345,
      "step": 1567,
      "time_per_iteration": 2.6235768795013428
    },
    {
      "auxiliary_loss_clip": 0.01116558,
      "auxiliary_loss_mlp": 0.00759106,
      "balance_loss_clip": 1.04813147,
      "balance_loss_mlp": 1.00112331,
      "epoch": 0.18854085252209463,
      "flos": 70318372625280.0,
      "grad_norm": 0.7392375009942009,
      "language_loss": 0.60438496,
      "learning_rate": 3.7422159870517025e-06,
      "loss": 0.62314159,
      "num_input_tokens_seen": 33301600,
      "step": 1568,
      "time_per_iteration": 3.1549465656280518
    },
    {
      "auxiliary_loss_clip": 0.010868,
      "auxiliary_loss_mlp": 0.01037063,
      "balance_loss_clip": 1.04680562,
      "balance_loss_mlp": 1.02616727,
      "epoch": 0.1886610954127337,
      "flos": 21289318410240.0,
      "grad_norm": 1.9138441115796982,
      "language_loss": 0.79256916,
      "learning_rate": 3.7418333085584717e-06,
      "loss": 0.81380779,
      "num_input_tokens_seen": 33322785,
      "step": 1569,
      "time_per_iteration": 3.7133326530456543
    },
    {
      "auxiliary_loss_clip": 0.01124575,
      "auxiliary_loss_mlp": 0.01040279,
      "balance_loss_clip": 1.05028951,
      "balance_loss_mlp": 1.0288173,
      "epoch": 0.18878133830337282,
      "flos": 17266819991040.0,
      "grad_norm": 2.232079279919486,
      "language_loss": 0.90509224,
      "learning_rate": 3.7414503658305128e-06,
      "loss": 0.92674077,
      "num_input_tokens_seen": 33340020,
      "step": 1570,
      "time_per_iteration": 2.8805487155914307
    },
    {
      "auxiliary_loss_clip": 0.01186317,
      "auxiliary_loss_mlp": 0.00764789,
      "balance_loss_clip": 1.05708671,
      "balance_loss_mlp": 1.00037384,
      "epoch": 0.1889015811940119,
      "flos": 25775207210880.0,
      "grad_norm": 2.271685955164892,
      "language_loss": 0.77894408,
      "learning_rate": 3.7410671589259185e-06,
      "loss": 0.79845512,
      "num_input_tokens_seen": 33358620,
      "step": 1571,
      "time_per_iteration": 3.5884220600128174
    },
    {
      "auxiliary_loss_clip": 0.01182058,
      "auxiliary_loss_mlp": 0.01038332,
      "balance_loss_clip": 1.06456923,
      "balance_loss_mlp": 1.02692366,
      "epoch": 0.18902182408465099,
      "flos": 21032197879680.0,
      "grad_norm": 1.7807604022606673,
      "language_loss": 0.79373628,
      "learning_rate": 3.7406836879028205e-06,
      "loss": 0.81594014,
      "num_input_tokens_seen": 33378845,
      "step": 1572,
      "time_per_iteration": 2.6047542095184326
    },
    {
      "auxiliary_loss_clip": 0.01190561,
      "auxiliary_loss_mlp": 0.0103238,
      "balance_loss_clip": 1.05596888,
      "balance_loss_mlp": 1.02144814,
      "epoch": 0.1891420669752901,
      "flos": 22272121411200.0,
      "grad_norm": 2.3018654646273693,
      "language_loss": 0.76814854,
      "learning_rate": 3.7402999528193907e-06,
      "loss": 0.79037791,
      "num_input_tokens_seen": 33398345,
      "step": 1573,
      "time_per_iteration": 2.557008981704712
    },
    {
      "auxiliary_loss_clip": 0.01163479,
      "auxiliary_loss_mlp": 0.0103284,
      "balance_loss_clip": 1.05811167,
      "balance_loss_mlp": 1.02152741,
      "epoch": 0.18926230986592918,
      "flos": 22017802141440.0,
      "grad_norm": 2.2090402749067928,
      "language_loss": 0.85516495,
      "learning_rate": 3.739915953733842e-06,
      "loss": 0.87712812,
      "num_input_tokens_seen": 33416390,
      "step": 1574,
      "time_per_iteration": 2.572950839996338
    },
    {
      "auxiliary_loss_clip": 0.01118891,
      "auxiliary_loss_mlp": 0.01036319,
      "balance_loss_clip": 1.04814529,
      "balance_loss_mlp": 1.02488708,
      "epoch": 0.18938255275656826,
      "flos": 24462672336000.0,
      "grad_norm": 1.592952800358907,
      "language_loss": 0.82032806,
      "learning_rate": 3.7395316907044264e-06,
      "loss": 0.84188014,
      "num_input_tokens_seen": 33437175,
      "step": 1575,
      "time_per_iteration": 2.67531418800354
    },
    {
      "auxiliary_loss_clip": 0.01195578,
      "auxiliary_loss_mlp": 0.01033161,
      "balance_loss_clip": 1.05963492,
      "balance_loss_mlp": 1.02187753,
      "epoch": 0.18950279564720737,
      "flos": 24427049022720.0,
      "grad_norm": 1.5921471044445694,
      "language_loss": 0.79576766,
      "learning_rate": 3.7391471637894364e-06,
      "loss": 0.81805503,
      "num_input_tokens_seen": 33459440,
      "step": 1576,
      "time_per_iteration": 2.57173490524292
    },
    {
      "auxiliary_loss_clip": 0.01220519,
      "auxiliary_loss_mlp": 0.01029621,
      "balance_loss_clip": 1.06355643,
      "balance_loss_mlp": 1.01879692,
      "epoch": 0.18962303853784646,
      "flos": 19756291898880.0,
      "grad_norm": 1.8148660682889146,
      "language_loss": 0.85077697,
      "learning_rate": 3.738762373047205e-06,
      "loss": 0.87327838,
      "num_input_tokens_seen": 33479360,
      "step": 1577,
      "time_per_iteration": 2.479051113128662
    },
    {
      "auxiliary_loss_clip": 0.01132214,
      "auxiliary_loss_mlp": 0.0076501,
      "balance_loss_clip": 1.04845071,
      "balance_loss_mlp": 1.00052571,
      "epoch": 0.18974328142848554,
      "flos": 21032054225280.0,
      "grad_norm": 1.6073382154593447,
      "language_loss": 0.83395118,
      "learning_rate": 3.738377318536103e-06,
      "loss": 0.85292339,
      "num_input_tokens_seen": 33499245,
      "step": 1578,
      "time_per_iteration": 2.6197404861450195
    },
    {
      "auxiliary_loss_clip": 0.01175897,
      "auxiliary_loss_mlp": 0.0103756,
      "balance_loss_clip": 1.05172276,
      "balance_loss_mlp": 1.0279814,
      "epoch": 0.18986352431912462,
      "flos": 12966122736000.0,
      "grad_norm": 2.278068529170855,
      "language_loss": 0.71892726,
      "learning_rate": 3.7379920003145447e-06,
      "loss": 0.74106181,
      "num_input_tokens_seen": 33513520,
      "step": 1579,
      "time_per_iteration": 2.5232062339782715
    },
    {
      "auxiliary_loss_clip": 0.01173904,
      "auxiliary_loss_mlp": 0.00765767,
      "balance_loss_clip": 1.06026709,
      "balance_loss_mlp": 1.00056779,
      "epoch": 0.18998376720976373,
      "flos": 23767908497280.0,
      "grad_norm": 1.8923465427562136,
      "language_loss": 0.83873242,
      "learning_rate": 3.7376064184409817e-06,
      "loss": 0.85812914,
      "num_input_tokens_seen": 33533100,
      "step": 1580,
      "time_per_iteration": 2.5846469402313232
    },
    {
      "auxiliary_loss_clip": 0.01222548,
      "auxiliary_loss_mlp": 0.01037817,
      "balance_loss_clip": 1.06552291,
      "balance_loss_mlp": 1.02689123,
      "epoch": 0.19010401010040281,
      "flos": 22966023323520.0,
      "grad_norm": 1.4009966035969492,
      "language_loss": 0.87112033,
      "learning_rate": 3.7372205729739063e-06,
      "loss": 0.89372396,
      "num_input_tokens_seen": 33554915,
      "step": 1581,
      "time_per_iteration": 2.4865758419036865
    },
    {
      "auxiliary_loss_clip": 0.01097246,
      "auxiliary_loss_mlp": 0.01040197,
      "balance_loss_clip": 1.04324508,
      "balance_loss_mlp": 1.02799547,
      "epoch": 0.1902242529910419,
      "flos": 19135647774720.0,
      "grad_norm": 2.605439079748829,
      "language_loss": 0.71790504,
      "learning_rate": 3.7368344639718514e-06,
      "loss": 0.73927951,
      "num_input_tokens_seen": 33572850,
      "step": 1582,
      "time_per_iteration": 2.706639528274536
    },
    {
      "auxiliary_loss_clip": 0.01167008,
      "auxiliary_loss_mlp": 0.00764545,
      "balance_loss_clip": 1.05535209,
      "balance_loss_mlp": 1.00038004,
      "epoch": 0.190344495881681,
      "flos": 25483935824640.0,
      "grad_norm": 1.6775370199863426,
      "language_loss": 0.80259025,
      "learning_rate": 3.7364480914933895e-06,
      "loss": 0.82190579,
      "num_input_tokens_seen": 33593090,
      "step": 1583,
      "time_per_iteration": 2.910705804824829
    },
    {
      "auxiliary_loss_clip": 0.01134997,
      "auxiliary_loss_mlp": 0.01039556,
      "balance_loss_clip": 1.0520426,
      "balance_loss_mlp": 1.02848744,
      "epoch": 0.1904647387723201,
      "flos": 26792843425920.0,
      "grad_norm": 1.7892120552343211,
      "language_loss": 0.81377506,
      "learning_rate": 3.7360614555971325e-06,
      "loss": 0.83552057,
      "num_input_tokens_seen": 33612745,
      "step": 1584,
      "time_per_iteration": 2.7063040733337402
    },
    {
      "auxiliary_loss_clip": 0.01148683,
      "auxiliary_loss_mlp": 0.01035174,
      "balance_loss_clip": 1.05106592,
      "balance_loss_mlp": 1.02424288,
      "epoch": 0.19058498166295917,
      "flos": 23987753688960.0,
      "grad_norm": 1.8716584272685584,
      "language_loss": 0.85336292,
      "learning_rate": 3.735674556341733e-06,
      "loss": 0.87520146,
      "num_input_tokens_seen": 33632360,
      "step": 1585,
      "time_per_iteration": 2.6252026557922363
    },
    {
      "auxiliary_loss_clip": 0.01163609,
      "auxiliary_loss_mlp": 0.01030803,
      "balance_loss_clip": 1.05457401,
      "balance_loss_mlp": 1.01935291,
      "epoch": 0.19070522455359826,
      "flos": 28293299280000.0,
      "grad_norm": 1.8983628532199,
      "language_loss": 0.82430673,
      "learning_rate": 3.7352873937858835e-06,
      "loss": 0.84625077,
      "num_input_tokens_seen": 33653895,
      "step": 1586,
      "time_per_iteration": 2.6255366802215576
    },
    {
      "auxiliary_loss_clip": 0.01149156,
      "auxiliary_loss_mlp": 0.01035026,
      "balance_loss_clip": 1.05575943,
      "balance_loss_mlp": 1.02281857,
      "epoch": 0.19082546744423737,
      "flos": 25660220797440.0,
      "grad_norm": 1.894374767759452,
      "language_loss": 0.72126758,
      "learning_rate": 3.734899967988316e-06,
      "loss": 0.74310946,
      "num_input_tokens_seen": 33672075,
      "step": 1587,
      "time_per_iteration": 2.618715524673462
    },
    {
      "auxiliary_loss_clip": 0.01206953,
      "auxiliary_loss_mlp": 0.00763614,
      "balance_loss_clip": 1.06364691,
      "balance_loss_mlp": 1.00047934,
      "epoch": 0.19094571033487645,
      "flos": 19719483436800.0,
      "grad_norm": 2.3131522226514125,
      "language_loss": 0.83707547,
      "learning_rate": 3.7345122790078026e-06,
      "loss": 0.85678113,
      "num_input_tokens_seen": 33689640,
      "step": 1588,
      "time_per_iteration": 2.523834466934204
    },
    {
      "auxiliary_loss_clip": 0.01185232,
      "auxiliary_loss_mlp": 0.01036503,
      "balance_loss_clip": 1.05877423,
      "balance_loss_mlp": 1.02477241,
      "epoch": 0.19106595322551553,
      "flos": 21616320850560.0,
      "grad_norm": 3.288392812331841,
      "language_loss": 0.93100131,
      "learning_rate": 3.7341243269031556e-06,
      "loss": 0.95321858,
      "num_input_tokens_seen": 33708630,
      "step": 1589,
      "time_per_iteration": 2.5163612365722656
    },
    {
      "auxiliary_loss_clip": 0.0118728,
      "auxiliary_loss_mlp": 0.01037732,
      "balance_loss_clip": 1.06237113,
      "balance_loss_mlp": 1.02663338,
      "epoch": 0.19118619611615464,
      "flos": 29896890059520.0,
      "grad_norm": 1.5034527238573623,
      "language_loss": 0.77410328,
      "learning_rate": 3.7337361117332275e-06,
      "loss": 0.7963534,
      "num_input_tokens_seen": 33730370,
      "step": 1590,
      "time_per_iteration": 2.5849359035491943
    },
    {
      "auxiliary_loss_clip": 0.0114319,
      "auxiliary_loss_mlp": 0.01036072,
      "balance_loss_clip": 1.05543792,
      "balance_loss_mlp": 1.02569485,
      "epoch": 0.19130643900679373,
      "flos": 17273428093440.0,
      "grad_norm": 1.909056665218753,
      "language_loss": 0.77229387,
      "learning_rate": 3.7333476335569087e-06,
      "loss": 0.79408646,
      "num_input_tokens_seen": 33748370,
      "step": 1591,
      "time_per_iteration": 3.355003833770752
    },
    {
      "auxiliary_loss_clip": 0.01198333,
      "auxiliary_loss_mlp": 0.00764993,
      "balance_loss_clip": 1.05973423,
      "balance_loss_mlp": 1.0004338,
      "epoch": 0.1914266818974328,
      "flos": 24826339584000.0,
      "grad_norm": 2.4536220788393117,
      "language_loss": 0.6728735,
      "learning_rate": 3.7329588924331325e-06,
      "loss": 0.69250673,
      "num_input_tokens_seen": 33769575,
      "step": 1592,
      "time_per_iteration": 2.5622951984405518
    },
    {
      "auxiliary_loss_clip": 0.01150916,
      "auxiliary_loss_mlp": 0.01041927,
      "balance_loss_clip": 1.05507958,
      "balance_loss_mlp": 1.03096604,
      "epoch": 0.1915469247880719,
      "flos": 18952467390720.0,
      "grad_norm": 1.8388359355660613,
      "language_loss": 0.82501745,
      "learning_rate": 3.732569888420871e-06,
      "loss": 0.84694588,
      "num_input_tokens_seen": 33789110,
      "step": 1593,
      "time_per_iteration": 2.610780715942383
    },
    {
      "auxiliary_loss_clip": 0.01167248,
      "auxiliary_loss_mlp": 0.01037827,
      "balance_loss_clip": 1.04843533,
      "balance_loss_mlp": 1.02675259,
      "epoch": 0.191667167678711,
      "flos": 21032952065280.0,
      "grad_norm": 4.742024050854939,
      "language_loss": 0.82742941,
      "learning_rate": 3.732180621579134e-06,
      "loss": 0.84948021,
      "num_input_tokens_seen": 33808325,
      "step": 1594,
      "time_per_iteration": 2.5262887477874756
    },
    {
      "auxiliary_loss_clip": 0.0120272,
      "auxiliary_loss_mlp": 0.01038526,
      "balance_loss_clip": 1.06353283,
      "balance_loss_mlp": 1.02723074,
      "epoch": 0.1917874105693501,
      "flos": 34237663914240.0,
      "grad_norm": 2.376522751101407,
      "language_loss": 0.81384802,
      "learning_rate": 3.7317910919669745e-06,
      "loss": 0.83626044,
      "num_input_tokens_seen": 33829520,
      "step": 1595,
      "time_per_iteration": 2.605659246444702
    },
    {
      "auxiliary_loss_clip": 0.01119818,
      "auxiliary_loss_mlp": 0.00764725,
      "balance_loss_clip": 1.04930615,
      "balance_loss_mlp": 1.00052285,
      "epoch": 0.19190765345998917,
      "flos": 23550613171200.0,
      "grad_norm": 2.043622041866764,
      "language_loss": 0.76767147,
      "learning_rate": 3.7314012996434826e-06,
      "loss": 0.7865169,
      "num_input_tokens_seen": 33848250,
      "step": 1596,
      "time_per_iteration": 3.3455212116241455
    },
    {
      "auxiliary_loss_clip": 0.01199923,
      "auxiliary_loss_mlp": 0.01025978,
      "balance_loss_clip": 1.06073308,
      "balance_loss_mlp": 1.01598799,
      "epoch": 0.19202789635062828,
      "flos": 19861330245120.0,
      "grad_norm": 2.0374411631941527,
      "language_loss": 0.80903804,
      "learning_rate": 3.7310112446677907e-06,
      "loss": 0.83129704,
      "num_input_tokens_seen": 33866160,
      "step": 1597,
      "time_per_iteration": 3.4313881397247314
    },
    {
      "auxiliary_loss_clip": 0.01168607,
      "auxiliary_loss_mlp": 0.01040043,
      "balance_loss_clip": 1.05923426,
      "balance_loss_mlp": 1.02967131,
      "epoch": 0.19214813924126736,
      "flos": 20922957642240.0,
      "grad_norm": 2.245723417578427,
      "language_loss": 0.68835199,
      "learning_rate": 3.7306209270990695e-06,
      "loss": 0.71043849,
      "num_input_tokens_seen": 33884165,
      "step": 1598,
      "time_per_iteration": 2.5351274013519287
    },
    {
      "auxiliary_loss_clip": 0.01139155,
      "auxiliary_loss_mlp": 0.01029003,
      "balance_loss_clip": 1.05063343,
      "balance_loss_mlp": 1.01962137,
      "epoch": 0.19226838213190645,
      "flos": 26359725231360.0,
      "grad_norm": 1.8732995603849365,
      "language_loss": 0.86810529,
      "learning_rate": 3.7302303469965292e-06,
      "loss": 0.88978684,
      "num_input_tokens_seen": 33903705,
      "step": 1599,
      "time_per_iteration": 2.6670050621032715
    },
    {
      "auxiliary_loss_clip": 0.01205278,
      "auxiliary_loss_mlp": 0.01041221,
      "balance_loss_clip": 1.06168699,
      "balance_loss_mlp": 1.0301826,
      "epoch": 0.19238862502254553,
      "flos": 20850525866880.0,
      "grad_norm": 1.7693120069420885,
      "language_loss": 0.70854211,
      "learning_rate": 3.7298395044194206e-06,
      "loss": 0.7310071,
      "num_input_tokens_seen": 33922515,
      "step": 1600,
      "time_per_iteration": 2.5003368854522705
    },
    {
      "auxiliary_loss_clip": 0.0117584,
      "auxiliary_loss_mlp": 0.01031373,
      "balance_loss_clip": 1.06019449,
      "balance_loss_mlp": 1.0205369,
      "epoch": 0.19250886791318464,
      "flos": 21726063878400.0,
      "grad_norm": 1.996579387901125,
      "language_loss": 0.94505787,
      "learning_rate": 3.7294483994270356e-06,
      "loss": 0.96713001,
      "num_input_tokens_seen": 33940840,
      "step": 1601,
      "time_per_iteration": 2.5302586555480957
    },
    {
      "auxiliary_loss_clip": 0.01171201,
      "auxiliary_loss_mlp": 0.01035593,
      "balance_loss_clip": 1.05360162,
      "balance_loss_mlp": 1.02599061,
      "epoch": 0.19262911080382372,
      "flos": 23367827836800.0,
      "grad_norm": 2.1792676767637107,
      "language_loss": 0.7812553,
      "learning_rate": 3.7290570320787033e-06,
      "loss": 0.80332327,
      "num_input_tokens_seen": 33960420,
      "step": 1602,
      "time_per_iteration": 2.563086986541748
    },
    {
      "auxiliary_loss_clip": 0.01171395,
      "auxiliary_loss_mlp": 0.01031321,
      "balance_loss_clip": 1.05767035,
      "balance_loss_mlp": 1.02055657,
      "epoch": 0.1927493536944628,
      "flos": 21943502858880.0,
      "grad_norm": 2.3833595069771785,
      "language_loss": 0.71725017,
      "learning_rate": 3.728665402433793e-06,
      "loss": 0.7392773,
      "num_input_tokens_seen": 33978990,
      "step": 1603,
      "time_per_iteration": 2.544039726257324
    },
    {
      "auxiliary_loss_clip": 0.01182009,
      "auxiliary_loss_mlp": 0.01039465,
      "balance_loss_clip": 1.05626285,
      "balance_loss_mlp": 1.02948129,
      "epoch": 0.19286959658510192,
      "flos": 16545590807040.0,
      "grad_norm": 2.4409024732744404,
      "language_loss": 0.86034429,
      "learning_rate": 3.7282735105517164e-06,
      "loss": 0.88255894,
      "num_input_tokens_seen": 33997115,
      "step": 1604,
      "time_per_iteration": 2.501490354537964
    },
    {
      "auxiliary_loss_clip": 0.01180988,
      "auxiliary_loss_mlp": 0.01035769,
      "balance_loss_clip": 1.05740511,
      "balance_loss_mlp": 1.02486753,
      "epoch": 0.192989839475741,
      "flos": 21616967295360.0,
      "grad_norm": 3.101987004551368,
      "language_loss": 0.67254436,
      "learning_rate": 3.727881356491922e-06,
      "loss": 0.69471192,
      "num_input_tokens_seen": 34015525,
      "step": 1605,
      "time_per_iteration": 2.5271897315979004
    },
    {
      "auxiliary_loss_clip": 0.01192478,
      "auxiliary_loss_mlp": 0.0103274,
      "balance_loss_clip": 1.06297362,
      "balance_loss_mlp": 1.02148676,
      "epoch": 0.19311008236638008,
      "flos": 19281516906240.0,
      "grad_norm": 1.8277092298372495,
      "language_loss": 0.76131427,
      "learning_rate": 3.7274889403139002e-06,
      "loss": 0.78356636,
      "num_input_tokens_seen": 34033150,
      "step": 1606,
      "time_per_iteration": 2.5044093132019043
    },
    {
      "auxiliary_loss_clip": 0.01156046,
      "auxiliary_loss_mlp": 0.01035643,
      "balance_loss_clip": 1.05055165,
      "balance_loss_mlp": 1.0251826,
      "epoch": 0.1932303252570192,
      "flos": 28652369587200.0,
      "grad_norm": 2.1857668812385107,
      "language_loss": 0.78154027,
      "learning_rate": 3.727096262077179e-06,
      "loss": 0.80345714,
      "num_input_tokens_seen": 34052145,
      "step": 1607,
      "time_per_iteration": 2.6053102016448975
    },
    {
      "auxiliary_loss_clip": 0.01175874,
      "auxiliary_loss_mlp": 0.01030967,
      "balance_loss_clip": 1.05590999,
      "balance_loss_mlp": 1.02079844,
      "epoch": 0.19335056814765827,
      "flos": 18368990864640.0,
      "grad_norm": 1.698610840395594,
      "language_loss": 0.8521592,
      "learning_rate": 3.7267033218413285e-06,
      "loss": 0.87422764,
      "num_input_tokens_seen": 34069940,
      "step": 1608,
      "time_per_iteration": 2.5626220703125
    },
    {
      "auxiliary_loss_clip": 0.01128823,
      "auxiliary_loss_mlp": 0.010371,
      "balance_loss_clip": 1.0491606,
      "balance_loss_mlp": 1.02656209,
      "epoch": 0.19347081103829736,
      "flos": 13260877741440.0,
      "grad_norm": 1.9857339871709656,
      "language_loss": 0.8125149,
      "learning_rate": 3.726310119665957e-06,
      "loss": 0.83417416,
      "num_input_tokens_seen": 34086275,
      "step": 1609,
      "time_per_iteration": 2.5841500759124756
    },
    {
      "auxiliary_loss_clip": 0.01163497,
      "auxiliary_loss_mlp": 0.0103716,
      "balance_loss_clip": 1.05720234,
      "balance_loss_mlp": 1.02655649,
      "epoch": 0.19359105392893644,
      "flos": 20300122788480.0,
      "grad_norm": 1.7725899009247434,
      "language_loss": 0.85540771,
      "learning_rate": 3.725916655610713e-06,
      "loss": 0.87741423,
      "num_input_tokens_seen": 34105605,
      "step": 1610,
      "time_per_iteration": 2.592543840408325
    },
    {
      "auxiliary_loss_clip": 0.01135158,
      "auxiliary_loss_mlp": 0.01033371,
      "balance_loss_clip": 1.04863,
      "balance_loss_mlp": 1.02229011,
      "epoch": 0.19371129681957555,
      "flos": 20484596062080.0,
      "grad_norm": 2.562501658675675,
      "language_loss": 0.75245857,
      "learning_rate": 3.725522929735284e-06,
      "loss": 0.77414382,
      "num_input_tokens_seen": 34122540,
      "step": 1611,
      "time_per_iteration": 2.6009628772735596
    },
    {
      "auxiliary_loss_clip": 0.01152888,
      "auxiliary_loss_mlp": 0.01040312,
      "balance_loss_clip": 1.05486393,
      "balance_loss_mlp": 1.03014326,
      "epoch": 0.19383153971021463,
      "flos": 30445497457920.0,
      "grad_norm": 2.207600392171983,
      "language_loss": 0.74196804,
      "learning_rate": 3.725128942099399e-06,
      "loss": 0.76390004,
      "num_input_tokens_seen": 34142940,
      "step": 1612,
      "time_per_iteration": 2.6987464427948
    },
    {
      "auxiliary_loss_clip": 0.01149988,
      "auxiliary_loss_mlp": 0.01039712,
      "balance_loss_clip": 1.05512619,
      "balance_loss_mlp": 1.02822578,
      "epoch": 0.19395178260085372,
      "flos": 24569937325440.0,
      "grad_norm": 2.263168545601864,
      "language_loss": 0.80142641,
      "learning_rate": 3.7247346927628245e-06,
      "loss": 0.82332343,
      "num_input_tokens_seen": 34162875,
      "step": 1613,
      "time_per_iteration": 2.6227493286132812
    },
    {
      "auxiliary_loss_clip": 0.01173481,
      "auxiliary_loss_mlp": 0.01033621,
      "balance_loss_clip": 1.05305672,
      "balance_loss_mlp": 1.02305341,
      "epoch": 0.19407202549149283,
      "flos": 28950608211840.0,
      "grad_norm": 1.7944035361746107,
      "language_loss": 0.791731,
      "learning_rate": 3.7243401817853694e-06,
      "loss": 0.81380194,
      "num_input_tokens_seen": 34183565,
      "step": 1614,
      "time_per_iteration": 2.6134984493255615
    },
    {
      "auxiliary_loss_clip": 0.01190188,
      "auxiliary_loss_mlp": 0.01042975,
      "balance_loss_clip": 1.05532598,
      "balance_loss_mlp": 1.03307498,
      "epoch": 0.1941922683821319,
      "flos": 18004497603840.0,
      "grad_norm": 1.871189979880921,
      "language_loss": 0.71608043,
      "learning_rate": 3.723945409226879e-06,
      "loss": 0.73841202,
      "num_input_tokens_seen": 34202055,
      "step": 1615,
      "time_per_iteration": 2.4885809421539307
    },
    {
      "auxiliary_loss_clip": 0.01138982,
      "auxiliary_loss_mlp": 0.01033488,
      "balance_loss_clip": 1.05060947,
      "balance_loss_mlp": 1.02286649,
      "epoch": 0.194312511272771,
      "flos": 9720337034880.0,
      "grad_norm": 13.05893792721544,
      "language_loss": 0.79536444,
      "learning_rate": 3.723550375147241e-06,
      "loss": 0.81708914,
      "num_input_tokens_seen": 34216830,
      "step": 1616,
      "time_per_iteration": 2.5593316555023193
    },
    {
      "auxiliary_loss_clip": 0.0121582,
      "auxiliary_loss_mlp": 0.00764959,
      "balance_loss_clip": 1.06095552,
      "balance_loss_mlp": 1.00046337,
      "epoch": 0.19443275416341008,
      "flos": 27016208150400.0,
      "grad_norm": 1.8390322368805143,
      "language_loss": 0.79599905,
      "learning_rate": 3.7231550796063816e-06,
      "loss": 0.81580687,
      "num_input_tokens_seen": 34236840,
      "step": 1617,
      "time_per_iteration": 2.4969944953918457
    },
    {
      "auxiliary_loss_clip": 0.01174301,
      "auxiliary_loss_mlp": 0.01041628,
      "balance_loss_clip": 1.06295407,
      "balance_loss_mlp": 1.03010607,
      "epoch": 0.1945529970540492,
      "flos": 15846625077120.0,
      "grad_norm": 1.7465424440587876,
      "language_loss": 0.65409112,
      "learning_rate": 3.722759522664266e-06,
      "loss": 0.6762504,
      "num_input_tokens_seen": 34254140,
      "step": 1618,
      "time_per_iteration": 3.2716782093048096
    },
    {
      "auxiliary_loss_clip": 0.01180541,
      "auxiliary_loss_mlp": 0.00764907,
      "balance_loss_clip": 1.05809689,
      "balance_loss_mlp": 1.00056672,
      "epoch": 0.19467323994468827,
      "flos": 19314985403520.0,
      "grad_norm": 2.0507877093816442,
      "language_loss": 0.82053483,
      "learning_rate": 3.7223637043809016e-06,
      "loss": 0.8399893,
      "num_input_tokens_seen": 34273120,
      "step": 1619,
      "time_per_iteration": 2.537863254547119
    },
    {
      "auxiliary_loss_clip": 0.01175305,
      "auxiliary_loss_mlp": 0.01041516,
      "balance_loss_clip": 1.06287801,
      "balance_loss_mlp": 1.03142476,
      "epoch": 0.19479348283532735,
      "flos": 24133227770880.0,
      "grad_norm": 1.8927501511858211,
      "language_loss": 0.86627793,
      "learning_rate": 3.7219676248163322e-06,
      "loss": 0.88844609,
      "num_input_tokens_seen": 34290285,
      "step": 1620,
      "time_per_iteration": 2.5778908729553223
    },
    {
      "auxiliary_loss_clip": 0.01109325,
      "auxiliary_loss_mlp": 0.01034254,
      "balance_loss_clip": 1.04539323,
      "balance_loss_mlp": 1.02354908,
      "epoch": 0.19491372572596646,
      "flos": 25775638174080.0,
      "grad_norm": 2.3810906249269626,
      "language_loss": 0.93286788,
      "learning_rate": 3.721571284030643e-06,
      "loss": 0.95430362,
      "num_input_tokens_seen": 34310095,
      "step": 1621,
      "time_per_iteration": 2.7197048664093018
    },
    {
      "auxiliary_loss_clip": 0.01189682,
      "auxiliary_loss_mlp": 0.01046514,
      "balance_loss_clip": 1.0573225,
      "balance_loss_mlp": 1.0360657,
      "epoch": 0.19503396861660555,
      "flos": 19645220067840.0,
      "grad_norm": 2.1868201189838405,
      "language_loss": 0.79439521,
      "learning_rate": 3.7211746820839587e-06,
      "loss": 0.8167572,
      "num_input_tokens_seen": 34327190,
      "step": 1622,
      "time_per_iteration": 4.219514608383179
    },
    {
      "auxiliary_loss_clip": 0.0118603,
      "auxiliary_loss_mlp": 0.01036533,
      "balance_loss_clip": 1.06009829,
      "balance_loss_mlp": 1.0250349,
      "epoch": 0.19515421150724463,
      "flos": 21033023892480.0,
      "grad_norm": 2.2274054365367277,
      "language_loss": 0.80655259,
      "learning_rate": 3.7207778190364437e-06,
      "loss": 0.82877821,
      "num_input_tokens_seen": 34345615,
      "step": 1623,
      "time_per_iteration": 3.2809736728668213
    },
    {
      "auxiliary_loss_clip": 0.01134173,
      "auxiliary_loss_mlp": 0.0103571,
      "balance_loss_clip": 1.05491948,
      "balance_loss_mlp": 1.02567267,
      "epoch": 0.1952744543978837,
      "flos": 32961255143040.0,
      "grad_norm": 1.718489482671285,
      "language_loss": 0.74003702,
      "learning_rate": 3.720380694948302e-06,
      "loss": 0.76173586,
      "num_input_tokens_seen": 34368500,
      "step": 1624,
      "time_per_iteration": 2.737372875213623
    },
    {
      "auxiliary_loss_clip": 0.01113099,
      "auxiliary_loss_mlp": 0.01003176,
      "balance_loss_clip": 1.04360461,
      "balance_loss_mlp": 1.0011971,
      "epoch": 0.19539469728852282,
      "flos": 64044312030720.0,
      "grad_norm": 1.036774728640262,
      "language_loss": 0.71238673,
      "learning_rate": 3.719983309879777e-06,
      "loss": 0.73354948,
      "num_input_tokens_seen": 34428280,
      "step": 1625,
      "time_per_iteration": 3.140871286392212
    },
    {
      "auxiliary_loss_clip": 0.0118702,
      "auxiliary_loss_mlp": 0.01037676,
      "balance_loss_clip": 1.05806172,
      "balance_loss_mlp": 1.02658391,
      "epoch": 0.1955149401791619,
      "flos": 13370908078080.0,
      "grad_norm": 2.519607162139007,
      "language_loss": 0.77688944,
      "learning_rate": 3.719585663891151e-06,
      "loss": 0.79913646,
      "num_input_tokens_seen": 34445815,
      "step": 1626,
      "time_per_iteration": 2.518704652786255
    },
    {
      "auxiliary_loss_clip": 0.01112272,
      "auxiliary_loss_mlp": 0.01042863,
      "balance_loss_clip": 1.04572916,
      "balance_loss_mlp": 1.03154397,
      "epoch": 0.195635183069801,
      "flos": 18728887184640.0,
      "grad_norm": 2.1175759779116965,
      "language_loss": 0.79034162,
      "learning_rate": 3.719187757042747e-06,
      "loss": 0.81189287,
      "num_input_tokens_seen": 34463635,
      "step": 1627,
      "time_per_iteration": 2.605221748352051
    },
    {
      "auxiliary_loss_clip": 0.0107635,
      "auxiliary_loss_mlp": 0.01001855,
      "balance_loss_clip": 1.03367257,
      "balance_loss_mlp": 0.99993557,
      "epoch": 0.1957554259604401,
      "flos": 69313952615040.0,
      "grad_norm": 0.7247970327044971,
      "language_loss": 0.54931569,
      "learning_rate": 3.7187895893949275e-06,
      "loss": 0.5700978,
      "num_input_tokens_seen": 34530105,
      "step": 1628,
      "time_per_iteration": 3.3033480644226074
    },
    {
      "auxiliary_loss_clip": 0.01169305,
      "auxiliary_loss_mlp": 0.01032611,
      "balance_loss_clip": 1.05410039,
      "balance_loss_mlp": 1.02212083,
      "epoch": 0.19587566885107918,
      "flos": 21069257736960.0,
      "grad_norm": 2.4415677507875175,
      "language_loss": 0.76474202,
      "learning_rate": 3.7183911610080937e-06,
      "loss": 0.78676111,
      "num_input_tokens_seen": 34546970,
      "step": 1629,
      "time_per_iteration": 2.584447145462036
    },
    {
      "auxiliary_loss_clip": 0.01169091,
      "auxiliary_loss_mlp": 0.01035765,
      "balance_loss_clip": 1.05924153,
      "balance_loss_mlp": 1.02449393,
      "epoch": 0.19599591174171827,
      "flos": 22194661731840.0,
      "grad_norm": 2.6659217629578245,
      "language_loss": 0.75020933,
      "learning_rate": 3.7179924719426872e-06,
      "loss": 0.77225792,
      "num_input_tokens_seen": 34564865,
      "step": 1630,
      "time_per_iteration": 2.5866341590881348
    },
    {
      "auxiliary_loss_clip": 0.01193333,
      "auxiliary_loss_mlp": 0.01035776,
      "balance_loss_clip": 1.05921888,
      "balance_loss_mlp": 1.02466595,
      "epoch": 0.19611615463235738,
      "flos": 23768375374080.0,
      "grad_norm": 2.339796253131173,
      "language_loss": 0.75539827,
      "learning_rate": 3.7175935222591885e-06,
      "loss": 0.77768934,
      "num_input_tokens_seen": 34584165,
      "step": 1631,
      "time_per_iteration": 2.5343430042266846
    },
    {
      "auxiliary_loss_clip": 0.01166604,
      "auxiliary_loss_mlp": 0.01038026,
      "balance_loss_clip": 1.06343091,
      "balance_loss_mlp": 1.02693987,
      "epoch": 0.19623639752299646,
      "flos": 28618218731520.0,
      "grad_norm": 1.8502289732702444,
      "language_loss": 0.74584919,
      "learning_rate": 3.717194312018118e-06,
      "loss": 0.76789558,
      "num_input_tokens_seen": 34603150,
      "step": 1632,
      "time_per_iteration": 2.6674070358276367
    },
    {
      "auxiliary_loss_clip": 0.01168635,
      "auxiliary_loss_mlp": 0.01028344,
      "balance_loss_clip": 1.05680192,
      "balance_loss_mlp": 1.01736474,
      "epoch": 0.19635664041363554,
      "flos": 21032700670080.0,
      "grad_norm": 1.9877100598203519,
      "language_loss": 0.76341563,
      "learning_rate": 3.716794841280036e-06,
      "loss": 0.78538537,
      "num_input_tokens_seen": 34621855,
      "step": 1633,
      "time_per_iteration": 2.5425777435302734
    },
    {
      "auxiliary_loss_clip": 0.01175477,
      "auxiliary_loss_mlp": 0.01035275,
      "balance_loss_clip": 1.05892622,
      "balance_loss_mlp": 1.02387273,
      "epoch": 0.19647688330427462,
      "flos": 18879748306560.0,
      "grad_norm": 2.286649282149183,
      "language_loss": 0.77889532,
      "learning_rate": 3.7163951101055407e-06,
      "loss": 0.80100286,
      "num_input_tokens_seen": 34639915,
      "step": 1634,
      "time_per_iteration": 2.5621702671051025
    },
    {
      "auxiliary_loss_clip": 0.01208726,
      "auxiliary_loss_mlp": 0.01034877,
      "balance_loss_clip": 1.06094241,
      "balance_loss_mlp": 1.02337289,
      "epoch": 0.19659712619491373,
      "flos": 24242503921920.0,
      "grad_norm": 1.9339792202003607,
      "language_loss": 0.79047823,
      "learning_rate": 3.715995118555273e-06,
      "loss": 0.81291437,
      "num_input_tokens_seen": 34659890,
      "step": 1635,
      "time_per_iteration": 2.516979455947876
    },
    {
      "auxiliary_loss_clip": 0.01202088,
      "auxiliary_loss_mlp": 0.01033883,
      "balance_loss_clip": 1.05871642,
      "balance_loss_mlp": 1.0218848,
      "epoch": 0.19671736908555282,
      "flos": 24717422568960.0,
      "grad_norm": 2.0387822064514425,
      "language_loss": 0.85893595,
      "learning_rate": 3.71559486668991e-06,
      "loss": 0.88129562,
      "num_input_tokens_seen": 34678750,
      "step": 1636,
      "time_per_iteration": 2.531385898590088
    },
    {
      "auxiliary_loss_clip": 0.01105392,
      "auxiliary_loss_mlp": 0.01039509,
      "balance_loss_clip": 1.05271244,
      "balance_loss_mlp": 1.0282377,
      "epoch": 0.1968376119761919,
      "flos": 23842279607040.0,
      "grad_norm": 1.8094846241296252,
      "language_loss": 0.77669436,
      "learning_rate": 3.715194354570169e-06,
      "loss": 0.79814339,
      "num_input_tokens_seen": 34698755,
      "step": 1637,
      "time_per_iteration": 2.6884257793426514
    },
    {
      "auxiliary_loss_clip": 0.0117439,
      "auxiliary_loss_mlp": 0.01040032,
      "balance_loss_clip": 1.05936122,
      "balance_loss_mlp": 1.02912402,
      "epoch": 0.196957854866831,
      "flos": 18113917409280.0,
      "grad_norm": 4.130204384186348,
      "language_loss": 0.83184004,
      "learning_rate": 3.714793582256809e-06,
      "loss": 0.85398424,
      "num_input_tokens_seen": 34715820,
      "step": 1638,
      "time_per_iteration": 2.5311548709869385
    },
    {
      "auxiliary_loss_clip": 0.01146633,
      "auxiliary_loss_mlp": 0.00764536,
      "balance_loss_clip": 1.05415344,
      "balance_loss_mlp": 1.00049317,
      "epoch": 0.1970780977574701,
      "flos": 21653129312640.0,
      "grad_norm": 2.1345992737318893,
      "language_loss": 0.85371292,
      "learning_rate": 3.7143925498106253e-06,
      "loss": 0.87282455,
      "num_input_tokens_seen": 34734360,
      "step": 1639,
      "time_per_iteration": 2.585815191268921
    },
    {
      "auxiliary_loss_clip": 0.01158868,
      "auxiliary_loss_mlp": 0.0103235,
      "balance_loss_clip": 1.05226803,
      "balance_loss_mlp": 1.02135289,
      "epoch": 0.19719834064810918,
      "flos": 20811813984000.0,
      "grad_norm": 1.918553312829744,
      "language_loss": 0.79270196,
      "learning_rate": 3.7139912572924558e-06,
      "loss": 0.81461412,
      "num_input_tokens_seen": 34753390,
      "step": 1640,
      "time_per_iteration": 2.5746662616729736
    },
    {
      "auxiliary_loss_clip": 0.01153836,
      "auxiliary_loss_mlp": 0.01035153,
      "balance_loss_clip": 1.0507201,
      "balance_loss_mlp": 1.02338088,
      "epoch": 0.19731858353874826,
      "flos": 23434800744960.0,
      "grad_norm": 3.189558212152088,
      "language_loss": 0.80599296,
      "learning_rate": 3.7135897047631744e-06,
      "loss": 0.82788289,
      "num_input_tokens_seen": 34771275,
      "step": 1641,
      "time_per_iteration": 2.6057252883911133
    },
    {
      "auxiliary_loss_clip": 0.01208407,
      "auxiliary_loss_mlp": 0.00765536,
      "balance_loss_clip": 1.06137562,
      "balance_loss_mlp": 1.0004034,
      "epoch": 0.19743882642938737,
      "flos": 23988184652160.0,
      "grad_norm": 1.9823028884187144,
      "language_loss": 0.76393664,
      "learning_rate": 3.713187892283698e-06,
      "loss": 0.78367603,
      "num_input_tokens_seen": 34790885,
      "step": 1642,
      "time_per_iteration": 2.5137948989868164
    },
    {
      "auxiliary_loss_clip": 0.0115901,
      "auxiliary_loss_mlp": 0.00765492,
      "balance_loss_clip": 1.05490351,
      "balance_loss_mlp": 1.00057387,
      "epoch": 0.19755906932002645,
      "flos": 15004340081280.0,
      "grad_norm": 2.34627364577497,
      "language_loss": 0.87196803,
      "learning_rate": 3.71278581991498e-06,
      "loss": 0.89121306,
      "num_input_tokens_seen": 34806745,
      "step": 1643,
      "time_per_iteration": 3.417463541030884
    },
    {
      "auxiliary_loss_clip": 0.01214499,
      "auxiliary_loss_mlp": 0.01034478,
      "balance_loss_clip": 1.06499171,
      "balance_loss_mlp": 1.02213442,
      "epoch": 0.19767931221066554,
      "flos": 19494466686720.0,
      "grad_norm": 1.9231858060808655,
      "language_loss": 0.79051411,
      "learning_rate": 3.712383487718015e-06,
      "loss": 0.81300384,
      "num_input_tokens_seen": 34824985,
      "step": 1644,
      "time_per_iteration": 2.4887959957122803
    },
    {
      "auxiliary_loss_clip": 0.01179514,
      "auxiliary_loss_mlp": 0.01045174,
      "balance_loss_clip": 1.05835533,
      "balance_loss_mlp": 1.0348506,
      "epoch": 0.19779955510130465,
      "flos": 25737895958400.0,
      "grad_norm": 1.8493796897741368,
      "language_loss": 0.8642903,
      "learning_rate": 3.7119808957538365e-06,
      "loss": 0.88653719,
      "num_input_tokens_seen": 34843980,
      "step": 1645,
      "time_per_iteration": 2.5633912086486816
    },
    {
      "auxiliary_loss_clip": 0.01147143,
      "auxiliary_loss_mlp": 0.01036751,
      "balance_loss_clip": 1.05525398,
      "balance_loss_mlp": 1.02490175,
      "epoch": 0.19791979799194373,
      "flos": 20777699041920.0,
      "grad_norm": 1.981110316114066,
      "language_loss": 0.79777801,
      "learning_rate": 3.711578044083517e-06,
      "loss": 0.81961691,
      "num_input_tokens_seen": 34860780,
      "step": 1646,
      "time_per_iteration": 2.618001699447632
    },
    {
      "auxiliary_loss_clip": 0.01184536,
      "auxiliary_loss_mlp": 0.01033531,
      "balance_loss_clip": 1.05681252,
      "balance_loss_mlp": 1.02257001,
      "epoch": 0.1980400408825828,
      "flos": 25589010084480.0,
      "grad_norm": 1.9258640752068876,
      "language_loss": 0.74485493,
      "learning_rate": 3.7111749327681698e-06,
      "loss": 0.7670356,
      "num_input_tokens_seen": 34880815,
      "step": 1647,
      "time_per_iteration": 2.5734853744506836
    },
    {
      "auxiliary_loss_clip": 0.01186949,
      "auxiliary_loss_mlp": 0.01035802,
      "balance_loss_clip": 1.06294823,
      "balance_loss_mlp": 1.02442908,
      "epoch": 0.1981602837732219,
      "flos": 23513840622720.0,
      "grad_norm": 2.2589531540597494,
      "language_loss": 0.86551875,
      "learning_rate": 3.7107715618689455e-06,
      "loss": 0.88774627,
      "num_input_tokens_seen": 34899790,
      "step": 1648,
      "time_per_iteration": 3.430835485458374
    },
    {
      "auxiliary_loss_clip": 0.01173553,
      "auxiliary_loss_mlp": 0.01034796,
      "balance_loss_clip": 1.057199,
      "balance_loss_mlp": 1.02373362,
      "epoch": 0.198280526663861,
      "flos": 23185365724800.0,
      "grad_norm": 2.110081829592474,
      "language_loss": 0.83506656,
      "learning_rate": 3.710367931447035e-06,
      "loss": 0.85715008,
      "num_input_tokens_seen": 34921570,
      "step": 1649,
      "time_per_iteration": 3.335034132003784
    },
    {
      "auxiliary_loss_clip": 0.01168667,
      "auxiliary_loss_mlp": 0.00764835,
      "balance_loss_clip": 1.0586158,
      "balance_loss_mlp": 1.00053918,
      "epoch": 0.1984007695545001,
      "flos": 21689470897920.0,
      "grad_norm": 2.4661138836580876,
      "language_loss": 0.86696333,
      "learning_rate": 3.70996404156367e-06,
      "loss": 0.88629836,
      "num_input_tokens_seen": 34941205,
      "step": 1650,
      "time_per_iteration": 2.6076276302337646
    },
    {
      "auxiliary_loss_clip": 0.01161741,
      "auxiliary_loss_mlp": 0.01040262,
      "balance_loss_clip": 1.05694056,
      "balance_loss_mlp": 1.02880013,
      "epoch": 0.19852101244513917,
      "flos": 36064008887040.0,
      "grad_norm": 1.679542704668755,
      "language_loss": 0.72701836,
      "learning_rate": 3.7095598922801187e-06,
      "loss": 0.74903834,
      "num_input_tokens_seen": 34963280,
      "step": 1651,
      "time_per_iteration": 2.7348029613494873
    },
    {
      "auxiliary_loss_clip": 0.01190504,
      "auxiliary_loss_mlp": 0.01030178,
      "balance_loss_clip": 1.06032825,
      "balance_loss_mlp": 1.0195508,
      "epoch": 0.19864125533577828,
      "flos": 23105894883840.0,
      "grad_norm": 2.5148764076274954,
      "language_loss": 0.76450497,
      "learning_rate": 3.7091554836576914e-06,
      "loss": 0.78671181,
      "num_input_tokens_seen": 34979955,
      "step": 1652,
      "time_per_iteration": 2.538243055343628
    },
    {
      "auxiliary_loss_clip": 0.01058863,
      "auxiliary_loss_mlp": 0.01038257,
      "balance_loss_clip": 1.03927898,
      "balance_loss_mlp": 1.02683663,
      "epoch": 0.19876149822641737,
      "flos": 24608505553920.0,
      "grad_norm": 1.9755455868172398,
      "language_loss": 0.82779068,
      "learning_rate": 3.708750815757736e-06,
      "loss": 0.84876186,
      "num_input_tokens_seen": 35000725,
      "step": 1653,
      "time_per_iteration": 2.9575114250183105
    },
    {
      "auxiliary_loss_clip": 0.01166606,
      "auxiliary_loss_mlp": 0.00765623,
      "balance_loss_clip": 1.05573142,
      "balance_loss_mlp": 1.00045967,
      "epoch": 0.19888174111705645,
      "flos": 32196645308160.0,
      "grad_norm": 2.682513048982441,
      "language_loss": 0.73083997,
      "learning_rate": 3.7083458886416407e-06,
      "loss": 0.75016224,
      "num_input_tokens_seen": 35019920,
      "step": 1654,
      "time_per_iteration": 3.6619997024536133
    },
    {
      "auxiliary_loss_clip": 0.01116021,
      "auxiliary_loss_mlp": 0.01036985,
      "balance_loss_clip": 1.04752648,
      "balance_loss_mlp": 1.02531457,
      "epoch": 0.19900198400769553,
      "flos": 24608469640320.0,
      "grad_norm": 2.3364816167876303,
      "language_loss": 0.88664424,
      "learning_rate": 3.707940702370832e-06,
      "loss": 0.9081744,
      "num_input_tokens_seen": 35040765,
      "step": 1655,
      "time_per_iteration": 2.6625452041625977
    },
    {
      "auxiliary_loss_clip": 0.01112675,
      "auxiliary_loss_mlp": 0.01004358,
      "balance_loss_clip": 1.04707384,
      "balance_loss_mlp": 1.00241458,
      "epoch": 0.19912222689833464,
      "flos": 67915805673600.0,
      "grad_norm": 0.759900502364181,
      "language_loss": 0.58284771,
      "learning_rate": 3.707535257006777e-06,
      "loss": 0.60401797,
      "num_input_tokens_seen": 35106390,
      "step": 1656,
      "time_per_iteration": 3.209782123565674
    },
    {
      "auxiliary_loss_clip": 0.01162815,
      "auxiliary_loss_mlp": 0.01033814,
      "balance_loss_clip": 1.05399752,
      "balance_loss_mlp": 1.0214107,
      "epoch": 0.19924246978897373,
      "flos": 15742340916480.0,
      "grad_norm": 2.191348082631784,
      "language_loss": 0.88790941,
      "learning_rate": 3.707129552610981e-06,
      "loss": 0.90987575,
      "num_input_tokens_seen": 35125040,
      "step": 1657,
      "time_per_iteration": 2.5599117279052734
    },
    {
      "auxiliary_loss_clip": 0.01177751,
      "auxiliary_loss_mlp": 0.01045505,
      "balance_loss_clip": 1.06504679,
      "balance_loss_mlp": 1.03301191,
      "epoch": 0.1993627126796128,
      "flos": 17566566986880.0,
      "grad_norm": 2.0591950431074673,
      "language_loss": 0.74010438,
      "learning_rate": 3.70672358924499e-06,
      "loss": 0.76233697,
      "num_input_tokens_seen": 35144280,
      "step": 1658,
      "time_per_iteration": 2.5370209217071533
    },
    {
      "auxiliary_loss_clip": 0.0110172,
      "auxiliary_loss_mlp": 0.00764902,
      "balance_loss_clip": 1.04835844,
      "balance_loss_mlp": 1.00038707,
      "epoch": 0.19948295557025192,
      "flos": 40843826680320.0,
      "grad_norm": 1.892629596689948,
      "language_loss": 0.78663665,
      "learning_rate": 3.706317366970386e-06,
      "loss": 0.80530292,
      "num_input_tokens_seen": 35165280,
      "step": 1659,
      "time_per_iteration": 2.868565559387207
    },
    {
      "auxiliary_loss_clip": 0.01191298,
      "auxiliary_loss_mlp": 0.0103932,
      "balance_loss_clip": 1.06008554,
      "balance_loss_mlp": 1.02794743,
      "epoch": 0.199603198460891,
      "flos": 25082418620160.0,
      "grad_norm": 2.3848805891827958,
      "language_loss": 0.83359456,
      "learning_rate": 3.705910885848795e-06,
      "loss": 0.85590076,
      "num_input_tokens_seen": 35183655,
      "step": 1660,
      "time_per_iteration": 2.5683863162994385
    },
    {
      "auxiliary_loss_clip": 0.01193451,
      "auxiliary_loss_mlp": 0.01036475,
      "balance_loss_clip": 1.05965137,
      "balance_loss_mlp": 1.02473855,
      "epoch": 0.19972344135153008,
      "flos": 20084120352000.0,
      "grad_norm": 1.7596112306995706,
      "language_loss": 0.84891313,
      "learning_rate": 3.705504145941879e-06,
      "loss": 0.87121236,
      "num_input_tokens_seen": 35201825,
      "step": 1661,
      "time_per_iteration": 2.4981751441955566
    },
    {
      "auxiliary_loss_clip": 0.0119539,
      "auxiliary_loss_mlp": 0.01035636,
      "balance_loss_clip": 1.05969965,
      "balance_loss_mlp": 1.02464461,
      "epoch": 0.1998436842421692,
      "flos": 23727472761600.0,
      "grad_norm": 2.3102493316002746,
      "language_loss": 0.7862795,
      "learning_rate": 3.7050971473113403e-06,
      "loss": 0.80858976,
      "num_input_tokens_seen": 35221600,
      "step": 1662,
      "time_per_iteration": 2.542916774749756
    },
    {
      "auxiliary_loss_clip": 0.01174462,
      "auxiliary_loss_mlp": 0.01033759,
      "balance_loss_clip": 1.05683553,
      "balance_loss_mlp": 1.02344799,
      "epoch": 0.19996392713280828,
      "flos": 36102361633920.0,
      "grad_norm": 1.7209579466858733,
      "language_loss": 0.80392349,
      "learning_rate": 3.7046898900189196e-06,
      "loss": 0.8260057,
      "num_input_tokens_seen": 35245935,
      "step": 1663,
      "time_per_iteration": 2.7209854125976562
    },
    {
      "auxiliary_loss_clip": 0.0121114,
      "auxiliary_loss_mlp": 0.01030595,
      "balance_loss_clip": 1.06786799,
      "balance_loss_mlp": 1.01912105,
      "epoch": 0.20008417002344736,
      "flos": 23657662679040.0,
      "grad_norm": 1.7540784523966824,
      "language_loss": 0.83297962,
      "learning_rate": 3.704282374126398e-06,
      "loss": 0.85539699,
      "num_input_tokens_seen": 35265615,
      "step": 1664,
      "time_per_iteration": 2.516698122024536
    },
    {
      "auxiliary_loss_clip": 0.01198832,
      "auxiliary_loss_mlp": 0.01038181,
      "balance_loss_clip": 1.06135464,
      "balance_loss_mlp": 1.02664733,
      "epoch": 0.20020441291408644,
      "flos": 21872076664320.0,
      "grad_norm": 1.699871093842925,
      "language_loss": 0.87801516,
      "learning_rate": 3.7038745996955954e-06,
      "loss": 0.90038526,
      "num_input_tokens_seen": 35284960,
      "step": 1665,
      "time_per_iteration": 2.956199884414673
    },
    {
      "auxiliary_loss_clip": 0.01062879,
      "auxiliary_loss_mlp": 0.01047242,
      "balance_loss_clip": 1.0394218,
      "balance_loss_mlp": 1.03428996,
      "epoch": 0.20032465580472555,
      "flos": 23179691376000.0,
      "grad_norm": 2.4684363525184745,
      "language_loss": 0.71803027,
      "learning_rate": 3.703466566788371e-06,
      "loss": 0.73913145,
      "num_input_tokens_seen": 35304090,
      "step": 1666,
      "time_per_iteration": 3.314638137817383
    },
    {
      "auxiliary_loss_clip": 0.0116183,
      "auxiliary_loss_mlp": 0.0103157,
      "balance_loss_clip": 1.0573535,
      "balance_loss_mlp": 1.01970243,
      "epoch": 0.20044489869536464,
      "flos": 23873521461120.0,
      "grad_norm": 1.8702545672864737,
      "language_loss": 0.74293292,
      "learning_rate": 3.703058275466622e-06,
      "loss": 0.76486689,
      "num_input_tokens_seen": 35323325,
      "step": 1667,
      "time_per_iteration": 3.9419190883636475
    },
    {
      "auxiliary_loss_clip": 0.01171561,
      "auxiliary_loss_mlp": 0.01031371,
      "balance_loss_clip": 1.05594409,
      "balance_loss_mlp": 1.02054048,
      "epoch": 0.20056514158600372,
      "flos": 21945226711680.0,
      "grad_norm": 2.026513115879843,
      "language_loss": 0.77820128,
      "learning_rate": 3.7026497257922877e-06,
      "loss": 0.80023056,
      "num_input_tokens_seen": 35343635,
      "step": 1668,
      "time_per_iteration": 3.749480962753296
    },
    {
      "auxiliary_loss_clip": 0.01144002,
      "auxiliary_loss_mlp": 0.01040347,
      "balance_loss_clip": 1.05183578,
      "balance_loss_mlp": 1.02904034,
      "epoch": 0.20068538447664283,
      "flos": 23879159896320.0,
      "grad_norm": 1.696893154532381,
      "language_loss": 0.85243297,
      "learning_rate": 3.7022409178273436e-06,
      "loss": 0.87427652,
      "num_input_tokens_seen": 35364615,
      "step": 1669,
      "time_per_iteration": 2.888080358505249
    },
    {
      "auxiliary_loss_clip": 0.01170804,
      "auxiliary_loss_mlp": 0.01037693,
      "balance_loss_clip": 1.05116773,
      "balance_loss_mlp": 1.02611792,
      "epoch": 0.2008056273672819,
      "flos": 18442823270400.0,
      "grad_norm": 1.7530995181130486,
      "language_loss": 0.78526676,
      "learning_rate": 3.7018318516338054e-06,
      "loss": 0.80735171,
      "num_input_tokens_seen": 35383775,
      "step": 1670,
      "time_per_iteration": 2.856137752532959
    },
    {
      "auxiliary_loss_clip": 0.01186154,
      "auxiliary_loss_mlp": 0.01040452,
      "balance_loss_clip": 1.05854535,
      "balance_loss_mlp": 1.02921629,
      "epoch": 0.200925870257921,
      "flos": 23659530186240.0,
      "grad_norm": 2.756465883870941,
      "language_loss": 0.82061726,
      "learning_rate": 3.7014225272737284e-06,
      "loss": 0.84288329,
      "num_input_tokens_seen": 35403000,
      "step": 1671,
      "time_per_iteration": 2.822138547897339
    },
    {
      "auxiliary_loss_clip": 0.01208019,
      "auxiliary_loss_mlp": 0.01037721,
      "balance_loss_clip": 1.06374192,
      "balance_loss_mlp": 1.02667046,
      "epoch": 0.20104611314856008,
      "flos": 16217115909120.0,
      "grad_norm": 4.367207019965177,
      "language_loss": 0.74234641,
      "learning_rate": 3.701012944809207e-06,
      "loss": 0.76480383,
      "num_input_tokens_seen": 35420115,
      "step": 1672,
      "time_per_iteration": 4.164104223251343
    },
    {
      "auxiliary_loss_clip": 0.01171404,
      "auxiliary_loss_mlp": 0.01035864,
      "balance_loss_clip": 1.05648518,
      "balance_loss_mlp": 1.02522433,
      "epoch": 0.2011663560391992,
      "flos": 21397373498880.0,
      "grad_norm": 2.039382699041982,
      "language_loss": 0.78920144,
      "learning_rate": 3.700603104302374e-06,
      "loss": 0.81127411,
      "num_input_tokens_seen": 35439925,
      "step": 1673,
      "time_per_iteration": 3.6824557781219482
    },
    {
      "auxiliary_loss_clip": 0.01075235,
      "auxiliary_loss_mlp": 0.01005704,
      "balance_loss_clip": 1.03244543,
      "balance_loss_mlp": 1.00368929,
      "epoch": 0.20128659892983827,
      "flos": 62229459409920.0,
      "grad_norm": 0.8980052714847889,
      "language_loss": 0.55964845,
      "learning_rate": 3.7001930058154027e-06,
      "loss": 0.58045781,
      "num_input_tokens_seen": 35504885,
      "step": 1674,
      "time_per_iteration": 4.522831916809082
    },
    {
      "auxiliary_loss_clip": 0.01134811,
      "auxiliary_loss_mlp": 0.01046928,
      "balance_loss_clip": 1.0563823,
      "balance_loss_mlp": 1.03483438,
      "epoch": 0.20140684182047736,
      "flos": 28438737448320.0,
      "grad_norm": 4.14176230381484,
      "language_loss": 0.79203701,
      "learning_rate": 3.6997826494105037e-06,
      "loss": 0.81385446,
      "num_input_tokens_seen": 35525330,
      "step": 1675,
      "time_per_iteration": 3.044438123703003
    },
    {
      "auxiliary_loss_clip": 0.01175592,
      "auxiliary_loss_mlp": 0.00765042,
      "balance_loss_clip": 1.05687618,
      "balance_loss_mlp": 1.00038338,
      "epoch": 0.20152708471111647,
      "flos": 28074064619520.0,
      "grad_norm": 2.101403697514675,
      "language_loss": 0.69591868,
      "learning_rate": 3.6993720351499286e-06,
      "loss": 0.715325,
      "num_input_tokens_seen": 35546455,
      "step": 1676,
      "time_per_iteration": 3.620248317718506
    },
    {
      "auxiliary_loss_clip": 0.01181812,
      "auxiliary_loss_mlp": 0.01035742,
      "balance_loss_clip": 1.056481,
      "balance_loss_mlp": 1.02506065,
      "epoch": 0.20164732760175555,
      "flos": 23549751244800.0,
      "grad_norm": 1.8338381709656126,
      "language_loss": 0.77133852,
      "learning_rate": 3.6989611630959666e-06,
      "loss": 0.79351401,
      "num_input_tokens_seen": 35565010,
      "step": 1677,
      "time_per_iteration": 2.8750741481781006
    },
    {
      "auxiliary_loss_clip": 0.01110966,
      "auxiliary_loss_mlp": 0.0100393,
      "balance_loss_clip": 1.03059912,
      "balance_loss_mlp": 1.00176048,
      "epoch": 0.20176757049239463,
      "flos": 71100616037760.0,
      "grad_norm": 0.6778426833465658,
      "language_loss": 0.58306241,
      "learning_rate": 3.6985500333109474e-06,
      "loss": 0.60421145,
      "num_input_tokens_seen": 35633340,
      "step": 1678,
      "time_per_iteration": 3.367764949798584
    },
    {
      "auxiliary_loss_clip": 0.01186484,
      "auxiliary_loss_mlp": 0.01033384,
      "balance_loss_clip": 1.06320763,
      "balance_loss_mlp": 1.02278018,
      "epoch": 0.20188781338303372,
      "flos": 21430159637760.0,
      "grad_norm": 2.268618054050469,
      "language_loss": 0.76687682,
      "learning_rate": 3.6981386458572385e-06,
      "loss": 0.78907549,
      "num_input_tokens_seen": 35651315,
      "step": 1679,
      "time_per_iteration": 2.7507872581481934
    },
    {
      "auxiliary_loss_clip": 0.01197778,
      "auxiliary_loss_mlp": 0.01035101,
      "balance_loss_clip": 1.06056952,
      "balance_loss_mlp": 1.02291143,
      "epoch": 0.20200805627367283,
      "flos": 11546215130880.0,
      "grad_norm": 2.217957890990232,
      "language_loss": 0.75855118,
      "learning_rate": 3.6977270007972468e-06,
      "loss": 0.78087997,
      "num_input_tokens_seen": 35668850,
      "step": 1680,
      "time_per_iteration": 2.7660398483276367
    },
    {
      "auxiliary_loss_clip": 0.01173122,
      "auxiliary_loss_mlp": 0.01035754,
      "balance_loss_clip": 1.05849266,
      "balance_loss_mlp": 1.02494216,
      "epoch": 0.2021282991643119,
      "flos": 28545391906560.0,
      "grad_norm": 2.5449590706321024,
      "language_loss": 0.72325599,
      "learning_rate": 3.6973150981934196e-06,
      "loss": 0.74534476,
      "num_input_tokens_seen": 35690080,
      "step": 1681,
      "time_per_iteration": 2.7877614498138428
    },
    {
      "auxiliary_loss_clip": 0.01131454,
      "auxiliary_loss_mlp": 0.01037387,
      "balance_loss_clip": 1.05014491,
      "balance_loss_mlp": 1.02511978,
      "epoch": 0.202248542054951,
      "flos": 17923446564480.0,
      "grad_norm": 2.439761779920082,
      "language_loss": 0.83478177,
      "learning_rate": 3.6969029381082415e-06,
      "loss": 0.85647017,
      "num_input_tokens_seen": 35706075,
      "step": 1682,
      "time_per_iteration": 2.7942821979522705
    },
    {
      "auxiliary_loss_clip": 0.01180792,
      "auxiliary_loss_mlp": 0.01027462,
      "balance_loss_clip": 1.05633521,
      "balance_loss_mlp": 1.01755023,
      "epoch": 0.2023687849455901,
      "flos": 19864634296320.0,
      "grad_norm": 6.393717696722472,
      "language_loss": 0.79621208,
      "learning_rate": 3.696490520604237e-06,
      "loss": 0.81829464,
      "num_input_tokens_seen": 35724765,
      "step": 1683,
      "time_per_iteration": 2.819718360900879
    },
    {
      "auxiliary_loss_clip": 0.0112212,
      "auxiliary_loss_mlp": 0.01030466,
      "balance_loss_clip": 1.04976785,
      "balance_loss_mlp": 1.020661,
      "epoch": 0.20248902783622919,
      "flos": 22564721600640.0,
      "grad_norm": 1.7942991629281715,
      "language_loss": 0.80767721,
      "learning_rate": 3.696077845743968e-06,
      "loss": 0.82920313,
      "num_input_tokens_seen": 35744355,
      "step": 1684,
      "time_per_iteration": 2.8505680561065674
    },
    {
      "auxiliary_loss_clip": 0.01150348,
      "auxiliary_loss_mlp": 0.01035836,
      "balance_loss_clip": 1.05087841,
      "balance_loss_mlp": 1.02299738,
      "epoch": 0.20260927072686827,
      "flos": 22709728805760.0,
      "grad_norm": 3.4550597470173026,
      "language_loss": 0.73140585,
      "learning_rate": 3.69566491359004e-06,
      "loss": 0.75326771,
      "num_input_tokens_seen": 35761000,
      "step": 1685,
      "time_per_iteration": 2.8570172786712646
    },
    {
      "auxiliary_loss_clip": 0.01143281,
      "auxiliary_loss_mlp": 0.01033649,
      "balance_loss_clip": 1.05285895,
      "balance_loss_mlp": 1.02262235,
      "epoch": 0.20272951361750738,
      "flos": 51023998650240.0,
      "grad_norm": 1.7023410209126686,
      "language_loss": 0.69445997,
      "learning_rate": 3.695251724205092e-06,
      "loss": 0.71622932,
      "num_input_tokens_seen": 35785360,
      "step": 1686,
      "time_per_iteration": 3.095844268798828
    },
    {
      "auxiliary_loss_clip": 0.01182912,
      "auxiliary_loss_mlp": 0.01034546,
      "balance_loss_clip": 1.05996394,
      "balance_loss_mlp": 1.02307796,
      "epoch": 0.20284975650814646,
      "flos": 26578133879040.0,
      "grad_norm": 1.92192665763565,
      "language_loss": 0.86416817,
      "learning_rate": 3.6948382776518054e-06,
      "loss": 0.88634276,
      "num_input_tokens_seen": 35806065,
      "step": 1687,
      "time_per_iteration": 2.747946262359619
    },
    {
      "auxiliary_loss_clip": 0.01101655,
      "auxiliary_loss_mlp": 0.01036338,
      "balance_loss_clip": 1.049034,
      "balance_loss_mlp": 1.02522182,
      "epoch": 0.20296999939878554,
      "flos": 16034222833920.0,
      "grad_norm": 2.3010617157827573,
      "language_loss": 0.79364145,
      "learning_rate": 3.6944245739929e-06,
      "loss": 0.8150214,
      "num_input_tokens_seen": 35822225,
      "step": 1688,
      "time_per_iteration": 3.1403985023498535
    },
    {
      "auxiliary_loss_clip": 0.01156883,
      "auxiliary_loss_mlp": 0.01038029,
      "balance_loss_clip": 1.05571473,
      "balance_loss_mlp": 1.025208,
      "epoch": 0.20309024228942463,
      "flos": 19203374868480.0,
      "grad_norm": 2.6281812269738807,
      "language_loss": 0.72040719,
      "learning_rate": 3.6940106132911332e-06,
      "loss": 0.7423563,
      "num_input_tokens_seen": 35839410,
      "step": 1689,
      "time_per_iteration": 3.1037068367004395
    },
    {
      "auxiliary_loss_clip": 0.01139004,
      "auxiliary_loss_mlp": 0.01039308,
      "balance_loss_clip": 1.04916108,
      "balance_loss_mlp": 1.02726829,
      "epoch": 0.20321048518006374,
      "flos": 22821087945600.0,
      "grad_norm": 1.9821659061752697,
      "language_loss": 0.88637066,
      "learning_rate": 3.6935963956093037e-06,
      "loss": 0.90815377,
      "num_input_tokens_seen": 35859495,
      "step": 1690,
      "time_per_iteration": 3.188859701156616
    },
    {
      "auxiliary_loss_clip": 0.01162416,
      "auxiliary_loss_mlp": 0.01041712,
      "balance_loss_clip": 1.05742741,
      "balance_loss_mlp": 1.03138828,
      "epoch": 0.20333072807070282,
      "flos": 19096397187840.0,
      "grad_norm": 1.793515518310224,
      "language_loss": 0.68917572,
      "learning_rate": 3.6931819210102474e-06,
      "loss": 0.71121705,
      "num_input_tokens_seen": 35878890,
      "step": 1691,
      "time_per_iteration": 2.9627983570098877
    },
    {
      "auxiliary_loss_clip": 0.01208991,
      "auxiliary_loss_mlp": 0.01039108,
      "balance_loss_clip": 1.06058598,
      "balance_loss_mlp": 1.02862382,
      "epoch": 0.2034509709613419,
      "flos": 18180962144640.0,
      "grad_norm": 1.8608333206651322,
      "language_loss": 0.84297043,
      "learning_rate": 3.6927671895568402e-06,
      "loss": 0.86545146,
      "num_input_tokens_seen": 35897950,
      "step": 1692,
      "time_per_iteration": 2.796652317047119
    },
    {
      "auxiliary_loss_clip": 0.01180809,
      "auxiliary_loss_mlp": 0.01039465,
      "balance_loss_clip": 1.05884337,
      "balance_loss_mlp": 1.02765727,
      "epoch": 0.20357121385198101,
      "flos": 22923899648640.0,
      "grad_norm": 3.1106436455379955,
      "language_loss": 0.86670369,
      "learning_rate": 3.692352201311996e-06,
      "loss": 0.88890642,
      "num_input_tokens_seen": 35916800,
      "step": 1693,
      "time_per_iteration": 2.854883909225464
    },
    {
      "auxiliary_loss_clip": 0.01133739,
      "auxiliary_loss_mlp": 0.01034333,
      "balance_loss_clip": 1.05253577,
      "balance_loss_mlp": 1.0218693,
      "epoch": 0.2036914567426201,
      "flos": 20922131629440.0,
      "grad_norm": 3.5029853915877767,
      "language_loss": 0.76682371,
      "learning_rate": 3.6919369563386687e-06,
      "loss": 0.78850448,
      "num_input_tokens_seen": 35936600,
      "step": 1694,
      "time_per_iteration": 3.7682456970214844
    },
    {
      "auxiliary_loss_clip": 0.01169493,
      "auxiliary_loss_mlp": 0.01034192,
      "balance_loss_clip": 1.05303431,
      "balance_loss_mlp": 1.02315891,
      "epoch": 0.20381169963325918,
      "flos": 15519155760000.0,
      "grad_norm": 2.0052718202918838,
      "language_loss": 0.79065287,
      "learning_rate": 3.69152145469985e-06,
      "loss": 0.81268972,
      "num_input_tokens_seen": 35953645,
      "step": 1695,
      "time_per_iteration": 2.748426675796509
    },
    {
      "auxiliary_loss_clip": 0.01169625,
      "auxiliary_loss_mlp": 0.01036708,
      "balance_loss_clip": 1.0543673,
      "balance_loss_mlp": 1.0243814,
      "epoch": 0.20393194252389826,
      "flos": 28833143760000.0,
      "grad_norm": 1.8905007907906302,
      "language_loss": 0.82474178,
      "learning_rate": 3.691105696458572e-06,
      "loss": 0.8468051,
      "num_input_tokens_seen": 35970940,
      "step": 1696,
      "time_per_iteration": 2.9206342697143555
    },
    {
      "auxiliary_loss_clip": 0.01158632,
      "auxiliary_loss_mlp": 0.01038633,
      "balance_loss_clip": 1.05286551,
      "balance_loss_mlp": 1.02738011,
      "epoch": 0.20405218541453737,
      "flos": 22488554810880.0,
      "grad_norm": 4.153108916099943,
      "language_loss": 0.67273152,
      "learning_rate": 3.690689681677904e-06,
      "loss": 0.69470417,
      "num_input_tokens_seen": 35989410,
      "step": 1697,
      "time_per_iteration": 2.8462464809417725
    },
    {
      "auxiliary_loss_clip": 0.01181262,
      "auxiliary_loss_mlp": 0.01047287,
      "balance_loss_clip": 1.05687618,
      "balance_loss_mlp": 1.03583705,
      "epoch": 0.20417242830517646,
      "flos": 25374408278400.0,
      "grad_norm": 1.8517767094029411,
      "language_loss": 0.88936043,
      "learning_rate": 3.690273410420956e-06,
      "loss": 0.91164595,
      "num_input_tokens_seen": 36009175,
      "step": 1698,
      "time_per_iteration": 4.836026430130005
    },
    {
      "auxiliary_loss_clip": 0.01172707,
      "auxiliary_loss_mlp": 0.01038721,
      "balance_loss_clip": 1.05576289,
      "balance_loss_mlp": 1.02745557,
      "epoch": 0.20429267119581554,
      "flos": 14793078240000.0,
      "grad_norm": 2.5544659333549804,
      "language_loss": 0.76540017,
      "learning_rate": 3.689856882750875e-06,
      "loss": 0.78751445,
      "num_input_tokens_seen": 36024375,
      "step": 1699,
      "time_per_iteration": 3.91256046295166
    },
    {
      "auxiliary_loss_clip": 0.01143562,
      "auxiliary_loss_mlp": 0.01036626,
      "balance_loss_clip": 1.05852151,
      "balance_loss_mlp": 1.0260942,
      "epoch": 0.20441291408645465,
      "flos": 17781851151360.0,
      "grad_norm": 1.684520091222651,
      "language_loss": 0.78824461,
      "learning_rate": 3.6894400987308486e-06,
      "loss": 0.81004643,
      "num_input_tokens_seen": 36041895,
      "step": 1700,
      "time_per_iteration": 2.805859327316284
    },
    {
      "auxiliary_loss_clip": 0.01152744,
      "auxiliary_loss_mlp": 0.01039782,
      "balance_loss_clip": 1.05519152,
      "balance_loss_mlp": 1.02651978,
      "epoch": 0.20453315697709373,
      "flos": 16435668211200.0,
      "grad_norm": 3.7955563449115255,
      "language_loss": 0.85287338,
      "learning_rate": 3.6890230584241024e-06,
      "loss": 0.8747986,
      "num_input_tokens_seen": 36058825,
      "step": 1701,
      "time_per_iteration": 2.859717607498169
    },
    {
      "auxiliary_loss_clip": 0.0110699,
      "auxiliary_loss_mlp": 0.00759164,
      "balance_loss_clip": 1.03138912,
      "balance_loss_mlp": 1.00016117,
      "epoch": 0.20465339986773282,
      "flos": 66713085653760.0,
      "grad_norm": 1.054795746606504,
      "language_loss": 0.66453969,
      "learning_rate": 3.6886057618939016e-06,
      "loss": 0.68320125,
      "num_input_tokens_seen": 36121645,
      "step": 1702,
      "time_per_iteration": 3.4474027156829834
    },
    {
      "auxiliary_loss_clip": 0.01129717,
      "auxiliary_loss_mlp": 0.0103996,
      "balance_loss_clip": 1.05027044,
      "balance_loss_mlp": 1.02916598,
      "epoch": 0.2047736427583719,
      "flos": 41974114924800.0,
      "grad_norm": 1.879442653379252,
      "language_loss": 0.6942637,
      "learning_rate": 3.6881882092035492e-06,
      "loss": 0.71596044,
      "num_input_tokens_seen": 36143030,
      "step": 1703,
      "time_per_iteration": 3.0114307403564453
    },
    {
      "auxiliary_loss_clip": 0.0105288,
      "auxiliary_loss_mlp": 0.0075927,
      "balance_loss_clip": 1.01963806,
      "balance_loss_mlp": 1.00016546,
      "epoch": 0.204893885649011,
      "flos": 69940878641280.0,
      "grad_norm": 0.9234772996381411,
      "language_loss": 0.61212409,
      "learning_rate": 3.6877704004163873e-06,
      "loss": 0.63024557,
      "num_input_tokens_seen": 36203435,
      "step": 1704,
      "time_per_iteration": 3.686976909637451
    },
    {
      "auxiliary_loss_clip": 0.01175637,
      "auxiliary_loss_mlp": 0.01041565,
      "balance_loss_clip": 1.05986261,
      "balance_loss_mlp": 1.03024626,
      "epoch": 0.2050141285396501,
      "flos": 22200012858240.0,
      "grad_norm": 2.2808261671377386,
      "language_loss": 0.77966017,
      "learning_rate": 3.6873523355957984e-06,
      "loss": 0.8018322,
      "num_input_tokens_seen": 36222435,
      "step": 1705,
      "time_per_iteration": 3.5739963054656982
    },
    {
      "auxiliary_loss_clip": 0.01090364,
      "auxiliary_loss_mlp": 0.01009964,
      "balance_loss_clip": 1.02839684,
      "balance_loss_mlp": 1.00765169,
      "epoch": 0.20513437143028918,
      "flos": 46283721730560.0,
      "grad_norm": 0.9799077473696763,
      "language_loss": 0.64044493,
      "learning_rate": 3.686934014805201e-06,
      "loss": 0.66144824,
      "num_input_tokens_seen": 36273065,
      "step": 1706,
      "time_per_iteration": 3.1826171875
    },
    {
      "auxiliary_loss_clip": 0.0121922,
      "auxiliary_loss_mlp": 0.00765604,
      "balance_loss_clip": 1.06386149,
      "balance_loss_mlp": 1.00049329,
      "epoch": 0.20525461432092829,
      "flos": 21904324099200.0,
      "grad_norm": 2.2235253538425948,
      "language_loss": 0.80961418,
      "learning_rate": 3.6865154381080552e-06,
      "loss": 0.82946241,
      "num_input_tokens_seen": 36293750,
      "step": 1707,
      "time_per_iteration": 2.6622250080108643
    },
    {
      "auxiliary_loss_clip": 0.01190011,
      "auxiliary_loss_mlp": 0.0102919,
      "balance_loss_clip": 1.05919111,
      "balance_loss_mlp": 1.01843166,
      "epoch": 0.20537485721156737,
      "flos": 21214264942080.0,
      "grad_norm": 2.061799034598792,
      "language_loss": 0.82627714,
      "learning_rate": 3.6860966055678585e-06,
      "loss": 0.84846914,
      "num_input_tokens_seen": 36310105,
      "step": 1708,
      "time_per_iteration": 2.713313341140747
    },
    {
      "auxiliary_loss_clip": 0.01194309,
      "auxiliary_loss_mlp": 0.01038869,
      "balance_loss_clip": 1.06177497,
      "balance_loss_mlp": 1.02614331,
      "epoch": 0.20549510010220645,
      "flos": 20191205773440.0,
      "grad_norm": 2.0867163583340735,
      "language_loss": 0.86276668,
      "learning_rate": 3.685677517248147e-06,
      "loss": 0.88509846,
      "num_input_tokens_seen": 36328995,
      "step": 1709,
      "time_per_iteration": 2.704258918762207
    },
    {
      "auxiliary_loss_clip": 0.01122168,
      "auxiliary_loss_mlp": 0.01040787,
      "balance_loss_clip": 1.04720092,
      "balance_loss_mlp": 1.02871656,
      "epoch": 0.20561534299284553,
      "flos": 17016702612480.0,
      "grad_norm": 2.3755912428821557,
      "language_loss": 0.80370188,
      "learning_rate": 3.6852581732124967e-06,
      "loss": 0.82533145,
      "num_input_tokens_seen": 36346340,
      "step": 1710,
      "time_per_iteration": 2.8448944091796875
    },
    {
      "auxiliary_loss_clip": 0.01167945,
      "auxiliary_loss_mlp": 0.01035948,
      "balance_loss_clip": 1.05906534,
      "balance_loss_mlp": 1.02495074,
      "epoch": 0.20573558588348465,
      "flos": 22890467064960.0,
      "grad_norm": 2.124536978354805,
      "language_loss": 0.76632744,
      "learning_rate": 3.6848385735245213e-06,
      "loss": 0.78836632,
      "num_input_tokens_seen": 36365430,
      "step": 1711,
      "time_per_iteration": 2.8163039684295654
    },
    {
      "auxiliary_loss_clip": 0.01162568,
      "auxiliary_loss_mlp": 0.00764798,
      "balance_loss_clip": 1.0498265,
      "balance_loss_mlp": 1.00052667,
      "epoch": 0.20585582877412373,
      "flos": 24643123286400.0,
      "grad_norm": 1.9147972969682483,
      "language_loss": 0.86328906,
      "learning_rate": 3.6844187182478734e-06,
      "loss": 0.88256276,
      "num_input_tokens_seen": 36386285,
      "step": 1712,
      "time_per_iteration": 2.7554900646209717
    },
    {
      "auxiliary_loss_clip": 0.01185551,
      "auxiliary_loss_mlp": 0.01033474,
      "balance_loss_clip": 1.05643463,
      "balance_loss_mlp": 1.0223937,
      "epoch": 0.2059760716647628,
      "flos": 24206952435840.0,
      "grad_norm": 1.7425804602758708,
      "language_loss": 0.74694461,
      "learning_rate": 3.683998607446246e-06,
      "loss": 0.76913488,
      "num_input_tokens_seen": 36404935,
      "step": 1713,
      "time_per_iteration": 2.7131643295288086
    },
    {
      "auxiliary_loss_clip": 0.01175982,
      "auxiliary_loss_mlp": 0.01044147,
      "balance_loss_clip": 1.05736244,
      "balance_loss_mlp": 1.03388906,
      "epoch": 0.20609631455540192,
      "flos": 20229522606720.0,
      "grad_norm": 1.848783507742748,
      "language_loss": 0.75024647,
      "learning_rate": 3.6835782411833686e-06,
      "loss": 0.77244776,
      "num_input_tokens_seen": 36424455,
      "step": 1714,
      "time_per_iteration": 2.794527769088745
    },
    {
      "auxiliary_loss_clip": 0.01205409,
      "auxiliary_loss_mlp": 0.0104641,
      "balance_loss_clip": 1.06010795,
      "balance_loss_mlp": 1.03498352,
      "epoch": 0.206216557446041,
      "flos": 19864957518720.0,
      "grad_norm": 1.7508647436465024,
      "language_loss": 0.7423532,
      "learning_rate": 3.68315761952301e-06,
      "loss": 0.76487148,
      "num_input_tokens_seen": 36441685,
      "step": 1715,
      "time_per_iteration": 2.681442975997925
    },
    {
      "auxiliary_loss_clip": 0.01158772,
      "auxiliary_loss_mlp": 0.01037026,
      "balance_loss_clip": 1.05813968,
      "balance_loss_mlp": 1.0257901,
      "epoch": 0.2063368003366801,
      "flos": 24096311568000.0,
      "grad_norm": 2.1348600711976724,
      "language_loss": 0.83117628,
      "learning_rate": 3.6827367425289797e-06,
      "loss": 0.85313421,
      "num_input_tokens_seen": 36461460,
      "step": 1716,
      "time_per_iteration": 2.7663142681121826
    },
    {
      "auxiliary_loss_clip": 0.01183864,
      "auxiliary_loss_mlp": 0.01032176,
      "balance_loss_clip": 1.06089818,
      "balance_loss_mlp": 1.02036858,
      "epoch": 0.2064570432273192,
      "flos": 20340163474560.0,
      "grad_norm": 2.2147605315778045,
      "language_loss": 0.7237764,
      "learning_rate": 3.6823156102651225e-06,
      "loss": 0.74593687,
      "num_input_tokens_seen": 36479615,
      "step": 1717,
      "time_per_iteration": 2.785019636154175
    },
    {
      "auxiliary_loss_clip": 0.01195011,
      "auxiliary_loss_mlp": 0.01036695,
      "balance_loss_clip": 1.06089234,
      "balance_loss_mlp": 1.024863,
      "epoch": 0.20657728611795828,
      "flos": 20520363029760.0,
      "grad_norm": 1.7432983316983757,
      "language_loss": 0.70643663,
      "learning_rate": 3.6818942227953257e-06,
      "loss": 0.72875375,
      "num_input_tokens_seen": 36500160,
      "step": 1718,
      "time_per_iteration": 2.7574284076690674
    },
    {
      "auxiliary_loss_clip": 0.01139586,
      "auxiliary_loss_mlp": 0.01037322,
      "balance_loss_clip": 1.05307806,
      "balance_loss_mlp": 1.02580655,
      "epoch": 0.20669752900859736,
      "flos": 21799285752960.0,
      "grad_norm": 2.0405481714113862,
      "language_loss": 0.69002593,
      "learning_rate": 3.681472580183512e-06,
      "loss": 0.71179509,
      "num_input_tokens_seen": 36518810,
      "step": 1719,
      "time_per_iteration": 2.787923812866211
    },
    {
      "auxiliary_loss_clip": 0.0120247,
      "auxiliary_loss_mlp": 0.01030647,
      "balance_loss_clip": 1.06527877,
      "balance_loss_mlp": 1.02071691,
      "epoch": 0.20681777189923645,
      "flos": 15122020014720.0,
      "grad_norm": 3.992291845855127,
      "language_loss": 0.86514115,
      "learning_rate": 3.6810506824936455e-06,
      "loss": 0.88747233,
      "num_input_tokens_seen": 36536890,
      "step": 1720,
      "time_per_iteration": 3.910470724105835
    },
    {
      "auxiliary_loss_clip": 0.01092401,
      "auxiliary_loss_mlp": 0.01004363,
      "balance_loss_clip": 1.02969456,
      "balance_loss_mlp": 1.00253952,
      "epoch": 0.20693801478987556,
      "flos": 56481021509760.0,
      "grad_norm": 1.057442385868383,
      "language_loss": 0.62574559,
      "learning_rate": 3.680628529789726e-06,
      "loss": 0.6467132,
      "num_input_tokens_seen": 36589300,
      "step": 1721,
      "time_per_iteration": 3.086782217025757
    },
    {
      "auxiliary_loss_clip": 0.01184508,
      "auxiliary_loss_mlp": 0.0103425,
      "balance_loss_clip": 1.05778635,
      "balance_loss_mlp": 1.0211308,
      "epoch": 0.20705825768051464,
      "flos": 21614201948160.0,
      "grad_norm": 1.8089333915962422,
      "language_loss": 0.86221695,
      "learning_rate": 3.680206122135796e-06,
      "loss": 0.88440454,
      "num_input_tokens_seen": 36609905,
      "step": 1722,
      "time_per_iteration": 2.757024049758911
    },
    {
      "auxiliary_loss_clip": 0.01143648,
      "auxiliary_loss_mlp": 0.01036945,
      "balance_loss_clip": 1.04993904,
      "balance_loss_mlp": 1.02644229,
      "epoch": 0.20717850057115372,
      "flos": 25848895962240.0,
      "grad_norm": 1.9548728205582924,
      "language_loss": 0.78472173,
      "learning_rate": 3.6797834595959323e-06,
      "loss": 0.80652761,
      "num_input_tokens_seen": 36629805,
      "step": 1723,
      "time_per_iteration": 2.8121018409729004
    },
    {
      "auxiliary_loss_clip": 0.01101932,
      "auxiliary_loss_mlp": 0.01035311,
      "balance_loss_clip": 1.04284,
      "balance_loss_mlp": 1.02244198,
      "epoch": 0.20729874346179283,
      "flos": 29130807767040.0,
      "grad_norm": 3.6475636008517487,
      "language_loss": 0.77628452,
      "learning_rate": 3.679360542234254e-06,
      "loss": 0.79765695,
      "num_input_tokens_seen": 36649150,
      "step": 1724,
      "time_per_iteration": 3.8911685943603516
    },
    {
      "auxiliary_loss_clip": 0.01178253,
      "auxiliary_loss_mlp": 0.01035903,
      "balance_loss_clip": 1.05858326,
      "balance_loss_mlp": 1.02566934,
      "epoch": 0.20741898635243192,
      "flos": 29023363209600.0,
      "grad_norm": 1.6531523695598447,
      "language_loss": 0.72171652,
      "learning_rate": 3.678937370114916e-06,
      "loss": 0.7438581,
      "num_input_tokens_seen": 36668955,
      "step": 1725,
      "time_per_iteration": 4.620708465576172
    },
    {
      "auxiliary_loss_clip": 0.01167546,
      "auxiliary_loss_mlp": 0.01031658,
      "balance_loss_clip": 1.06051123,
      "balance_loss_mlp": 1.02227008,
      "epoch": 0.207539229243071,
      "flos": 15559447841280.0,
      "grad_norm": 2.146993200380475,
      "language_loss": 0.78766865,
      "learning_rate": 3.678513943302114e-06,
      "loss": 0.80966067,
      "num_input_tokens_seen": 36685730,
      "step": 1726,
      "time_per_iteration": 2.719727039337158
    },
    {
      "auxiliary_loss_clip": 0.01208628,
      "auxiliary_loss_mlp": 0.0104369,
      "balance_loss_clip": 1.06322074,
      "balance_loss_mlp": 1.03288937,
      "epoch": 0.20765947213371008,
      "flos": 20521081301760.0,
      "grad_norm": 1.8056881747735851,
      "language_loss": 0.84830987,
      "learning_rate": 3.678090261860082e-06,
      "loss": 0.87083304,
      "num_input_tokens_seen": 36705460,
      "step": 1727,
      "time_per_iteration": 2.6784114837646484
    },
    {
      "auxiliary_loss_clip": 0.01204999,
      "auxiliary_loss_mlp": 0.01035104,
      "balance_loss_clip": 1.0597527,
      "balance_loss_mlp": 1.02395773,
      "epoch": 0.2077797150243492,
      "flos": 19354415558400.0,
      "grad_norm": 2.4417211723260794,
      "language_loss": 0.78009492,
      "learning_rate": 3.6776663258530906e-06,
      "loss": 0.80249596,
      "num_input_tokens_seen": 36724110,
      "step": 1728,
      "time_per_iteration": 2.665834903717041
    },
    {
      "auxiliary_loss_clip": 0.01165114,
      "auxiliary_loss_mlp": 0.01037537,
      "balance_loss_clip": 1.05881369,
      "balance_loss_mlp": 1.02709413,
      "epoch": 0.20789995791498828,
      "flos": 21829952989440.0,
      "grad_norm": 2.184901830217271,
      "language_loss": 0.71209288,
      "learning_rate": 3.6772421353454516e-06,
      "loss": 0.73411942,
      "num_input_tokens_seen": 36742705,
      "step": 1729,
      "time_per_iteration": 2.664289712905884
    },
    {
      "auxiliary_loss_clip": 0.01154479,
      "auxiliary_loss_mlp": 0.01042021,
      "balance_loss_clip": 1.05460167,
      "balance_loss_mlp": 1.03077328,
      "epoch": 0.20802020080562736,
      "flos": 23148844571520.0,
      "grad_norm": 1.856314300533127,
      "language_loss": 0.88478869,
      "learning_rate": 3.6768176904015153e-06,
      "loss": 0.90675372,
      "num_input_tokens_seen": 36762510,
      "step": 1730,
      "time_per_iteration": 2.7542104721069336
    },
    {
      "auxiliary_loss_clip": 0.01154863,
      "auxiliary_loss_mlp": 0.01040859,
      "balance_loss_clip": 1.05199933,
      "balance_loss_mlp": 1.03014803,
      "epoch": 0.20814044369626647,
      "flos": 23072677781760.0,
      "grad_norm": 2.462566606496811,
      "language_loss": 0.59958184,
      "learning_rate": 3.6763929910856674e-06,
      "loss": 0.62153906,
      "num_input_tokens_seen": 36780960,
      "step": 1731,
      "time_per_iteration": 2.8253707885742188
    },
    {
      "auxiliary_loss_clip": 0.01164635,
      "auxiliary_loss_mlp": 0.0104363,
      "balance_loss_clip": 1.05780208,
      "balance_loss_mlp": 1.03177476,
      "epoch": 0.20826068658690555,
      "flos": 19608016556160.0,
      "grad_norm": 2.113669270170485,
      "language_loss": 0.77889204,
      "learning_rate": 3.6759680374623365e-06,
      "loss": 0.80097473,
      "num_input_tokens_seen": 36798875,
      "step": 1732,
      "time_per_iteration": 2.716578245162964
    },
    {
      "auxiliary_loss_clip": 0.01172179,
      "auxiliary_loss_mlp": 0.01040896,
      "balance_loss_clip": 1.06309438,
      "balance_loss_mlp": 1.03022099,
      "epoch": 0.20838092947754464,
      "flos": 25374049142400.0,
      "grad_norm": 2.2555683822287453,
      "language_loss": 0.74829,
      "learning_rate": 3.675542829595986e-06,
      "loss": 0.77042079,
      "num_input_tokens_seen": 36818540,
      "step": 1733,
      "time_per_iteration": 2.786162853240967
    },
    {
      "auxiliary_loss_clip": 0.01179331,
      "auxiliary_loss_mlp": 0.01034479,
      "balance_loss_clip": 1.05602741,
      "balance_loss_mlp": 1.02387571,
      "epoch": 0.20850117236818372,
      "flos": 24061729749120.0,
      "grad_norm": 1.5993334631263414,
      "language_loss": 0.79414105,
      "learning_rate": 3.6751173675511213e-06,
      "loss": 0.81627917,
      "num_input_tokens_seen": 36840585,
      "step": 1734,
      "time_per_iteration": 2.763509750366211
    },
    {
      "auxiliary_loss_clip": 0.01172978,
      "auxiliary_loss_mlp": 0.01035457,
      "balance_loss_clip": 1.05525887,
      "balance_loss_mlp": 1.02509809,
      "epoch": 0.20862141525882283,
      "flos": 20077799558400.0,
      "grad_norm": 2.2178247298638993,
      "language_loss": 0.87179124,
      "learning_rate": 3.674691651392283e-06,
      "loss": 0.8938756,
      "num_input_tokens_seen": 36858255,
      "step": 1735,
      "time_per_iteration": 2.674074411392212
    },
    {
      "auxiliary_loss_clip": 0.01221285,
      "auxiliary_loss_mlp": 0.01035594,
      "balance_loss_clip": 1.06459975,
      "balance_loss_mlp": 1.02427471,
      "epoch": 0.2087416581494619,
      "flos": 39015183237120.0,
      "grad_norm": 2.167605929571604,
      "language_loss": 0.75982165,
      "learning_rate": 3.674265681184053e-06,
      "loss": 0.78239042,
      "num_input_tokens_seen": 36881515,
      "step": 1736,
      "time_per_iteration": 2.749682664871216
    },
    {
      "auxiliary_loss_clip": 0.01206037,
      "auxiliary_loss_mlp": 0.01033787,
      "balance_loss_clip": 1.05958366,
      "balance_loss_mlp": 1.02258134,
      "epoch": 0.208861901040101,
      "flos": 26101994169600.0,
      "grad_norm": 1.7521623702321762,
      "language_loss": 0.86466062,
      "learning_rate": 3.6738394569910504e-06,
      "loss": 0.88705885,
      "num_input_tokens_seen": 36902055,
      "step": 1737,
      "time_per_iteration": 2.788294553756714
    },
    {
      "auxiliary_loss_clip": 0.01157,
      "auxiliary_loss_mlp": 0.01033994,
      "balance_loss_clip": 1.05642557,
      "balance_loss_mlp": 1.02243066,
      "epoch": 0.2089821439307401,
      "flos": 28398732675840.0,
      "grad_norm": 2.078523627464914,
      "language_loss": 0.82897246,
      "learning_rate": 3.6734129788779333e-06,
      "loss": 0.85088247,
      "num_input_tokens_seen": 36921230,
      "step": 1738,
      "time_per_iteration": 2.8292245864868164
    },
    {
      "auxiliary_loss_clip": 0.01191882,
      "auxiliary_loss_mlp": 0.00764577,
      "balance_loss_clip": 1.05943739,
      "balance_loss_mlp": 1.00045705,
      "epoch": 0.2091023868213792,
      "flos": 21069616872960.0,
      "grad_norm": 1.8752681035282033,
      "language_loss": 0.90146023,
      "learning_rate": 3.6729862469093976e-06,
      "loss": 0.9210248,
      "num_input_tokens_seen": 36940325,
      "step": 1739,
      "time_per_iteration": 2.677258253097534
    },
    {
      "auxiliary_loss_clip": 0.01156067,
      "auxiliary_loss_mlp": 0.01034287,
      "balance_loss_clip": 1.05370402,
      "balance_loss_mlp": 1.02374327,
      "epoch": 0.20922262971201827,
      "flos": 22455481363200.0,
      "grad_norm": 3.9212229613075076,
      "language_loss": 0.82925498,
      "learning_rate": 3.6725592611501782e-06,
      "loss": 0.85115862,
      "num_input_tokens_seen": 36959000,
      "step": 1740,
      "time_per_iteration": 2.814260244369507
    },
    {
      "auxiliary_loss_clip": 0.0112851,
      "auxiliary_loss_mlp": 0.00764915,
      "balance_loss_clip": 1.04792738,
      "balance_loss_mlp": 1.00047994,
      "epoch": 0.20934287260265738,
      "flos": 27852244179840.0,
      "grad_norm": 1.9560276872451257,
      "language_loss": 0.76476705,
      "learning_rate": 3.6721320216650496e-06,
      "loss": 0.78370136,
      "num_input_tokens_seen": 36979615,
      "step": 1741,
      "time_per_iteration": 2.8046481609344482
    },
    {
      "auxiliary_loss_clip": 0.01147348,
      "auxiliary_loss_mlp": 0.01033527,
      "balance_loss_clip": 1.05065513,
      "balance_loss_mlp": 1.02176738,
      "epoch": 0.20946311549329646,
      "flos": 16435309075200.0,
      "grad_norm": 1.8951183794793984,
      "language_loss": 0.83540016,
      "learning_rate": 3.6717045285188215e-06,
      "loss": 0.85720897,
      "num_input_tokens_seen": 36997310,
      "step": 1742,
      "time_per_iteration": 2.7314674854278564
    },
    {
      "auxiliary_loss_clip": 0.01163754,
      "auxiliary_loss_mlp": 0.01036875,
      "balance_loss_clip": 1.05560458,
      "balance_loss_mlp": 1.02468574,
      "epoch": 0.20958335838393555,
      "flos": 22492720788480.0,
      "grad_norm": 2.386026080173291,
      "language_loss": 0.8691929,
      "learning_rate": 3.671276781776346e-06,
      "loss": 0.89119917,
      "num_input_tokens_seen": 37015965,
      "step": 1743,
      "time_per_iteration": 2.747870445251465
    },
    {
      "auxiliary_loss_clip": 0.01173518,
      "auxiliary_loss_mlp": 0.01031861,
      "balance_loss_clip": 1.05312455,
      "balance_loss_mlp": 1.02132332,
      "epoch": 0.20970360127457463,
      "flos": 25224768218880.0,
      "grad_norm": 1.9433466389302498,
      "language_loss": 0.67651606,
      "learning_rate": 3.6708487815025128e-06,
      "loss": 0.69856989,
      "num_input_tokens_seen": 37036545,
      "step": 1744,
      "time_per_iteration": 2.7323248386383057
    },
    {
      "auxiliary_loss_clip": 0.01178853,
      "auxiliary_loss_mlp": 0.01033657,
      "balance_loss_clip": 1.05607009,
      "balance_loss_mlp": 1.0226959,
      "epoch": 0.20982384416521374,
      "flos": 18479164855680.0,
      "grad_norm": 2.1887627056138093,
      "language_loss": 0.74536073,
      "learning_rate": 3.6704205277622463e-06,
      "loss": 0.76748586,
      "num_input_tokens_seen": 37054985,
      "step": 1745,
      "time_per_iteration": 2.7089967727661133
    },
    {
      "auxiliary_loss_clip": 0.01164057,
      "auxiliary_loss_mlp": 0.01034851,
      "balance_loss_clip": 1.05228996,
      "balance_loss_mlp": 1.02396762,
      "epoch": 0.20994408705585282,
      "flos": 25373546352000.0,
      "grad_norm": 1.8552985082399653,
      "language_loss": 0.80298507,
      "learning_rate": 3.6699920206205146e-06,
      "loss": 0.82497412,
      "num_input_tokens_seen": 37075725,
      "step": 1746,
      "time_per_iteration": 3.6208131313323975
    },
    {
      "auxiliary_loss_clip": 0.01177455,
      "auxiliary_loss_mlp": 0.01036816,
      "balance_loss_clip": 1.05427837,
      "balance_loss_mlp": 1.02652216,
      "epoch": 0.2100643299464919,
      "flos": 21320955313920.0,
      "grad_norm": 1.7969039635232635,
      "language_loss": 0.81980252,
      "learning_rate": 3.669563260142321e-06,
      "loss": 0.84194523,
      "num_input_tokens_seen": 37094615,
      "step": 1747,
      "time_per_iteration": 2.6900503635406494
    },
    {
      "auxiliary_loss_clip": 0.01202605,
      "auxiliary_loss_mlp": 0.01040261,
      "balance_loss_clip": 1.05873346,
      "balance_loss_mlp": 1.02875137,
      "epoch": 0.21018457283713102,
      "flos": 19354379644800.0,
      "grad_norm": 2.0247526538847453,
      "language_loss": 0.84249192,
      "learning_rate": 3.6691342463927083e-06,
      "loss": 0.86492062,
      "num_input_tokens_seen": 37113610,
      "step": 1748,
      "time_per_iteration": 2.6639487743377686
    },
    {
      "auxiliary_loss_clip": 0.0116839,
      "auxiliary_loss_mlp": 0.01037163,
      "balance_loss_clip": 1.05140066,
      "balance_loss_mlp": 1.02614188,
      "epoch": 0.2103048157277701,
      "flos": 28330035914880.0,
      "grad_norm": 1.661814687781152,
      "language_loss": 0.81686974,
      "learning_rate": 3.668704979436758e-06,
      "loss": 0.83892524,
      "num_input_tokens_seen": 37133705,
      "step": 1749,
      "time_per_iteration": 2.7726869583129883
    },
    {
      "auxiliary_loss_clip": 0.01160091,
      "auxiliary_loss_mlp": 0.01030265,
      "balance_loss_clip": 1.04890025,
      "balance_loss_mlp": 1.01944721,
      "epoch": 0.21042505861840918,
      "flos": 17457290835840.0,
      "grad_norm": 2.6216178492860496,
      "language_loss": 0.78820473,
      "learning_rate": 3.668275459339588e-06,
      "loss": 0.8101083,
      "num_input_tokens_seen": 37152185,
      "step": 1750,
      "time_per_iteration": 4.694090127944946
    },
    {
      "auxiliary_loss_clip": 0.01205805,
      "auxiliary_loss_mlp": 0.01031839,
      "balance_loss_clip": 1.05995727,
      "balance_loss_mlp": 1.02123523,
      "epoch": 0.21054530150904827,
      "flos": 14209817195520.0,
      "grad_norm": 1.7956235277965784,
      "language_loss": 0.80183744,
      "learning_rate": 3.667845686166358e-06,
      "loss": 0.82421398,
      "num_input_tokens_seen": 37169110,
      "step": 1751,
      "time_per_iteration": 2.815218210220337
    },
    {
      "auxiliary_loss_clip": 0.0116141,
      "auxiliary_loss_mlp": 0.01042588,
      "balance_loss_clip": 1.05554855,
      "balance_loss_mlp": 1.02983832,
      "epoch": 0.21066554439968738,
      "flos": 18618210403200.0,
      "grad_norm": 1.789102652778967,
      "language_loss": 0.85680377,
      "learning_rate": 3.6674156599822634e-06,
      "loss": 0.87884378,
      "num_input_tokens_seen": 37184905,
      "step": 1752,
      "time_per_iteration": 3.6540212631225586
    },
    {
      "auxiliary_loss_clip": 0.01200787,
      "auxiliary_loss_mlp": 0.01039629,
      "balance_loss_clip": 1.06099248,
      "balance_loss_mlp": 1.02881706,
      "epoch": 0.21078578729032646,
      "flos": 23658883741440.0,
      "grad_norm": 2.0783803582696825,
      "language_loss": 0.81830657,
      "learning_rate": 3.666985380852539e-06,
      "loss": 0.84071076,
      "num_input_tokens_seen": 37203910,
      "step": 1753,
      "time_per_iteration": 2.693232536315918
    },
    {
      "auxiliary_loss_clip": 0.01147279,
      "auxiliary_loss_mlp": 0.01035714,
      "balance_loss_clip": 1.04848802,
      "balance_loss_mlp": 1.02354836,
      "epoch": 0.21090603018096554,
      "flos": 29346379240320.0,
      "grad_norm": 2.247726200075243,
      "language_loss": 0.74115384,
      "learning_rate": 3.6665548488424576e-06,
      "loss": 0.7629838,
      "num_input_tokens_seen": 37222670,
      "step": 1754,
      "time_per_iteration": 2.8366706371307373
    },
    {
      "auxiliary_loss_clip": 0.01177229,
      "auxiliary_loss_mlp": 0.00765474,
      "balance_loss_clip": 1.05589569,
      "balance_loss_mlp": 1.00051498,
      "epoch": 0.21102627307160465,
      "flos": 23261245205760.0,
      "grad_norm": 2.2110448827862617,
      "language_loss": 0.87816608,
      "learning_rate": 3.6661240640173307e-06,
      "loss": 0.89759308,
      "num_input_tokens_seen": 37244140,
      "step": 1755,
      "time_per_iteration": 2.7526674270629883
    },
    {
      "auxiliary_loss_clip": 0.011052,
      "auxiliary_loss_mlp": 0.01009537,
      "balance_loss_clip": 1.03006196,
      "balance_loss_mlp": 1.00754607,
      "epoch": 0.21114651596224374,
      "flos": 54633454577280.0,
      "grad_norm": 0.8524270760463251,
      "language_loss": 0.5784235,
      "learning_rate": 3.6656930264425085e-06,
      "loss": 0.59957081,
      "num_input_tokens_seen": 37308185,
      "step": 1756,
      "time_per_iteration": 3.2784411907196045
    },
    {
      "auxiliary_loss_clip": 0.01201185,
      "auxiliary_loss_mlp": 0.01038259,
      "balance_loss_clip": 1.06054783,
      "balance_loss_mlp": 1.02745271,
      "epoch": 0.21126675885288282,
      "flos": 21543314457600.0,
      "grad_norm": 1.7223202342331303,
      "language_loss": 0.75348842,
      "learning_rate": 3.665261736183378e-06,
      "loss": 0.77588284,
      "num_input_tokens_seen": 37328220,
      "step": 1757,
      "time_per_iteration": 2.657240390777588
    },
    {
      "auxiliary_loss_clip": 0.01128203,
      "auxiliary_loss_mlp": 0.01029738,
      "balance_loss_clip": 1.04944777,
      "balance_loss_mlp": 1.01829338,
      "epoch": 0.2113870017435219,
      "flos": 10961876678400.0,
      "grad_norm": 2.5258085583845085,
      "language_loss": 0.89034498,
      "learning_rate": 3.664830193305366e-06,
      "loss": 0.91192448,
      "num_input_tokens_seen": 37345995,
      "step": 1758,
      "time_per_iteration": 2.768983840942383
    },
    {
      "auxiliary_loss_clip": 0.01169356,
      "auxiliary_loss_mlp": 0.00765768,
      "balance_loss_clip": 1.05163193,
      "balance_loss_mlp": 1.00047159,
      "epoch": 0.211507244634161,
      "flos": 16653825463680.0,
      "grad_norm": 2.669314872662254,
      "language_loss": 0.77042067,
      "learning_rate": 3.6643983978739373e-06,
      "loss": 0.78977191,
      "num_input_tokens_seen": 37362610,
      "step": 1759,
      "time_per_iteration": 2.695024013519287
    },
    {
      "auxiliary_loss_clip": 0.01165093,
      "auxiliary_loss_mlp": 0.01040937,
      "balance_loss_clip": 1.0582552,
      "balance_loss_mlp": 1.02969587,
      "epoch": 0.2116274875248001,
      "flos": 20954091755520.0,
      "grad_norm": 1.845681716278264,
      "language_loss": 0.82514036,
      "learning_rate": 3.663966349954596e-06,
      "loss": 0.84720063,
      "num_input_tokens_seen": 37382790,
      "step": 1760,
      "time_per_iteration": 2.7670724391937256
    },
    {
      "auxiliary_loss_clip": 0.01080479,
      "auxiliary_loss_mlp": 0.01003088,
      "balance_loss_clip": 1.02910244,
      "balance_loss_mlp": 1.00119257,
      "epoch": 0.21174773041543918,
      "flos": 68196949424640.0,
      "grad_norm": 0.7956568152722244,
      "language_loss": 0.59722,
      "learning_rate": 3.6635340496128816e-06,
      "loss": 0.61805564,
      "num_input_tokens_seen": 37439720,
      "step": 1761,
      "time_per_iteration": 3.205204725265503
    },
    {
      "auxiliary_loss_clip": 0.01154509,
      "auxiliary_loss_mlp": 0.01033473,
      "balance_loss_clip": 1.05051374,
      "balance_loss_mlp": 1.02295303,
      "epoch": 0.2118679733060783,
      "flos": 20668315150080.0,
      "grad_norm": 1.7982041464725889,
      "language_loss": 0.92696702,
      "learning_rate": 3.6631014969143747e-06,
      "loss": 0.94884682,
      "num_input_tokens_seen": 37459410,
      "step": 1762,
      "time_per_iteration": 2.718318462371826
    },
    {
      "auxiliary_loss_clip": 0.01156193,
      "auxiliary_loss_mlp": 0.01034972,
      "balance_loss_clip": 1.05119812,
      "balance_loss_mlp": 1.02352834,
      "epoch": 0.21198821619671737,
      "flos": 23223431162880.0,
      "grad_norm": 1.7961890762087163,
      "language_loss": 0.88783866,
      "learning_rate": 3.662668691924693e-06,
      "loss": 0.90975034,
      "num_input_tokens_seen": 37480460,
      "step": 1763,
      "time_per_iteration": 2.792208194732666
    },
    {
      "auxiliary_loss_clip": 0.01140568,
      "auxiliary_loss_mlp": 0.01034725,
      "balance_loss_clip": 1.05416477,
      "balance_loss_mlp": 1.02296495,
      "epoch": 0.21210845908735645,
      "flos": 24498547044480.0,
      "grad_norm": 2.1692503074142255,
      "language_loss": 0.70830512,
      "learning_rate": 3.6622356347094927e-06,
      "loss": 0.73005801,
      "num_input_tokens_seen": 37502025,
      "step": 1764,
      "time_per_iteration": 2.891463041305542
    },
    {
      "auxiliary_loss_clip": 0.01166752,
      "auxiliary_loss_mlp": 0.00765853,
      "balance_loss_clip": 1.05307198,
      "balance_loss_mlp": 1.00038099,
      "epoch": 0.21222870197799554,
      "flos": 27089789160960.0,
      "grad_norm": 1.9464038052714747,
      "language_loss": 0.79207236,
      "learning_rate": 3.6618023253344684e-06,
      "loss": 0.81139845,
      "num_input_tokens_seen": 37520885,
      "step": 1765,
      "time_per_iteration": 2.815608263015747
    },
    {
      "auxiliary_loss_clip": 0.01220248,
      "auxiliary_loss_mlp": 0.01037439,
      "balance_loss_clip": 1.06259906,
      "balance_loss_mlp": 1.02588785,
      "epoch": 0.21234894486863465,
      "flos": 16873850223360.0,
      "grad_norm": 1.610841239061033,
      "language_loss": 0.83419442,
      "learning_rate": 3.6613687638653527e-06,
      "loss": 0.85677123,
      "num_input_tokens_seen": 37539055,
      "step": 1766,
      "time_per_iteration": 2.5742006301879883
    },
    {
      "auxiliary_loss_clip": 0.0113445,
      "auxiliary_loss_mlp": 0.01036517,
      "balance_loss_clip": 1.04819894,
      "balance_loss_mlp": 1.02430367,
      "epoch": 0.21246918775927373,
      "flos": 23474949171840.0,
      "grad_norm": 2.3266207480846166,
      "language_loss": 0.77792484,
      "learning_rate": 3.660934950367916e-06,
      "loss": 0.79963446,
      "num_input_tokens_seen": 37558300,
      "step": 1767,
      "time_per_iteration": 2.819441556930542
    },
    {
      "auxiliary_loss_clip": 0.01127395,
      "auxiliary_loss_mlp": 0.01043413,
      "balance_loss_clip": 1.04812288,
      "balance_loss_mlp": 1.03234482,
      "epoch": 0.21258943064991281,
      "flos": 22382295402240.0,
      "grad_norm": 1.7061541374235243,
      "language_loss": 0.83322674,
      "learning_rate": 3.660500884907968e-06,
      "loss": 0.85493481,
      "num_input_tokens_seen": 37579040,
      "step": 1768,
      "time_per_iteration": 2.786411762237549
    },
    {
      "auxiliary_loss_clip": 0.01076853,
      "auxiliary_loss_mlp": 0.01004005,
      "balance_loss_clip": 1.02715695,
      "balance_loss_mlp": 1.00179958,
      "epoch": 0.21270967354055192,
      "flos": 59440168679040.0,
      "grad_norm": 0.8211538095868576,
      "language_loss": 0.60063672,
      "learning_rate": 3.660066567551356e-06,
      "loss": 0.6214453,
      "num_input_tokens_seen": 37639185,
      "step": 1769,
      "time_per_iteration": 3.182314157485962
    },
    {
      "auxiliary_loss_clip": 0.01175622,
      "auxiliary_loss_mlp": 0.01036561,
      "balance_loss_clip": 1.05498242,
      "balance_loss_mlp": 1.02536702,
      "epoch": 0.212829916431191,
      "flos": 21544032729600.0,
      "grad_norm": 2.3239446629058094,
      "language_loss": 0.84256124,
      "learning_rate": 3.6596319983639657e-06,
      "loss": 0.86468303,
      "num_input_tokens_seen": 37657765,
      "step": 1770,
      "time_per_iteration": 2.7138829231262207
    },
    {
      "auxiliary_loss_clip": 0.01157174,
      "auxiliary_loss_mlp": 0.0103619,
      "balance_loss_clip": 1.05131125,
      "balance_loss_mlp": 1.02470458,
      "epoch": 0.2129501593218301,
      "flos": 28987739896320.0,
      "grad_norm": 1.5783767546295044,
      "language_loss": 0.86435282,
      "learning_rate": 3.6591971774117214e-06,
      "loss": 0.88628644,
      "num_input_tokens_seen": 37680740,
      "step": 1771,
      "time_per_iteration": 3.636378526687622
    },
    {
      "auxiliary_loss_clip": 0.01153309,
      "auxiliary_loss_mlp": 0.01051034,
      "balance_loss_clip": 1.0512929,
      "balance_loss_mlp": 1.0395304,
      "epoch": 0.2130704022124692,
      "flos": 18806993308800.0,
      "grad_norm": 2.120836390060668,
      "language_loss": 0.8007524,
      "learning_rate": 3.6587621047605833e-06,
      "loss": 0.82279587,
      "num_input_tokens_seen": 37697910,
      "step": 1772,
      "time_per_iteration": 2.722334623336792
    },
    {
      "auxiliary_loss_clip": 0.01187988,
      "auxiliary_loss_mlp": 0.01037985,
      "balance_loss_clip": 1.06162965,
      "balance_loss_mlp": 1.02715433,
      "epoch": 0.21319064510310828,
      "flos": 13918150759680.0,
      "grad_norm": 2.1776993458704728,
      "language_loss": 0.86548352,
      "learning_rate": 3.6583267804765542e-06,
      "loss": 0.88774323,
      "num_input_tokens_seen": 37712245,
      "step": 1773,
      "time_per_iteration": 2.6814358234405518
    },
    {
      "auxiliary_loss_clip": 0.01203089,
      "auxiliary_loss_mlp": 0.01040553,
      "balance_loss_clip": 1.05995488,
      "balance_loss_mlp": 1.02817965,
      "epoch": 0.21331088799374737,
      "flos": 20959694277120.0,
      "grad_norm": 2.0627834069321245,
      "language_loss": 0.85750782,
      "learning_rate": 3.6578912046256702e-06,
      "loss": 0.87994432,
      "num_input_tokens_seen": 37730765,
      "step": 1774,
      "time_per_iteration": 2.6324989795684814
    },
    {
      "auxiliary_loss_clip": 0.01181373,
      "auxiliary_loss_mlp": 0.0103675,
      "balance_loss_clip": 1.05909002,
      "balance_loss_mlp": 1.02454937,
      "epoch": 0.21343113088438645,
      "flos": 18624638937600.0,
      "grad_norm": 2.0885082660209493,
      "language_loss": 0.76075029,
      "learning_rate": 3.6574553772740083e-06,
      "loss": 0.78293151,
      "num_input_tokens_seen": 37748695,
      "step": 1775,
      "time_per_iteration": 3.7297940254211426
    },
    {
      "auxiliary_loss_clip": 0.01070151,
      "auxiliary_loss_mlp": 0.0075859,
      "balance_loss_clip": 1.02511573,
      "balance_loss_mlp": 1.00032568,
      "epoch": 0.21355137377502556,
      "flos": 67413128791680.0,
      "grad_norm": 0.8657045634101034,
      "language_loss": 0.61830974,
      "learning_rate": 3.657019298487684e-06,
      "loss": 0.63659716,
      "num_input_tokens_seen": 37813705,
      "step": 1776,
      "time_per_iteration": 4.408231258392334
    },
    {
      "auxiliary_loss_clip": 0.01172373,
      "auxiliary_loss_mlp": 0.010433,
      "balance_loss_clip": 1.0535512,
      "balance_loss_mlp": 1.0306282,
      "epoch": 0.21367161666566464,
      "flos": 34532095697280.0,
      "grad_norm": 2.5403978752499747,
      "language_loss": 0.83446842,
      "learning_rate": 3.6565829683328495e-06,
      "loss": 0.8566252,
      "num_input_tokens_seen": 37836330,
      "step": 1777,
      "time_per_iteration": 3.7246506214141846
    },
    {
      "auxiliary_loss_clip": 0.01203802,
      "auxiliary_loss_mlp": 0.01033272,
      "balance_loss_clip": 1.05889058,
      "balance_loss_mlp": 1.02181649,
      "epoch": 0.21379185955630373,
      "flos": 18989347680000.0,
      "grad_norm": 2.0112725902548374,
      "language_loss": 0.86065197,
      "learning_rate": 3.6561463868756965e-06,
      "loss": 0.88302267,
      "num_input_tokens_seen": 37855030,
      "step": 1778,
      "time_per_iteration": 2.621340274810791
    },
    {
      "auxiliary_loss_clip": 0.01122726,
      "auxiliary_loss_mlp": 0.01035093,
      "balance_loss_clip": 1.04768658,
      "balance_loss_mlp": 1.02280867,
      "epoch": 0.21391210244694284,
      "flos": 28218497207040.0,
      "grad_norm": 1.7791193732456205,
      "language_loss": 0.78019381,
      "learning_rate": 3.655709554182452e-06,
      "loss": 0.801772,
      "num_input_tokens_seen": 37875370,
      "step": 1779,
      "time_per_iteration": 2.8930907249450684
    },
    {
      "auxiliary_loss_clip": 0.01184221,
      "auxiliary_loss_mlp": 0.01040233,
      "balance_loss_clip": 1.05399466,
      "balance_loss_mlp": 1.02830052,
      "epoch": 0.21403234533758192,
      "flos": 17455064192640.0,
      "grad_norm": 1.821464443544204,
      "language_loss": 0.84217179,
      "learning_rate": 3.6552724703193855e-06,
      "loss": 0.86441636,
      "num_input_tokens_seen": 37892560,
      "step": 1780,
      "time_per_iteration": 2.671490430831909
    },
    {
      "auxiliary_loss_clip": 0.01085752,
      "auxiliary_loss_mlp": 0.01003924,
      "balance_loss_clip": 1.02449107,
      "balance_loss_mlp": 1.00187325,
      "epoch": 0.214152588228221,
      "flos": 51637606686720.0,
      "grad_norm": 0.7932041098310382,
      "language_loss": 0.55861104,
      "learning_rate": 3.654835135352801e-06,
      "loss": 0.57950783,
      "num_input_tokens_seen": 37947370,
      "step": 1781,
      "time_per_iteration": 3.1677303314208984
    },
    {
      "auxiliary_loss_clip": 0.01173502,
      "auxiliary_loss_mlp": 0.01037005,
      "balance_loss_clip": 1.05227768,
      "balance_loss_mlp": 1.02554333,
      "epoch": 0.21427283111886009,
      "flos": 19496154625920.0,
      "grad_norm": 5.876025673361648,
      "language_loss": 0.87460446,
      "learning_rate": 3.654397549349043e-06,
      "loss": 0.8967095,
      "num_input_tokens_seen": 37964745,
      "step": 1782,
      "time_per_iteration": 2.6891651153564453
    },
    {
      "auxiliary_loss_clip": 0.01161573,
      "auxiliary_loss_mlp": 0.01041726,
      "balance_loss_clip": 1.05337048,
      "balance_loss_mlp": 1.02954304,
      "epoch": 0.2143930740094992,
      "flos": 20084802710400.0,
      "grad_norm": 2.5675124620322243,
      "language_loss": 0.7514267,
      "learning_rate": 3.653959712374491e-06,
      "loss": 0.77345967,
      "num_input_tokens_seen": 37982850,
      "step": 1783,
      "time_per_iteration": 2.7720730304718018
    },
    {
      "auxiliary_loss_clip": 0.01186337,
      "auxiliary_loss_mlp": 0.01037398,
      "balance_loss_clip": 1.05796385,
      "balance_loss_mlp": 1.02621007,
      "epoch": 0.21451331690013828,
      "flos": 21798603394560.0,
      "grad_norm": 1.860941241720039,
      "language_loss": 0.82990241,
      "learning_rate": 3.6535216244955663e-06,
      "loss": 0.85213971,
      "num_input_tokens_seen": 38002745,
      "step": 1784,
      "time_per_iteration": 2.7066638469696045
    },
    {
      "auxiliary_loss_clip": 0.0117596,
      "auxiliary_loss_mlp": 0.01035636,
      "balance_loss_clip": 1.05162096,
      "balance_loss_mlp": 1.02407861,
      "epoch": 0.21463355979077736,
      "flos": 32853882412800.0,
      "grad_norm": 2.2015614562546877,
      "language_loss": 0.70916373,
      "learning_rate": 3.653083285778726e-06,
      "loss": 0.73127967,
      "num_input_tokens_seen": 38024115,
      "step": 1785,
      "time_per_iteration": 2.904989004135132
    },
    {
      "auxiliary_loss_clip": 0.01173782,
      "auxiliary_loss_mlp": 0.01042194,
      "balance_loss_clip": 1.05641055,
      "balance_loss_mlp": 1.03027332,
      "epoch": 0.21475380268141647,
      "flos": 21543817248000.0,
      "grad_norm": 2.15898892102148,
      "language_loss": 0.812289,
      "learning_rate": 3.6526446962904653e-06,
      "loss": 0.8344487,
      "num_input_tokens_seen": 38042830,
      "step": 1786,
      "time_per_iteration": 2.7682366371154785
    },
    {
      "auxiliary_loss_clip": 0.01141921,
      "auxiliary_loss_mlp": 0.01037492,
      "balance_loss_clip": 1.05127597,
      "balance_loss_mlp": 1.02570236,
      "epoch": 0.21487404557205556,
      "flos": 32159082660480.0,
      "grad_norm": 1.7113016023265613,
      "language_loss": 0.74074572,
      "learning_rate": 3.652205856097318e-06,
      "loss": 0.76253986,
      "num_input_tokens_seen": 38066015,
      "step": 1787,
      "time_per_iteration": 2.9366443157196045
    },
    {
      "auxiliary_loss_clip": 0.0116239,
      "auxiliary_loss_mlp": 0.01038104,
      "balance_loss_clip": 1.0547092,
      "balance_loss_mlp": 1.02557516,
      "epoch": 0.21499428846269464,
      "flos": 12673091583360.0,
      "grad_norm": 6.953478779138558,
      "language_loss": 0.78974581,
      "learning_rate": 3.651766765265856e-06,
      "loss": 0.81175071,
      "num_input_tokens_seen": 38083025,
      "step": 1788,
      "time_per_iteration": 2.6956708431243896
    },
    {
      "auxiliary_loss_clip": 0.01154971,
      "auxiliary_loss_mlp": 0.0076532,
      "balance_loss_clip": 1.0500176,
      "balance_loss_mlp": 1.00034142,
      "epoch": 0.21511453135333372,
      "flos": 23471573293440.0,
      "grad_norm": 2.625495758754628,
      "language_loss": 0.8105675,
      "learning_rate": 3.65132742386269e-06,
      "loss": 0.82977039,
      "num_input_tokens_seen": 38098245,
      "step": 1789,
      "time_per_iteration": 2.729624032974243
    },
    {
      "auxiliary_loss_clip": 0.01171729,
      "auxiliary_loss_mlp": 0.01037691,
      "balance_loss_clip": 1.05316484,
      "balance_loss_mlp": 1.0263778,
      "epoch": 0.21523477424397283,
      "flos": 26943560893440.0,
      "grad_norm": 1.9055027602054273,
      "language_loss": 0.84923631,
      "learning_rate": 3.6508878319544656e-06,
      "loss": 0.8713305,
      "num_input_tokens_seen": 38118460,
      "step": 1790,
      "time_per_iteration": 2.774273633956909
    },
    {
      "auxiliary_loss_clip": 0.01140487,
      "auxiliary_loss_mlp": 0.01036308,
      "balance_loss_clip": 1.05084372,
      "balance_loss_mlp": 1.02467334,
      "epoch": 0.21535501713461191,
      "flos": 18916161719040.0,
      "grad_norm": 2.6101075083150347,
      "language_loss": 0.81459206,
      "learning_rate": 3.65044798960787e-06,
      "loss": 0.83636004,
      "num_input_tokens_seen": 38136800,
      "step": 1791,
      "time_per_iteration": 2.7863340377807617
    },
    {
      "auxiliary_loss_clip": 0.01183922,
      "auxiliary_loss_mlp": 0.01039015,
      "balance_loss_clip": 1.05742455,
      "balance_loss_mlp": 1.02794027,
      "epoch": 0.215475260025251,
      "flos": 17895113712000.0,
      "grad_norm": 1.9704878924460436,
      "language_loss": 0.78627813,
      "learning_rate": 3.650007896889627e-06,
      "loss": 0.80850756,
      "num_input_tokens_seen": 38155380,
      "step": 1792,
      "time_per_iteration": 2.6855647563934326
    },
    {
      "auxiliary_loss_clip": 0.01179731,
      "auxiliary_loss_mlp": 0.01042885,
      "balance_loss_clip": 1.05779314,
      "balance_loss_mlp": 1.03147674,
      "epoch": 0.2155955029158901,
      "flos": 16654292340480.0,
      "grad_norm": 2.115128278382645,
      "language_loss": 0.80726731,
      "learning_rate": 3.6495675538664974e-06,
      "loss": 0.8294934,
      "num_input_tokens_seen": 38174395,
      "step": 1793,
      "time_per_iteration": 2.7121572494506836
    },
    {
      "auxiliary_loss_clip": 0.01180838,
      "auxiliary_loss_mlp": 0.01031985,
      "balance_loss_clip": 1.05697119,
      "balance_loss_mlp": 1.02129149,
      "epoch": 0.2157157458065292,
      "flos": 23621213352960.0,
      "grad_norm": 1.7770305604455272,
      "language_loss": 0.82799208,
      "learning_rate": 3.649126960605282e-06,
      "loss": 0.85012031,
      "num_input_tokens_seen": 38195380,
      "step": 1794,
      "time_per_iteration": 2.7218005657196045
    },
    {
      "auxiliary_loss_clip": 0.01165617,
      "auxiliary_loss_mlp": 0.01034642,
      "balance_loss_clip": 1.04919457,
      "balance_loss_mlp": 1.02262568,
      "epoch": 0.21583598869716827,
      "flos": 22127078292480.0,
      "grad_norm": 2.5975350036507754,
      "language_loss": 0.83648205,
      "learning_rate": 3.6486861171728174e-06,
      "loss": 0.85848469,
      "num_input_tokens_seen": 38213775,
      "step": 1795,
      "time_per_iteration": 2.6838912963867188
    },
    {
      "auxiliary_loss_clip": 0.01189866,
      "auxiliary_loss_mlp": 0.0104468,
      "balance_loss_clip": 1.05690813,
      "balance_loss_mlp": 1.03239536,
      "epoch": 0.21595623158780738,
      "flos": 23441229279360.0,
      "grad_norm": 1.8254952572308227,
      "language_loss": 0.78713548,
      "learning_rate": 3.6482450236359803e-06,
      "loss": 0.80948097,
      "num_input_tokens_seen": 38235630,
      "step": 1796,
      "time_per_iteration": 2.7140047550201416
    },
    {
      "auxiliary_loss_clip": 0.01198224,
      "auxiliary_loss_mlp": 0.01038947,
      "balance_loss_clip": 1.06008983,
      "balance_loss_mlp": 1.02825427,
      "epoch": 0.21607647447844647,
      "flos": 26906501036160.0,
      "grad_norm": 2.2173187131201093,
      "language_loss": 0.77928913,
      "learning_rate": 3.647803680061683e-06,
      "loss": 0.8016609,
      "num_input_tokens_seen": 38256045,
      "step": 1797,
      "time_per_iteration": 3.6328043937683105
    },
    {
      "auxiliary_loss_clip": 0.01201401,
      "auxiliary_loss_mlp": 0.01038942,
      "balance_loss_clip": 1.05896783,
      "balance_loss_mlp": 1.02718854,
      "epoch": 0.21619671736908555,
      "flos": 14495378319360.0,
      "grad_norm": 5.791481531349088,
      "language_loss": 0.75099051,
      "learning_rate": 3.6473620865168776e-06,
      "loss": 0.77339393,
      "num_input_tokens_seen": 38272915,
      "step": 1798,
      "time_per_iteration": 2.612157106399536
    },
    {
      "auxiliary_loss_clip": 0.01185501,
      "auxiliary_loss_mlp": 0.01035569,
      "balance_loss_clip": 1.05834341,
      "balance_loss_mlp": 1.02466094,
      "epoch": 0.21631696025972463,
      "flos": 17931096161280.0,
      "grad_norm": 2.6809773668188774,
      "language_loss": 0.81282651,
      "learning_rate": 3.646920243068554e-06,
      "loss": 0.83503723,
      "num_input_tokens_seen": 38290810,
      "step": 1799,
      "time_per_iteration": 2.7417197227478027
    },
    {
      "auxiliary_loss_clip": 0.01153081,
      "auxiliary_loss_mlp": 0.00765473,
      "balance_loss_clip": 1.06015706,
      "balance_loss_mlp": 1.00036561,
      "epoch": 0.21643720315036374,
      "flos": 24462385027200.0,
      "grad_norm": 2.512695315801428,
      "language_loss": 0.74437064,
      "learning_rate": 3.6464781497837384e-06,
      "loss": 0.76355624,
      "num_input_tokens_seen": 38312785,
      "step": 1800,
      "time_per_iteration": 2.7934679985046387
    },
    {
      "auxiliary_loss_clip": 0.01215427,
      "auxiliary_loss_mlp": 0.01040695,
      "balance_loss_clip": 1.05971646,
      "balance_loss_mlp": 1.0286963,
      "epoch": 0.21655744604100283,
      "flos": 28474432588800.0,
      "grad_norm": 18.804856745043427,
      "language_loss": 0.72865367,
      "learning_rate": 3.6460358067294965e-06,
      "loss": 0.75121486,
      "num_input_tokens_seen": 38334015,
      "step": 1801,
      "time_per_iteration": 3.674217700958252
    },
    {
      "auxiliary_loss_clip": 0.01192561,
      "auxiliary_loss_mlp": 0.01034924,
      "balance_loss_clip": 1.05827653,
      "balance_loss_mlp": 1.0219543,
      "epoch": 0.2166776889316419,
      "flos": 20152960767360.0,
      "grad_norm": 1.9966453879353059,
      "language_loss": 0.77491164,
      "learning_rate": 3.645593213972932e-06,
      "loss": 0.79718649,
      "num_input_tokens_seen": 38352920,
      "step": 1802,
      "time_per_iteration": 3.608604907989502
    },
    {
      "auxiliary_loss_clip": 0.01101372,
      "auxiliary_loss_mlp": 0.01034567,
      "balance_loss_clip": 1.04081976,
      "balance_loss_mlp": 1.02313447,
      "epoch": 0.21679793182228102,
      "flos": 15193482122880.0,
      "grad_norm": 1.9313783091855445,
      "language_loss": 0.80163944,
      "learning_rate": 3.6451503715811852e-06,
      "loss": 0.82299876,
      "num_input_tokens_seen": 38371230,
      "step": 1803,
      "time_per_iteration": 3.785277843475342
    },
    {
      "auxiliary_loss_clip": 0.01132138,
      "auxiliary_loss_mlp": 0.01041958,
      "balance_loss_clip": 1.05011559,
      "balance_loss_mlp": 1.0309726,
      "epoch": 0.2169181747129201,
      "flos": 17384464010880.0,
      "grad_norm": 1.9716444966751066,
      "language_loss": 0.80071288,
      "learning_rate": 3.6447072796214345e-06,
      "loss": 0.8224538,
      "num_input_tokens_seen": 38389795,
      "step": 1804,
      "time_per_iteration": 3.1186299324035645
    },
    {
      "auxiliary_loss_clip": 0.0109883,
      "auxiliary_loss_mlp": 0.01006101,
      "balance_loss_clip": 1.02701378,
      "balance_loss_mlp": 1.00406253,
      "epoch": 0.21703841760355919,
      "flos": 58760955429120.0,
      "grad_norm": 0.9827028682988537,
      "language_loss": 0.63165367,
      "learning_rate": 3.644263938160898e-06,
      "loss": 0.65270299,
      "num_input_tokens_seen": 38445760,
      "step": 1805,
      "time_per_iteration": 3.1266465187072754
    },
    {
      "auxiliary_loss_clip": 0.01206363,
      "auxiliary_loss_mlp": 0.0104493,
      "balance_loss_clip": 1.06127119,
      "balance_loss_mlp": 1.03253222,
      "epoch": 0.21715866049419827,
      "flos": 22418457419520.0,
      "grad_norm": 2.685260916583857,
      "language_loss": 0.72020864,
      "learning_rate": 3.6438203472668293e-06,
      "loss": 0.74272156,
      "num_input_tokens_seen": 38465405,
      "step": 1806,
      "time_per_iteration": 2.6602675914764404
    },
    {
      "auxiliary_loss_clip": 0.0119413,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 1.05443192,
      "balance_loss_mlp": 1.01770878,
      "epoch": 0.21727890338483738,
      "flos": 17237732952960.0,
      "grad_norm": 1.865513998256247,
      "language_loss": 0.81819814,
      "learning_rate": 3.6433765070065206e-06,
      "loss": 0.84041786,
      "num_input_tokens_seen": 38483195,
      "step": 1807,
      "time_per_iteration": 2.628082752227783
    },
    {
      "auxiliary_loss_clip": 0.01118323,
      "auxiliary_loss_mlp": 0.01034601,
      "balance_loss_clip": 1.04658329,
      "balance_loss_mlp": 1.02324009,
      "epoch": 0.21739914627547646,
      "flos": 13434792416640.0,
      "grad_norm": 2.340440757888809,
      "language_loss": 0.87650305,
      "learning_rate": 3.6429324174473025e-06,
      "loss": 0.89803231,
      "num_input_tokens_seen": 38496735,
      "step": 1808,
      "time_per_iteration": 2.783611536026001
    },
    {
      "auxiliary_loss_clip": 0.01147549,
      "auxiliary_loss_mlp": 0.01030959,
      "balance_loss_clip": 1.04785585,
      "balance_loss_mlp": 1.01987863,
      "epoch": 0.21751938916611555,
      "flos": 20959514709120.0,
      "grad_norm": 2.0861686373612787,
      "language_loss": 0.85040259,
      "learning_rate": 3.6424880786565425e-06,
      "loss": 0.87218761,
      "num_input_tokens_seen": 38512880,
      "step": 1809,
      "time_per_iteration": 2.733973979949951
    },
    {
      "auxiliary_loss_clip": 0.01136462,
      "auxiliary_loss_mlp": 0.00766029,
      "balance_loss_clip": 1.04841638,
      "balance_loss_mlp": 1.00033259,
      "epoch": 0.21763963205675466,
      "flos": 27599936071680.0,
      "grad_norm": 2.116328436360984,
      "language_loss": 0.79629505,
      "learning_rate": 3.6420434907016482e-06,
      "loss": 0.81531996,
      "num_input_tokens_seen": 38532570,
      "step": 1810,
      "time_per_iteration": 2.8336687088012695
    },
    {
      "auxiliary_loss_clip": 0.01169048,
      "auxiliary_loss_mlp": 0.01034058,
      "balance_loss_clip": 1.05960822,
      "balance_loss_mlp": 1.02313209,
      "epoch": 0.21775987494739374,
      "flos": 21430411032960.0,
      "grad_norm": 1.6748250445047315,
      "language_loss": 0.81244385,
      "learning_rate": 3.6415986536500606e-06,
      "loss": 0.83447486,
      "num_input_tokens_seen": 38550900,
      "step": 1811,
      "time_per_iteration": 2.753641128540039
    },
    {
      "auxiliary_loss_clip": 0.0117688,
      "auxiliary_loss_mlp": 0.01031426,
      "balance_loss_clip": 1.05653298,
      "balance_loss_mlp": 1.02060163,
      "epoch": 0.21788011783803282,
      "flos": 18332972501760.0,
      "grad_norm": 1.7426095010193192,
      "language_loss": 0.80478847,
      "learning_rate": 3.641153567569263e-06,
      "loss": 0.82687151,
      "num_input_tokens_seen": 38569215,
      "step": 1812,
      "time_per_iteration": 2.783820867538452
    },
    {
      "auxiliary_loss_clip": 0.01141992,
      "auxiliary_loss_mlp": 0.01032972,
      "balance_loss_clip": 1.05000973,
      "balance_loss_mlp": 1.02189732,
      "epoch": 0.2180003607286719,
      "flos": 30262748037120.0,
      "grad_norm": 2.423851757577313,
      "language_loss": 0.95316631,
      "learning_rate": 3.640708232526774e-06,
      "loss": 0.97491598,
      "num_input_tokens_seen": 38587870,
      "step": 1813,
      "time_per_iteration": 2.8321826457977295
    },
    {
      "auxiliary_loss_clip": 0.01135113,
      "auxiliary_loss_mlp": 0.01038359,
      "balance_loss_clip": 1.04641938,
      "balance_loss_mlp": 1.02639031,
      "epoch": 0.21812060361931102,
      "flos": 25480272637440.0,
      "grad_norm": 1.9667786502604003,
      "language_loss": 0.78513885,
      "learning_rate": 3.6402626485901504e-06,
      "loss": 0.80687356,
      "num_input_tokens_seen": 38606965,
      "step": 1814,
      "time_per_iteration": 2.8207473754882812
    },
    {
      "auxiliary_loss_clip": 0.01198635,
      "auxiliary_loss_mlp": 0.01039595,
      "balance_loss_clip": 1.06282234,
      "balance_loss_mlp": 1.02958119,
      "epoch": 0.2182408465099501,
      "flos": 21908166854400.0,
      "grad_norm": 1.8999897994999693,
      "language_loss": 0.78703964,
      "learning_rate": 3.639816815826988e-06,
      "loss": 0.8094219,
      "num_input_tokens_seen": 38626290,
      "step": 1815,
      "time_per_iteration": 2.712893486022949
    },
    {
      "auxiliary_loss_clip": 0.01171256,
      "auxiliary_loss_mlp": 0.0104284,
      "balance_loss_clip": 1.05265141,
      "balance_loss_mlp": 1.03163409,
      "epoch": 0.21836108940058918,
      "flos": 23657339456640.0,
      "grad_norm": 2.1969398744273163,
      "language_loss": 0.78060406,
      "learning_rate": 3.6393707343049176e-06,
      "loss": 0.80274498,
      "num_input_tokens_seen": 38646620,
      "step": 1816,
      "time_per_iteration": 2.6810617446899414
    },
    {
      "auxiliary_loss_clip": 0.01143285,
      "auxiliary_loss_mlp": 0.01039942,
      "balance_loss_clip": 1.05172932,
      "balance_loss_mlp": 1.02853966,
      "epoch": 0.2184813322912283,
      "flos": 24681009156480.0,
      "grad_norm": 2.621756988355913,
      "language_loss": 0.73163903,
      "learning_rate": 3.6389244040916104e-06,
      "loss": 0.75347131,
      "num_input_tokens_seen": 38665695,
      "step": 1817,
      "time_per_iteration": 2.8512349128723145
    },
    {
      "auxiliary_loss_clip": 0.01122603,
      "auxiliary_loss_mlp": 0.01040216,
      "balance_loss_clip": 1.04816377,
      "balance_loss_mlp": 1.02802682,
      "epoch": 0.21860157518186737,
      "flos": 26574650259840.0,
      "grad_norm": 2.108267110360847,
      "language_loss": 0.79582614,
      "learning_rate": 3.6384778252547747e-06,
      "loss": 0.81745434,
      "num_input_tokens_seen": 38681575,
      "step": 1818,
      "time_per_iteration": 2.871464967727661
    },
    {
      "auxiliary_loss_clip": 0.011347,
      "auxiliary_loss_mlp": 0.01036588,
      "balance_loss_clip": 1.05433917,
      "balance_loss_mlp": 1.02460194,
      "epoch": 0.21872181807250646,
      "flos": 20886292834560.0,
      "grad_norm": 2.2492300295973293,
      "language_loss": 0.78071439,
      "learning_rate": 3.638030997862155e-06,
      "loss": 0.80242723,
      "num_input_tokens_seen": 38700510,
      "step": 1819,
      "time_per_iteration": 2.844710111618042
    },
    {
      "auxiliary_loss_clip": 0.01069183,
      "auxiliary_loss_mlp": 0.0100235,
      "balance_loss_clip": 1.02776158,
      "balance_loss_mlp": 1.00065732,
      "epoch": 0.21884206096314554,
      "flos": 61209452897280.0,
      "grad_norm": 0.757853788161933,
      "language_loss": 0.59411681,
      "learning_rate": 3.6375839219815356e-06,
      "loss": 0.6148321,
      "num_input_tokens_seen": 38758310,
      "step": 1820,
      "time_per_iteration": 3.253613233566284
    },
    {
      "auxiliary_loss_clip": 0.01169084,
      "auxiliary_loss_mlp": 0.01031989,
      "balance_loss_clip": 1.05715501,
      "balance_loss_mlp": 1.02051544,
      "epoch": 0.21896230385378465,
      "flos": 23473835850240.0,
      "grad_norm": 2.0066201276020945,
      "language_loss": 0.82438886,
      "learning_rate": 3.6371365976807375e-06,
      "loss": 0.84639955,
      "num_input_tokens_seen": 38778705,
      "step": 1821,
      "time_per_iteration": 2.737171173095703
    },
    {
      "auxiliary_loss_clip": 0.01139144,
      "auxiliary_loss_mlp": 0.01032391,
      "balance_loss_clip": 1.04982543,
      "balance_loss_mlp": 1.02124536,
      "epoch": 0.21908254674442373,
      "flos": 25081915829760.0,
      "grad_norm": 1.6860208085192028,
      "language_loss": 0.83402479,
      "learning_rate": 3.6366890250276185e-06,
      "loss": 0.85574013,
      "num_input_tokens_seen": 38799660,
      "step": 1822,
      "time_per_iteration": 3.0961427688598633
    },
    {
      "auxiliary_loss_clip": 0.01158941,
      "auxiliary_loss_mlp": 0.01035348,
      "balance_loss_clip": 1.05304432,
      "balance_loss_mlp": 1.02496481,
      "epoch": 0.21920278963506282,
      "flos": 23513768795520.0,
      "grad_norm": 1.9255482555205954,
      "language_loss": 0.90317518,
      "learning_rate": 3.6362412040900764e-06,
      "loss": 0.92511809,
      "num_input_tokens_seen": 38819450,
      "step": 1823,
      "time_per_iteration": 3.7413086891174316
    },
    {
      "auxiliary_loss_clip": 0.01155769,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 1.0506016,
      "balance_loss_mlp": 1.02046931,
      "epoch": 0.21932303252570193,
      "flos": 29242238734080.0,
      "grad_norm": 2.86144029820175,
      "language_loss": 0.79874307,
      "learning_rate": 3.635793134936044e-06,
      "loss": 0.82061231,
      "num_input_tokens_seen": 38840460,
      "step": 1824,
      "time_per_iteration": 2.873940944671631
    },
    {
      "auxiliary_loss_clip": 0.01168894,
      "auxiliary_loss_mlp": 0.01035572,
      "balance_loss_clip": 1.05700445,
      "balance_loss_mlp": 1.02499223,
      "epoch": 0.219443275416341,
      "flos": 20806857907200.0,
      "grad_norm": 1.7968930529753282,
      "language_loss": 0.73217332,
      "learning_rate": 3.635344817633494e-06,
      "loss": 0.75421798,
      "num_input_tokens_seen": 38859775,
      "step": 1825,
      "time_per_iteration": 2.727259635925293
    },
    {
      "auxiliary_loss_clip": 0.01138247,
      "auxiliary_loss_mlp": 0.01038944,
      "balance_loss_clip": 1.04868746,
      "balance_loss_mlp": 1.02825665,
      "epoch": 0.2195635183069801,
      "flos": 14501555458560.0,
      "grad_norm": 2.367066912977883,
      "language_loss": 0.75332028,
      "learning_rate": 3.634896252250436e-06,
      "loss": 0.77509218,
      "num_input_tokens_seen": 38876540,
      "step": 1826,
      "time_per_iteration": 2.8203063011169434
    },
    {
      "auxiliary_loss_clip": 0.01177604,
      "auxiliary_loss_mlp": 0.01037192,
      "balance_loss_clip": 1.05599439,
      "balance_loss_mlp": 1.02621865,
      "epoch": 0.2196837611976192,
      "flos": 24243473589120.0,
      "grad_norm": 1.8126874886460143,
      "language_loss": 0.82244277,
      "learning_rate": 3.6344474388549157e-06,
      "loss": 0.84459078,
      "num_input_tokens_seen": 38896195,
      "step": 1827,
      "time_per_iteration": 3.83203387260437
    },
    {
      "auxiliary_loss_clip": 0.01181511,
      "auxiliary_loss_mlp": 0.00765131,
      "balance_loss_clip": 1.05774689,
      "balance_loss_mlp": 1.00036287,
      "epoch": 0.2198040040882583,
      "flos": 18074523168000.0,
      "grad_norm": 4.146404030766358,
      "language_loss": 0.80262315,
      "learning_rate": 3.6339983775150183e-06,
      "loss": 0.82208955,
      "num_input_tokens_seen": 38912755,
      "step": 1828,
      "time_per_iteration": 3.609074592590332
    },
    {
      "auxiliary_loss_clip": 0.0120569,
      "auxiliary_loss_mlp": 0.01037126,
      "balance_loss_clip": 1.06219125,
      "balance_loss_mlp": 1.02488923,
      "epoch": 0.21992424697889737,
      "flos": 17784185535360.0,
      "grad_norm": 3.056144144586961,
      "language_loss": 0.84042042,
      "learning_rate": 3.6335490682988664e-06,
      "loss": 0.86284864,
      "num_input_tokens_seen": 38928365,
      "step": 1829,
      "time_per_iteration": 3.6118383407592773
    },
    {
      "auxiliary_loss_clip": 0.01184924,
      "auxiliary_loss_mlp": 0.0076585,
      "balance_loss_clip": 1.05716014,
      "balance_loss_mlp": 1.00039291,
      "epoch": 0.22004448986953645,
      "flos": 17638495971840.0,
      "grad_norm": 1.8828708931943205,
      "language_loss": 0.82882965,
      "learning_rate": 3.63309951127462e-06,
      "loss": 0.84833741,
      "num_input_tokens_seen": 38945275,
      "step": 1830,
      "time_per_iteration": 2.6918935775756836
    },
    {
      "auxiliary_loss_clip": 0.01188092,
      "auxiliary_loss_mlp": 0.01036743,
      "balance_loss_clip": 1.05746937,
      "balance_loss_mlp": 1.02365947,
      "epoch": 0.22016473276017556,
      "flos": 22275533203200.0,
      "grad_norm": 1.8741378408844953,
      "language_loss": 0.75216436,
      "learning_rate": 3.6326497065104757e-06,
      "loss": 0.77441269,
      "num_input_tokens_seen": 38965740,
      "step": 1831,
      "time_per_iteration": 2.7657206058502197
    },
    {
      "auxiliary_loss_clip": 0.01215785,
      "auxiliary_loss_mlp": 0.01034649,
      "balance_loss_clip": 1.06168735,
      "balance_loss_mlp": 1.02405083,
      "epoch": 0.22028497565081465,
      "flos": 25556259859200.0,
      "grad_norm": 2.461880758248314,
      "language_loss": 0.77577162,
      "learning_rate": 3.6321996540746697e-06,
      "loss": 0.79827595,
      "num_input_tokens_seen": 38984815,
      "step": 1832,
      "time_per_iteration": 2.663907766342163
    },
    {
      "auxiliary_loss_clip": 0.0118597,
      "auxiliary_loss_mlp": 0.0103858,
      "balance_loss_clip": 1.05980325,
      "balance_loss_mlp": 1.0277915,
      "epoch": 0.22040521854145373,
      "flos": 36247332925440.0,
      "grad_norm": 2.03112491026417,
      "language_loss": 0.80850977,
      "learning_rate": 3.6317493540354733e-06,
      "loss": 0.83075523,
      "num_input_tokens_seen": 39008230,
      "step": 1833,
      "time_per_iteration": 2.8141472339630127
    },
    {
      "auxiliary_loss_clip": 0.01125392,
      "auxiliary_loss_mlp": 0.01045232,
      "balance_loss_clip": 1.0539546,
      "balance_loss_mlp": 1.03211904,
      "epoch": 0.22052546143209284,
      "flos": 11838420270720.0,
      "grad_norm": 2.297776444290168,
      "language_loss": 0.77141112,
      "learning_rate": 3.6312988064611976e-06,
      "loss": 0.7931174,
      "num_input_tokens_seen": 39026540,
      "step": 1834,
      "time_per_iteration": 2.796003580093384
    },
    {
      "auxiliary_loss_clip": 0.01132879,
      "auxiliary_loss_mlp": 0.01033019,
      "balance_loss_clip": 1.04830563,
      "balance_loss_mlp": 1.0213902,
      "epoch": 0.22064570432273192,
      "flos": 24209250906240.0,
      "grad_norm": 1.7996014308492991,
      "language_loss": 0.81267774,
      "learning_rate": 3.6308480114201896e-06,
      "loss": 0.83433676,
      "num_input_tokens_seen": 39048460,
      "step": 1835,
      "time_per_iteration": 2.8243062496185303
    },
    {
      "auxiliary_loss_clip": 0.01146758,
      "auxiliary_loss_mlp": 0.00765554,
      "balance_loss_clip": 1.05509555,
      "balance_loss_mlp": 1.00035369,
      "epoch": 0.220765947213371,
      "flos": 17931347556480.0,
      "grad_norm": 1.87919746168898,
      "language_loss": 0.76745355,
      "learning_rate": 3.630396968980835e-06,
      "loss": 0.78657663,
      "num_input_tokens_seen": 39066335,
      "step": 1836,
      "time_per_iteration": 2.7676312923431396
    },
    {
      "auxiliary_loss_clip": 0.01148218,
      "auxiliary_loss_mlp": 0.01037936,
      "balance_loss_clip": 1.04696155,
      "balance_loss_mlp": 1.02574706,
      "epoch": 0.2208861901040101,
      "flos": 26757040544640.0,
      "grad_norm": 2.254474260994776,
      "language_loss": 0.83662122,
      "learning_rate": 3.6299456792115575e-06,
      "loss": 0.85848278,
      "num_input_tokens_seen": 39087590,
      "step": 1837,
      "time_per_iteration": 2.772610902786255
    },
    {
      "auxiliary_loss_clip": 0.01186836,
      "auxiliary_loss_mlp": 0.01036069,
      "balance_loss_clip": 1.06192243,
      "balance_loss_mlp": 1.02457738,
      "epoch": 0.2210064329946492,
      "flos": 17817977255040.0,
      "grad_norm": 1.8612550773361785,
      "language_loss": 0.80926454,
      "learning_rate": 3.629494142180815e-06,
      "loss": 0.83149356,
      "num_input_tokens_seen": 39106335,
      "step": 1838,
      "time_per_iteration": 2.652824878692627
    },
    {
      "auxiliary_loss_clip": 0.01139902,
      "auxiliary_loss_mlp": 0.01031986,
      "balance_loss_clip": 1.05020905,
      "balance_loss_mlp": 1.02128696,
      "epoch": 0.22112667588528828,
      "flos": 17967401832960.0,
      "grad_norm": 2.4261828052191388,
      "language_loss": 0.8536073,
      "learning_rate": 3.6290423579571075e-06,
      "loss": 0.87532616,
      "num_input_tokens_seen": 39122875,
      "step": 1839,
      "time_per_iteration": 2.7587814331054688
    },
    {
      "auxiliary_loss_clip": 0.01173899,
      "auxiliary_loss_mlp": 0.01031838,
      "balance_loss_clip": 1.05440974,
      "balance_loss_mlp": 1.02054334,
      "epoch": 0.22124691877592736,
      "flos": 18369206346240.0,
      "grad_norm": 3.151809181799272,
      "language_loss": 0.80358034,
      "learning_rate": 3.6285903266089694e-06,
      "loss": 0.8256377,
      "num_input_tokens_seen": 39142150,
      "step": 1840,
      "time_per_iteration": 2.7496907711029053
    },
    {
      "auxiliary_loss_clip": 0.01168066,
      "auxiliary_loss_mlp": 0.0103633,
      "balance_loss_clip": 1.05704236,
      "balance_loss_mlp": 1.02471948,
      "epoch": 0.22136716166656648,
      "flos": 20813286441600.0,
      "grad_norm": 1.8945199887821305,
      "language_loss": 0.77643287,
      "learning_rate": 3.628138048204974e-06,
      "loss": 0.79847682,
      "num_input_tokens_seen": 39162835,
      "step": 1841,
      "time_per_iteration": 2.763044834136963
    },
    {
      "auxiliary_loss_clip": 0.0117671,
      "auxiliary_loss_mlp": 0.01036096,
      "balance_loss_clip": 1.05724454,
      "balance_loss_mlp": 1.0233351,
      "epoch": 0.22148740455720556,
      "flos": 17675699483520.0,
      "grad_norm": 4.505772032932523,
      "language_loss": 0.76147342,
      "learning_rate": 3.6276855228137304e-06,
      "loss": 0.78360146,
      "num_input_tokens_seen": 39181040,
      "step": 1842,
      "time_per_iteration": 2.7207272052764893
    },
    {
      "auxiliary_loss_clip": 0.01183221,
      "auxiliary_loss_mlp": 0.01043235,
      "balance_loss_clip": 1.05719316,
      "balance_loss_mlp": 1.03040266,
      "epoch": 0.22160764744784464,
      "flos": 21726710323200.0,
      "grad_norm": 3.7629668773277563,
      "language_loss": 0.81978738,
      "learning_rate": 3.6272327505038874e-06,
      "loss": 0.84205198,
      "num_input_tokens_seen": 39197505,
      "step": 1843,
      "time_per_iteration": 2.7117769718170166
    },
    {
      "auxiliary_loss_clip": 0.01087383,
      "auxiliary_loss_mlp": 0.01027701,
      "balance_loss_clip": 1.03880811,
      "balance_loss_mlp": 1.01763344,
      "epoch": 0.22172789033848372,
      "flos": 23764712186880.0,
      "grad_norm": 1.9475333238472639,
      "language_loss": 0.78044516,
      "learning_rate": 3.626779731344131e-06,
      "loss": 0.80159599,
      "num_input_tokens_seen": 39217295,
      "step": 1844,
      "time_per_iteration": 2.9324047565460205
    },
    {
      "auxiliary_loss_clip": 0.0111368,
      "auxiliary_loss_mlp": 0.00765095,
      "balance_loss_clip": 1.04406083,
      "balance_loss_mlp": 1.00040984,
      "epoch": 0.22184813322912283,
      "flos": 16982300361600.0,
      "grad_norm": 2.034462271500043,
      "language_loss": 0.85242879,
      "learning_rate": 3.6263264654031814e-06,
      "loss": 0.87121654,
      "num_input_tokens_seen": 39234195,
      "step": 1845,
      "time_per_iteration": 2.8267874717712402
    },
    {
      "auxiliary_loss_clip": 0.0109301,
      "auxiliary_loss_mlp": 0.01001715,
      "balance_loss_clip": 1.02742481,
      "balance_loss_mlp": 0.99995035,
      "epoch": 0.22196837611976192,
      "flos": 61823740314240.0,
      "grad_norm": 0.6960442327762848,
      "language_loss": 0.59161621,
      "learning_rate": 3.6258729527498008e-06,
      "loss": 0.61256349,
      "num_input_tokens_seen": 39295040,
      "step": 1846,
      "time_per_iteration": 3.2375173568725586
    },
    {
      "auxiliary_loss_clip": 0.01203834,
      "auxiliary_loss_mlp": 0.01034926,
      "balance_loss_clip": 1.06122088,
      "balance_loss_mlp": 1.02248049,
      "epoch": 0.222088619010401,
      "flos": 25558019625600.0,
      "grad_norm": 5.414245192082764,
      "language_loss": 0.64795375,
      "learning_rate": 3.6254191934527854e-06,
      "loss": 0.67034131,
      "num_input_tokens_seen": 39314395,
      "step": 1847,
      "time_per_iteration": 2.731729030609131
    },
    {
      "auxiliary_loss_clip": 0.01218919,
      "auxiliary_loss_mlp": 0.01036278,
      "balance_loss_clip": 1.06304014,
      "balance_loss_mlp": 1.02461374,
      "epoch": 0.2222088619010401,
      "flos": 19318612677120.0,
      "grad_norm": 3.118227787455605,
      "language_loss": 0.6543178,
      "learning_rate": 3.6249651875809715e-06,
      "loss": 0.67686981,
      "num_input_tokens_seen": 39334275,
      "step": 1848,
      "time_per_iteration": 2.687856674194336
    },
    {
      "auxiliary_loss_clip": 0.0116163,
      "auxiliary_loss_mlp": 0.01039122,
      "balance_loss_clip": 1.05822659,
      "balance_loss_mlp": 1.02788687,
      "epoch": 0.2223291047916792,
      "flos": 19099342103040.0,
      "grad_norm": 1.9741553272838472,
      "language_loss": 0.89182591,
      "learning_rate": 3.62451093520323e-06,
      "loss": 0.91383344,
      "num_input_tokens_seen": 39352180,
      "step": 1849,
      "time_per_iteration": 3.6626665592193604
    },
    {
      "auxiliary_loss_clip": 0.01149198,
      "auxiliary_loss_mlp": 0.01035618,
      "balance_loss_clip": 1.05062699,
      "balance_loss_mlp": 1.02460885,
      "epoch": 0.22244934768231828,
      "flos": 20850418126080.0,
      "grad_norm": 2.1440115530719313,
      "language_loss": 0.90771228,
      "learning_rate": 3.6240564363884714e-06,
      "loss": 0.92956048,
      "num_input_tokens_seen": 39372125,
      "step": 1850,
      "time_per_iteration": 2.7956862449645996
    },
    {
      "auxiliary_loss_clip": 0.01215469,
      "auxiliary_loss_mlp": 0.01041281,
      "balance_loss_clip": 1.05948997,
      "balance_loss_mlp": 1.02939034,
      "epoch": 0.2225695905729574,
      "flos": 15632921111040.0,
      "grad_norm": 1.7194598117074142,
      "language_loss": 0.70394766,
      "learning_rate": 3.623601691205643e-06,
      "loss": 0.72651517,
      "num_input_tokens_seen": 39391200,
      "step": 1851,
      "time_per_iteration": 2.6930181980133057
    },
    {
      "auxiliary_loss_clip": 0.01200843,
      "auxiliary_loss_mlp": 0.01033032,
      "balance_loss_clip": 1.06061649,
      "balance_loss_mlp": 1.02187383,
      "epoch": 0.22268983346359647,
      "flos": 25373582265600.0,
      "grad_norm": 2.359613301134343,
      "language_loss": 0.81292379,
      "learning_rate": 3.623146699723729e-06,
      "loss": 0.83526248,
      "num_input_tokens_seen": 39410660,
      "step": 1852,
      "time_per_iteration": 2.7014834880828857
    },
    {
      "auxiliary_loss_clip": 0.01135336,
      "auxiliary_loss_mlp": 0.01037279,
      "balance_loss_clip": 1.05481148,
      "balance_loss_mlp": 1.02465487,
      "epoch": 0.22281007635423555,
      "flos": 13261452359040.0,
      "grad_norm": 1.7487970894229283,
      "language_loss": 0.77300268,
      "learning_rate": 3.6226914620117507e-06,
      "loss": 0.79472882,
      "num_input_tokens_seen": 39429280,
      "step": 1853,
      "time_per_iteration": 3.9710824489593506
    },
    {
      "auxiliary_loss_clip": 0.01157783,
      "auxiliary_loss_mlp": 0.01031253,
      "balance_loss_clip": 1.05526948,
      "balance_loss_mlp": 1.02106082,
      "epoch": 0.22293031924487464,
      "flos": 15340536403200.0,
      "grad_norm": 2.047249207429962,
      "language_loss": 0.80410767,
      "learning_rate": 3.622235978138768e-06,
      "loss": 0.82599807,
      "num_input_tokens_seen": 39446905,
      "step": 1854,
      "time_per_iteration": 4.609593152999878
    },
    {
      "auxiliary_loss_clip": 0.01186755,
      "auxiliary_loss_mlp": 0.0104188,
      "balance_loss_clip": 1.05643439,
      "balance_loss_mlp": 1.03048897,
      "epoch": 0.22305056213551375,
      "flos": 22564649773440.0,
      "grad_norm": 1.8842768941018806,
      "language_loss": 0.81045377,
      "learning_rate": 3.621780248173877e-06,
      "loss": 0.83274007,
      "num_input_tokens_seen": 39465105,
      "step": 1855,
      "time_per_iteration": 2.7432572841644287
    },
    {
      "auxiliary_loss_clip": 0.01048182,
      "auxiliary_loss_mlp": 0.01001921,
      "balance_loss_clip": 1.02259684,
      "balance_loss_mlp": 1.00027585,
      "epoch": 0.22317080502615283,
      "flos": 64880419887360.0,
      "grad_norm": 0.825785720690032,
      "language_loss": 0.61064267,
      "learning_rate": 3.6213242721862125e-06,
      "loss": 0.63114369,
      "num_input_tokens_seen": 39523560,
      "step": 1856,
      "time_per_iteration": 3.356231689453125
    },
    {
      "auxiliary_loss_clip": 0.01094848,
      "auxiliary_loss_mlp": 0.01034267,
      "balance_loss_clip": 1.046188,
      "balance_loss_mlp": 1.02285266,
      "epoch": 0.2232910479167919,
      "flos": 25775997310080.0,
      "grad_norm": 1.601714720352016,
      "language_loss": 0.75135803,
      "learning_rate": 3.620868050244945e-06,
      "loss": 0.77264917,
      "num_input_tokens_seen": 39544040,
      "step": 1857,
      "time_per_iteration": 2.94618821144104
    },
    {
      "auxiliary_loss_clip": 0.01175118,
      "auxiliary_loss_mlp": 0.01038105,
      "balance_loss_clip": 1.05758452,
      "balance_loss_mlp": 1.02546906,
      "epoch": 0.22341129080743102,
      "flos": 23251799928960.0,
      "grad_norm": 1.8918941310198616,
      "language_loss": 0.77152175,
      "learning_rate": 3.6204115824192817e-06,
      "loss": 0.79365396,
      "num_input_tokens_seen": 39561515,
      "step": 1858,
      "time_per_iteration": 2.952453136444092
    },
    {
      "auxiliary_loss_clip": 0.01201935,
      "auxiliary_loss_mlp": 0.01041679,
      "balance_loss_clip": 1.05522633,
      "balance_loss_mlp": 1.02963269,
      "epoch": 0.2235315336980701,
      "flos": 21214552250880.0,
      "grad_norm": 4.568991882465262,
      "language_loss": 0.76464397,
      "learning_rate": 3.619954868778471e-06,
      "loss": 0.78708017,
      "num_input_tokens_seen": 39578210,
      "step": 1859,
      "time_per_iteration": 2.6959338188171387
    },
    {
      "auxiliary_loss_clip": 0.01165195,
      "auxiliary_loss_mlp": 0.01038687,
      "balance_loss_clip": 1.05567074,
      "balance_loss_mlp": 1.02714801,
      "epoch": 0.2236517765887092,
      "flos": 19901945548800.0,
      "grad_norm": 1.7832312316641061,
      "language_loss": 0.82944322,
      "learning_rate": 3.6194979093917944e-06,
      "loss": 0.85148203,
      "num_input_tokens_seen": 39597625,
      "step": 1860,
      "time_per_iteration": 2.6674234867095947
    },
    {
      "auxiliary_loss_clip": 0.0120089,
      "auxiliary_loss_mlp": 0.01038956,
      "balance_loss_clip": 1.0636332,
      "balance_loss_mlp": 1.0278095,
      "epoch": 0.22377201947934827,
      "flos": 23214847812480.0,
      "grad_norm": 2.11521426809911,
      "language_loss": 0.8681556,
      "learning_rate": 3.6190407043285724e-06,
      "loss": 0.89055407,
      "num_input_tokens_seen": 39615360,
      "step": 1861,
      "time_per_iteration": 2.687783718109131
    },
    {
      "auxiliary_loss_clip": 0.01215451,
      "auxiliary_loss_mlp": 0.01037206,
      "balance_loss_clip": 1.05977786,
      "balance_loss_mlp": 1.02549374,
      "epoch": 0.22389226236998738,
      "flos": 26794244056320.0,
      "grad_norm": 2.2807919895534794,
      "language_loss": 0.75810218,
      "learning_rate": 3.618583253658163e-06,
      "loss": 0.78062874,
      "num_input_tokens_seen": 39635460,
      "step": 1862,
      "time_per_iteration": 2.6365702152252197
    },
    {
      "auxiliary_loss_clip": 0.01107273,
      "auxiliary_loss_mlp": 0.01038828,
      "balance_loss_clip": 1.04805732,
      "balance_loss_mlp": 1.02738404,
      "epoch": 0.22401250526062647,
      "flos": 24170359455360.0,
      "grad_norm": 2.0357131968217135,
      "language_loss": 0.8656103,
      "learning_rate": 3.618125557449961e-06,
      "loss": 0.88707137,
      "num_input_tokens_seen": 39653515,
      "step": 1863,
      "time_per_iteration": 2.8311967849731445
    },
    {
      "auxiliary_loss_clip": 0.01168923,
      "auxiliary_loss_mlp": 0.00765385,
      "balance_loss_clip": 1.05298114,
      "balance_loss_mlp": 1.00079525,
      "epoch": 0.22413274815126555,
      "flos": 16759761649920.0,
      "grad_norm": 3.066492387463542,
      "language_loss": 0.8300671,
      "learning_rate": 3.6176676157733983e-06,
      "loss": 0.84941018,
      "num_input_tokens_seen": 39668525,
      "step": 1864,
      "time_per_iteration": 2.6989941596984863
    },
    {
      "auxiliary_loss_clip": 0.01114093,
      "auxiliary_loss_mlp": 0.01036683,
      "balance_loss_clip": 1.04953527,
      "balance_loss_mlp": 1.0250119,
      "epoch": 0.22425299104190466,
      "flos": 21360205900800.0,
      "grad_norm": 2.4577363393808715,
      "language_loss": 0.75512779,
      "learning_rate": 3.6172094286979443e-06,
      "loss": 0.77663553,
      "num_input_tokens_seen": 39685895,
      "step": 1865,
      "time_per_iteration": 2.8552558422088623
    },
    {
      "auxiliary_loss_clip": 0.01148015,
      "auxiliary_loss_mlp": 0.01032699,
      "balance_loss_clip": 1.04848671,
      "balance_loss_mlp": 1.02122557,
      "epoch": 0.22437323393254374,
      "flos": 32165547108480.0,
      "grad_norm": 3.875738373098573,
      "language_loss": 0.81537318,
      "learning_rate": 3.6167509962931064e-06,
      "loss": 0.83718038,
      "num_input_tokens_seen": 39711595,
      "step": 1866,
      "time_per_iteration": 2.9015724658966064
    },
    {
      "auxiliary_loss_clip": 0.01123932,
      "auxiliary_loss_mlp": 0.0104105,
      "balance_loss_clip": 1.05398393,
      "balance_loss_mlp": 1.02926016,
      "epoch": 0.22449347682318282,
      "flos": 18002809664640.0,
      "grad_norm": 2.5014545802153125,
      "language_loss": 0.76766741,
      "learning_rate": 3.6162923186284276e-06,
      "loss": 0.78931725,
      "num_input_tokens_seen": 39727555,
      "step": 1867,
      "time_per_iteration": 2.796851634979248
    },
    {
      "auxiliary_loss_clip": 0.0113305,
      "auxiliary_loss_mlp": 0.00765772,
      "balance_loss_clip": 1.05151379,
      "balance_loss_mlp": 1.00083232,
      "epoch": 0.2246137197138219,
      "flos": 18697286194560.0,
      "grad_norm": 2.6985733493548443,
      "language_loss": 0.85715079,
      "learning_rate": 3.6158333957734888e-06,
      "loss": 0.87613904,
      "num_input_tokens_seen": 39746145,
      "step": 1868,
      "time_per_iteration": 2.8202006816864014
    },
    {
      "auxiliary_loss_clip": 0.01099514,
      "auxiliary_loss_mlp": 0.01041764,
      "balance_loss_clip": 1.0409193,
      "balance_loss_mlp": 1.03111839,
      "epoch": 0.22473396260446102,
      "flos": 15590653781760.0,
      "grad_norm": 2.1534727433335386,
      "language_loss": 0.8268162,
      "learning_rate": 3.6153742277979088e-06,
      "loss": 0.84822893,
      "num_input_tokens_seen": 39763575,
      "step": 1869,
      "time_per_iteration": 2.8205811977386475
    },
    {
      "auxiliary_loss_clip": 0.01153398,
      "auxiliary_loss_mlp": 0.01032096,
      "balance_loss_clip": 1.0519141,
      "balance_loss_mlp": 1.02107549,
      "epoch": 0.2248542054951001,
      "flos": 14465501182080.0,
      "grad_norm": 2.2275502870950055,
      "language_loss": 0.77943516,
      "learning_rate": 3.6149148147713434e-06,
      "loss": 0.80129004,
      "num_input_tokens_seen": 39781810,
      "step": 1870,
      "time_per_iteration": 2.741093397140503
    },
    {
      "auxiliary_loss_clip": 0.01157736,
      "auxiliary_loss_mlp": 0.01041812,
      "balance_loss_clip": 1.05447435,
      "balance_loss_mlp": 1.03067219,
      "epoch": 0.22497444838573918,
      "flos": 19243882431360.0,
      "grad_norm": 2.0789510748198436,
      "language_loss": 0.86895341,
      "learning_rate": 3.614455156763484e-06,
      "loss": 0.89094889,
      "num_input_tokens_seen": 39800115,
      "step": 1871,
      "time_per_iteration": 2.7682714462280273
    },
    {
      "auxiliary_loss_clip": 0.01177028,
      "auxiliary_loss_mlp": 0.01039706,
      "balance_loss_clip": 1.05603898,
      "balance_loss_mlp": 1.02844632,
      "epoch": 0.2250946912763783,
      "flos": 16910299549440.0,
      "grad_norm": 2.09132282044132,
      "language_loss": 0.71228468,
      "learning_rate": 3.613995253844061e-06,
      "loss": 0.73445201,
      "num_input_tokens_seen": 39817795,
      "step": 1872,
      "time_per_iteration": 2.7451839447021484
    },
    {
      "auxiliary_loss_clip": 0.01148551,
      "auxiliary_loss_mlp": 0.01036539,
      "balance_loss_clip": 1.05542803,
      "balance_loss_mlp": 1.0250535,
      "epoch": 0.22521493416701738,
      "flos": 24681368292480.0,
      "grad_norm": 1.9830854426821147,
      "language_loss": 0.80511564,
      "learning_rate": 3.6135351060828414e-06,
      "loss": 0.82696658,
      "num_input_tokens_seen": 39838270,
      "step": 1873,
      "time_per_iteration": 2.784085273742676
    },
    {
      "auxiliary_loss_clip": 0.01185971,
      "auxiliary_loss_mlp": 0.01041841,
      "balance_loss_clip": 1.05663395,
      "balance_loss_mlp": 1.02834654,
      "epoch": 0.22533517705765646,
      "flos": 17821963664640.0,
      "grad_norm": 2.030708143521327,
      "language_loss": 0.68896687,
      "learning_rate": 3.6130747135496285e-06,
      "loss": 0.71124494,
      "num_input_tokens_seen": 39857270,
      "step": 1874,
      "time_per_iteration": 2.965451717376709
    },
    {
      "auxiliary_loss_clip": 0.012002,
      "auxiliary_loss_mlp": 0.00765896,
      "balance_loss_clip": 1.05931282,
      "balance_loss_mlp": 1.00069797,
      "epoch": 0.22545541994829554,
      "flos": 33691390899840.0,
      "grad_norm": 20.257734320058255,
      "language_loss": 0.65949422,
      "learning_rate": 3.6126140763142646e-06,
      "loss": 0.67915523,
      "num_input_tokens_seen": 39882300,
      "step": 1875,
      "time_per_iteration": 2.7869207859039307
    },
    {
      "auxiliary_loss_clip": 0.01198889,
      "auxiliary_loss_mlp": 0.01042637,
      "balance_loss_clip": 1.0587337,
      "balance_loss_mlp": 1.03158677,
      "epoch": 0.22557566283893465,
      "flos": 19171594310400.0,
      "grad_norm": 2.666082098792006,
      "language_loss": 0.85868323,
      "learning_rate": 3.6121531944466275e-06,
      "loss": 0.88109851,
      "num_input_tokens_seen": 39899625,
      "step": 1876,
      "time_per_iteration": 3.5852601528167725
    },
    {
      "auxiliary_loss_clip": 0.01155711,
      "auxiliary_loss_mlp": 0.00765388,
      "balance_loss_clip": 1.04975557,
      "balance_loss_mlp": 1.00083792,
      "epoch": 0.22569590572957374,
      "flos": 20773281669120.0,
      "grad_norm": 2.2372235242858167,
      "language_loss": 0.78463149,
      "learning_rate": 3.611692068016633e-06,
      "loss": 0.80384243,
      "num_input_tokens_seen": 39915955,
      "step": 1877,
      "time_per_iteration": 2.7052929401397705
    },
    {
      "auxiliary_loss_clip": 0.01192385,
      "auxiliary_loss_mlp": 0.01040264,
      "balance_loss_clip": 1.05919886,
      "balance_loss_mlp": 1.02821839,
      "epoch": 0.22581614862021282,
      "flos": 18442715529600.0,
      "grad_norm": 2.8290254017047363,
      "language_loss": 0.74645644,
      "learning_rate": 3.611230697094233e-06,
      "loss": 0.76878291,
      "num_input_tokens_seen": 39932655,
      "step": 1878,
      "time_per_iteration": 2.715045690536499
    },
    {
      "auxiliary_loss_clip": 0.01167235,
      "auxiliary_loss_mlp": 0.01039163,
      "balance_loss_clip": 1.05771863,
      "balance_loss_mlp": 1.02801728,
      "epoch": 0.22593639151085193,
      "flos": 20048389297920.0,
      "grad_norm": 1.921790777267898,
      "language_loss": 0.87540424,
      "learning_rate": 3.6107690817494173e-06,
      "loss": 0.89746821,
      "num_input_tokens_seen": 39952875,
      "step": 1879,
      "time_per_iteration": 3.7492334842681885
    },
    {
      "auxiliary_loss_clip": 0.01190811,
      "auxiliary_loss_mlp": 0.01038268,
      "balance_loss_clip": 1.05675292,
      "balance_loss_mlp": 1.02692509,
      "epoch": 0.226056634401491,
      "flos": 13115116350720.0,
      "grad_norm": 3.452493296135719,
      "language_loss": 0.70389992,
      "learning_rate": 3.6103072220522117e-06,
      "loss": 0.72619069,
      "num_input_tokens_seen": 39968405,
      "step": 1880,
      "time_per_iteration": 4.47825288772583
    },
    {
      "auxiliary_loss_clip": 0.0118304,
      "auxiliary_loss_mlp": 0.01031123,
      "balance_loss_clip": 1.05820441,
      "balance_loss_mlp": 1.02053177,
      "epoch": 0.2261768772921301,
      "flos": 18988378012800.0,
      "grad_norm": 2.20671823784061,
      "language_loss": 0.91804183,
      "learning_rate": 3.609845118072682e-06,
      "loss": 0.94018346,
      "num_input_tokens_seen": 39987075,
      "step": 1881,
      "time_per_iteration": 2.686361074447632
    },
    {
      "auxiliary_loss_clip": 0.01188845,
      "auxiliary_loss_mlp": 0.00766196,
      "balance_loss_clip": 1.06048751,
      "balance_loss_mlp": 1.00076032,
      "epoch": 0.2262971201827692,
      "flos": 19974054101760.0,
      "grad_norm": 1.9121272439524004,
      "language_loss": 0.79955935,
      "learning_rate": 3.6093827698809276e-06,
      "loss": 0.8191098,
      "num_input_tokens_seen": 40006175,
      "step": 1882,
      "time_per_iteration": 2.68677020072937
    },
    {
      "auxiliary_loss_clip": 0.01169576,
      "auxiliary_loss_mlp": 0.01038406,
      "balance_loss_clip": 1.05549121,
      "balance_loss_mlp": 1.02782655,
      "epoch": 0.2264173630734083,
      "flos": 16654543735680.0,
      "grad_norm": 2.802392912579538,
      "language_loss": 0.85013878,
      "learning_rate": 3.6089201775470864e-06,
      "loss": 0.87221861,
      "num_input_tokens_seen": 40021630,
      "step": 1883,
      "time_per_iteration": 2.7371175289154053
    },
    {
      "auxiliary_loss_clip": 0.01193737,
      "auxiliary_loss_mlp": 0.0104357,
      "balance_loss_clip": 1.05541658,
      "balance_loss_mlp": 1.03231692,
      "epoch": 0.22653760596404737,
      "flos": 24389809597440.0,
      "grad_norm": 1.5256265710061914,
      "language_loss": 0.77471584,
      "learning_rate": 3.6084573411413334e-06,
      "loss": 0.79708892,
      "num_input_tokens_seen": 40041025,
      "step": 1884,
      "time_per_iteration": 2.6542043685913086
    },
    {
      "auxiliary_loss_clip": 0.01203722,
      "auxiliary_loss_mlp": 0.01039226,
      "balance_loss_clip": 1.0623244,
      "balance_loss_mlp": 1.02732301,
      "epoch": 0.22665784885468646,
      "flos": 18332541538560.0,
      "grad_norm": 2.1932353179052906,
      "language_loss": 0.81141317,
      "learning_rate": 3.607994260733881e-06,
      "loss": 0.83384264,
      "num_input_tokens_seen": 40060265,
      "step": 1885,
      "time_per_iteration": 2.623683452606201
    },
    {
      "auxiliary_loss_clip": 0.01149634,
      "auxiliary_loss_mlp": 0.01039393,
      "balance_loss_clip": 1.05202651,
      "balance_loss_mlp": 1.02855694,
      "epoch": 0.22677809174532557,
      "flos": 24058102475520.0,
      "grad_norm": 1.6723338244857977,
      "language_loss": 0.74797833,
      "learning_rate": 3.6075309363949776e-06,
      "loss": 0.76986855,
      "num_input_tokens_seen": 40079435,
      "step": 1886,
      "time_per_iteration": 2.7165582180023193
    },
    {
      "auxiliary_loss_clip": 0.01116934,
      "auxiliary_loss_mlp": 0.00765797,
      "balance_loss_clip": 1.04509318,
      "balance_loss_mlp": 1.00068617,
      "epoch": 0.22689833463596465,
      "flos": 20374242503040.0,
      "grad_norm": 1.844736973383848,
      "language_loss": 0.80641633,
      "learning_rate": 3.6070673681949094e-06,
      "loss": 0.82524365,
      "num_input_tokens_seen": 40097800,
      "step": 1887,
      "time_per_iteration": 2.801823854446411
    },
    {
      "auxiliary_loss_clip": 0.01081542,
      "auxiliary_loss_mlp": 0.01033382,
      "balance_loss_clip": 1.04630005,
      "balance_loss_mlp": 1.02218795,
      "epoch": 0.22701857752660373,
      "flos": 30120398438400.0,
      "grad_norm": 1.7160360479232006,
      "language_loss": 0.81260121,
      "learning_rate": 3.606603556203999e-06,
      "loss": 0.83375043,
      "num_input_tokens_seen": 40122745,
      "step": 1888,
      "time_per_iteration": 3.31288743019104
    },
    {
      "auxiliary_loss_clip": 0.01167996,
      "auxiliary_loss_mlp": 0.0104809,
      "balance_loss_clip": 1.05478835,
      "balance_loss_mlp": 1.03742123,
      "epoch": 0.22713882041724284,
      "flos": 22492182084480.0,
      "grad_norm": 2.0720814082017407,
      "language_loss": 0.8319478,
      "learning_rate": 3.6061395004926066e-06,
      "loss": 0.85410863,
      "num_input_tokens_seen": 40141680,
      "step": 1889,
      "time_per_iteration": 3.104095935821533
    },
    {
      "auxiliary_loss_clip": 0.01153999,
      "auxiliary_loss_mlp": 0.01040938,
      "balance_loss_clip": 1.05178368,
      "balance_loss_mlp": 1.02907062,
      "epoch": 0.22725906330788193,
      "flos": 20521548178560.0,
      "grad_norm": 4.7465792045839645,
      "language_loss": 0.84843332,
      "learning_rate": 3.605675201131129e-06,
      "loss": 0.87038267,
      "num_input_tokens_seen": 40160140,
      "step": 1890,
      "time_per_iteration": 2.7495293617248535
    },
    {
      "auxiliary_loss_clip": 0.01130936,
      "auxiliary_loss_mlp": 0.01038672,
      "balance_loss_clip": 1.04898381,
      "balance_loss_mlp": 1.02747869,
      "epoch": 0.227379306198521,
      "flos": 18989922297600.0,
      "grad_norm": 2.2825929219179635,
      "language_loss": 0.79807389,
      "learning_rate": 3.60521065819e-06,
      "loss": 0.81976998,
      "num_input_tokens_seen": 40177450,
      "step": 1891,
      "time_per_iteration": 2.72760272026062
    },
    {
      "auxiliary_loss_clip": 0.01163413,
      "auxiliary_loss_mlp": 0.01034559,
      "balance_loss_clip": 1.05302751,
      "balance_loss_mlp": 1.0239259,
      "epoch": 0.2274995490891601,
      "flos": 21798351999360.0,
      "grad_norm": 1.847355834009663,
      "language_loss": 0.87675893,
      "learning_rate": 3.60474587173969e-06,
      "loss": 0.89873862,
      "num_input_tokens_seen": 40195935,
      "step": 1892,
      "time_per_iteration": 2.712831735610962
    },
    {
      "auxiliary_loss_clip": 0.01111798,
      "auxiliary_loss_mlp": 0.01035596,
      "balance_loss_clip": 1.05182409,
      "balance_loss_mlp": 1.02473021,
      "epoch": 0.2276197919797992,
      "flos": 19058654972160.0,
      "grad_norm": 2.1311948774067875,
      "language_loss": 0.84099197,
      "learning_rate": 3.6042808418507084e-06,
      "loss": 0.86246598,
      "num_input_tokens_seen": 40213620,
      "step": 1893,
      "time_per_iteration": 3.1579596996307373
    },
    {
      "auxiliary_loss_clip": 0.01155637,
      "auxiliary_loss_mlp": 0.01035002,
      "balance_loss_clip": 1.05489945,
      "balance_loss_mlp": 1.02419007,
      "epoch": 0.22774003487043828,
      "flos": 18806777827200.0,
      "grad_norm": 2.157173194078489,
      "language_loss": 0.77085602,
      "learning_rate": 3.6038155685935976e-06,
      "loss": 0.7927624,
      "num_input_tokens_seen": 40230190,
      "step": 1894,
      "time_per_iteration": 2.937166213989258
    },
    {
      "auxiliary_loss_clip": 0.01131172,
      "auxiliary_loss_mlp": 0.01042352,
      "balance_loss_clip": 1.05132818,
      "balance_loss_mlp": 1.03114653,
      "epoch": 0.22786027776107737,
      "flos": 23002544476800.0,
      "grad_norm": 2.7146380758087187,
      "language_loss": 0.70208484,
      "learning_rate": 3.6033500520389404e-06,
      "loss": 0.72382009,
      "num_input_tokens_seen": 40246860,
      "step": 1895,
      "time_per_iteration": 2.7828621864318848
    },
    {
      "auxiliary_loss_clip": 0.01084298,
      "auxiliary_loss_mlp": 0.01002507,
      "balance_loss_clip": 1.02914596,
      "balance_loss_mlp": 1.00071895,
      "epoch": 0.22798052065171648,
      "flos": 66706872600960.0,
      "grad_norm": 0.7903189321383433,
      "language_loss": 0.64811838,
      "learning_rate": 3.6028842922573553e-06,
      "loss": 0.66898644,
      "num_input_tokens_seen": 40311005,
      "step": 1896,
      "time_per_iteration": 3.3369147777557373
    },
    {
      "auxiliary_loss_clip": 0.01088011,
      "auxiliary_loss_mlp": 0.00758109,
      "balance_loss_clip": 1.0264014,
      "balance_loss_mlp": 1.00086558,
      "epoch": 0.22810076354235556,
      "flos": 62080896758400.0,
      "grad_norm": 0.8745431024852642,
      "language_loss": 0.62912023,
      "learning_rate": 3.602418289319497e-06,
      "loss": 0.64758146,
      "num_input_tokens_seen": 40369560,
      "step": 1897,
      "time_per_iteration": 3.2105209827423096
    },
    {
      "auxiliary_loss_clip": 0.01163808,
      "auxiliary_loss_mlp": 0.01038135,
      "balance_loss_clip": 1.05285239,
      "balance_loss_mlp": 1.02647054,
      "epoch": 0.22822100643299464,
      "flos": 23876358635520.0,
      "grad_norm": 1.785789714558695,
      "language_loss": 0.73425114,
      "learning_rate": 3.601952043296059e-06,
      "loss": 0.75627053,
      "num_input_tokens_seen": 40389555,
      "step": 1898,
      "time_per_iteration": 2.8078560829162598
    },
    {
      "auxiliary_loss_clip": 0.0111455,
      "auxiliary_loss_mlp": 0.01038898,
      "balance_loss_clip": 1.04903841,
      "balance_loss_mlp": 1.02706623,
      "epoch": 0.22834124932363373,
      "flos": 20991331180800.0,
      "grad_norm": 2.0712995566729706,
      "language_loss": 0.80723548,
      "learning_rate": 3.6014855542577696e-06,
      "loss": 0.82876992,
      "num_input_tokens_seen": 40406765,
      "step": 1899,
      "time_per_iteration": 2.8369626998901367
    },
    {
      "auxiliary_loss_clip": 0.01163352,
      "auxiliary_loss_mlp": 0.01039556,
      "balance_loss_clip": 1.05858707,
      "balance_loss_mlp": 1.02802873,
      "epoch": 0.22846149221427284,
      "flos": 24901572620160.0,
      "grad_norm": 1.7196449402624208,
      "language_loss": 0.84308094,
      "learning_rate": 3.6010188222753943e-06,
      "loss": 0.86511004,
      "num_input_tokens_seen": 40427535,
      "step": 1900,
      "time_per_iteration": 2.7713589668273926
    },
    {
      "auxiliary_loss_clip": 0.01081795,
      "auxiliary_loss_mlp": 0.01002366,
      "balance_loss_clip": 1.02428854,
      "balance_loss_mlp": 1.00070882,
      "epoch": 0.22858173510491192,
      "flos": 56132294319360.0,
      "grad_norm": 0.9031811892166622,
      "language_loss": 0.64181757,
      "learning_rate": 3.6005518474197372e-06,
      "loss": 0.66265917,
      "num_input_tokens_seen": 40479580,
      "step": 1901,
      "time_per_iteration": 3.134834051132202
    },
    {
      "auxiliary_loss_clip": 0.01096865,
      "auxiliary_loss_mlp": 0.01035269,
      "balance_loss_clip": 1.04314578,
      "balance_loss_mlp": 1.02338934,
      "epoch": 0.228701977995551,
      "flos": 24170826332160.0,
      "grad_norm": 2.1932215216369886,
      "language_loss": 0.78348774,
      "learning_rate": 3.6000846297616373e-06,
      "loss": 0.80480909,
      "num_input_tokens_seen": 40497880,
      "step": 1902,
      "time_per_iteration": 4.19816517829895
    },
    {
      "auxiliary_loss_clip": 0.01174449,
      "auxiliary_loss_mlp": 0.01043742,
      "balance_loss_clip": 1.05638623,
      "balance_loss_mlp": 1.03128469,
      "epoch": 0.22882222088619011,
      "flos": 21387892308480.0,
      "grad_norm": 2.2514978775130516,
      "language_loss": 0.72199047,
      "learning_rate": 3.5996171693719717e-06,
      "loss": 0.74417239,
      "num_input_tokens_seen": 40513975,
      "step": 1903,
      "time_per_iteration": 3.2594919204711914
    },
    {
      "auxiliary_loss_clip": 0.01073089,
      "auxiliary_loss_mlp": 0.0100606,
      "balance_loss_clip": 1.02117002,
      "balance_loss_mlp": 1.00435531,
      "epoch": 0.2289424637768292,
      "flos": 64589615377920.0,
      "grad_norm": 0.8297547950660292,
      "language_loss": 0.6477465,
      "learning_rate": 3.5991494663216528e-06,
      "loss": 0.66853797,
      "num_input_tokens_seen": 40576960,
      "step": 1904,
      "time_per_iteration": 4.592101812362671
    },
    {
      "auxiliary_loss_clip": 0.01163863,
      "auxiliary_loss_mlp": 0.0103864,
      "balance_loss_clip": 1.05242562,
      "balance_loss_mlp": 1.02723193,
      "epoch": 0.22906270666746828,
      "flos": 22163419877760.0,
      "grad_norm": 2.1777462235118814,
      "language_loss": 0.87436414,
      "learning_rate": 3.5986815206816314e-06,
      "loss": 0.89638925,
      "num_input_tokens_seen": 40595780,
      "step": 1905,
      "time_per_iteration": 5.21498441696167
    },
    {
      "auxiliary_loss_clip": 0.01150006,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.05571556,
      "balance_loss_mlp": 1.0227654,
      "epoch": 0.2291829495581074,
      "flos": 25772334122880.0,
      "grad_norm": 1.8715640171071937,
      "language_loss": 0.74834269,
      "learning_rate": 3.598213332522895e-06,
      "loss": 0.77017772,
      "num_input_tokens_seen": 40615810,
      "step": 1906,
      "time_per_iteration": 2.8583552837371826
    },
    {
      "auxiliary_loss_clip": 0.01139115,
      "auxiliary_loss_mlp": 0.01040911,
      "balance_loss_clip": 1.0474087,
      "balance_loss_mlp": 1.02997351,
      "epoch": 0.22930319244874647,
      "flos": 31172760126720.0,
      "grad_norm": 2.0067441955140075,
      "language_loss": 0.7790451,
      "learning_rate": 3.597744901916466e-06,
      "loss": 0.80084538,
      "num_input_tokens_seen": 40637095,
      "step": 1907,
      "time_per_iteration": 2.83247971534729
    },
    {
      "auxiliary_loss_clip": 0.01151753,
      "auxiliary_loss_mlp": 0.01037853,
      "balance_loss_clip": 1.05004621,
      "balance_loss_mlp": 1.02638531,
      "epoch": 0.22942343533938556,
      "flos": 23254098399360.0,
      "grad_norm": 2.0536672216238436,
      "language_loss": 0.76950455,
      "learning_rate": 3.5972762289334058e-06,
      "loss": 0.79140061,
      "num_input_tokens_seen": 40656725,
      "step": 1908,
      "time_per_iteration": 2.7729945182800293
    },
    {
      "auxiliary_loss_clip": 0.01137005,
      "auxiliary_loss_mlp": 0.01035576,
      "balance_loss_clip": 1.04579997,
      "balance_loss_mlp": 1.02394736,
      "epoch": 0.22954367823002464,
      "flos": 14610903436800.0,
      "grad_norm": 2.2125107917563716,
      "language_loss": 0.85150659,
      "learning_rate": 3.5968073136448116e-06,
      "loss": 0.87323236,
      "num_input_tokens_seen": 40674745,
      "step": 1909,
      "time_per_iteration": 2.7684874534606934
    },
    {
      "auxiliary_loss_clip": 0.01151678,
      "auxiliary_loss_mlp": 0.01036781,
      "balance_loss_clip": 1.0483377,
      "balance_loss_mlp": 1.02463901,
      "epoch": 0.22966392112066375,
      "flos": 16763604405120.0,
      "grad_norm": 1.9736158703676356,
      "language_loss": 0.91333449,
      "learning_rate": 3.596338156121818e-06,
      "loss": 0.93521905,
      "num_input_tokens_seen": 40693630,
      "step": 1910,
      "time_per_iteration": 2.721193313598633
    },
    {
      "auxiliary_loss_clip": 0.01068282,
      "auxiliary_loss_mlp": 0.01003759,
      "balance_loss_clip": 1.02861333,
      "balance_loss_mlp": 1.00206614,
      "epoch": 0.22978416401130283,
      "flos": 67474247783040.0,
      "grad_norm": 0.7438881163104339,
      "language_loss": 0.59320831,
      "learning_rate": 3.595868756435595e-06,
      "loss": 0.61392868,
      "num_input_tokens_seen": 40761310,
      "step": 1911,
      "time_per_iteration": 3.456148624420166
    },
    {
      "auxiliary_loss_clip": 0.01200013,
      "auxiliary_loss_mlp": 0.0104344,
      "balance_loss_clip": 1.06109524,
      "balance_loss_mlp": 1.03218651,
      "epoch": 0.22990440690194192,
      "flos": 19865137086720.0,
      "grad_norm": 2.3632872158089717,
      "language_loss": 0.80319929,
      "learning_rate": 3.5953991146573504e-06,
      "loss": 0.82563376,
      "num_input_tokens_seen": 40779955,
      "step": 1912,
      "time_per_iteration": 2.7166035175323486
    },
    {
      "auxiliary_loss_clip": 0.01154304,
      "auxiliary_loss_mlp": 0.0103953,
      "balance_loss_clip": 1.05229139,
      "balance_loss_mlp": 1.02760923,
      "epoch": 0.23002464979258103,
      "flos": 13289246507520.0,
      "grad_norm": 2.6212082698989847,
      "language_loss": 0.83471423,
      "learning_rate": 3.5949292308583294e-06,
      "loss": 0.85665256,
      "num_input_tokens_seen": 40793200,
      "step": 1913,
      "time_per_iteration": 2.687222480773926
    },
    {
      "auxiliary_loss_clip": 0.01197216,
      "auxiliary_loss_mlp": 0.01043246,
      "balance_loss_clip": 1.06118774,
      "balance_loss_mlp": 1.03099751,
      "epoch": 0.2301448926832201,
      "flos": 22163779013760.0,
      "grad_norm": 1.972451877949099,
      "language_loss": 0.80575716,
      "learning_rate": 3.594459105109811e-06,
      "loss": 0.82816172,
      "num_input_tokens_seen": 40812380,
      "step": 1914,
      "time_per_iteration": 2.7499237060546875
    },
    {
      "auxiliary_loss_clip": 0.01152788,
      "auxiliary_loss_mlp": 0.01036123,
      "balance_loss_clip": 1.05723047,
      "balance_loss_mlp": 1.02514958,
      "epoch": 0.2302651355738592,
      "flos": 20704477167360.0,
      "grad_norm": 1.7932324781740896,
      "language_loss": 0.81348658,
      "learning_rate": 3.593988737483115e-06,
      "loss": 0.83537567,
      "num_input_tokens_seen": 40832320,
      "step": 1915,
      "time_per_iteration": 2.800713062286377
    },
    {
      "auxiliary_loss_clip": 0.01200184,
      "auxiliary_loss_mlp": 0.01032522,
      "balance_loss_clip": 1.05852365,
      "balance_loss_mlp": 1.02172148,
      "epoch": 0.23038537846449827,
      "flos": 18588943797120.0,
      "grad_norm": 1.9551471120090382,
      "language_loss": 0.78147644,
      "learning_rate": 3.5935181280495947e-06,
      "loss": 0.8038035,
      "num_input_tokens_seen": 40850900,
      "step": 1916,
      "time_per_iteration": 2.7005906105041504
    },
    {
      "auxiliary_loss_clip": 0.01086064,
      "auxiliary_loss_mlp": 0.01006791,
      "balance_loss_clip": 1.04056883,
      "balance_loss_mlp": 1.00483632,
      "epoch": 0.23050562135513739,
      "flos": 64224260190720.0,
      "grad_norm": 0.8081329297031766,
      "language_loss": 0.54277235,
      "learning_rate": 3.5930472768806412e-06,
      "loss": 0.56370091,
      "num_input_tokens_seen": 40909570,
      "step": 1917,
      "time_per_iteration": 3.2751824855804443
    },
    {
      "auxiliary_loss_clip": 0.01161746,
      "auxiliary_loss_mlp": 0.01042475,
      "balance_loss_clip": 1.05560374,
      "balance_loss_mlp": 1.03057241,
      "epoch": 0.23062586424577647,
      "flos": 17313396952320.0,
      "grad_norm": 1.8830408108556744,
      "language_loss": 0.77003413,
      "learning_rate": 3.5925761840476826e-06,
      "loss": 0.79207635,
      "num_input_tokens_seen": 40928180,
      "step": 1918,
      "time_per_iteration": 2.8120715618133545
    },
    {
      "auxiliary_loss_clip": 0.01139043,
      "auxiliary_loss_mlp": 0.01045103,
      "balance_loss_clip": 1.04972148,
      "balance_loss_mlp": 1.03332555,
      "epoch": 0.23074610713641555,
      "flos": 27855979194240.0,
      "grad_norm": 2.1201742390102436,
      "language_loss": 0.81493676,
      "learning_rate": 3.592104849622183e-06,
      "loss": 0.83677828,
      "num_input_tokens_seen": 40950435,
      "step": 1919,
      "time_per_iteration": 2.866150379180908
    },
    {
      "auxiliary_loss_clip": 0.0111725,
      "auxiliary_loss_mlp": 0.00765111,
      "balance_loss_clip": 1.04503059,
      "balance_loss_mlp": 1.00046825,
      "epoch": 0.23086635002705466,
      "flos": 28841798937600.0,
      "grad_norm": 1.51158725755505,
      "language_loss": 0.73398465,
      "learning_rate": 3.591633273675644e-06,
      "loss": 0.75280821,
      "num_input_tokens_seen": 40972670,
      "step": 1920,
      "time_per_iteration": 3.0855298042297363
    },
    {
      "auxiliary_loss_clip": 0.01082402,
      "auxiliary_loss_mlp": 0.01001638,
      "balance_loss_clip": 1.01674294,
      "balance_loss_mlp": 1.00002897,
      "epoch": 0.23098659291769374,
      "flos": 62923681566720.0,
      "grad_norm": 0.9063560950583892,
      "language_loss": 0.58157122,
      "learning_rate": 3.591161456279602e-06,
      "loss": 0.60241163,
      "num_input_tokens_seen": 41018215,
      "step": 1921,
      "time_per_iteration": 3.124210834503174
    },
    {
      "auxiliary_loss_clip": 0.01133157,
      "auxiliary_loss_mlp": 0.01037634,
      "balance_loss_clip": 1.05084741,
      "balance_loss_mlp": 1.02694714,
      "epoch": 0.23110683580833283,
      "flos": 23476816679040.0,
      "grad_norm": 1.5348295120844233,
      "language_loss": 0.80312359,
      "learning_rate": 3.590689397505633e-06,
      "loss": 0.82483149,
      "num_input_tokens_seen": 41039125,
      "step": 1922,
      "time_per_iteration": 2.854947566986084
    },
    {
      "auxiliary_loss_clip": 0.01155517,
      "auxiliary_loss_mlp": 0.01039483,
      "balance_loss_clip": 1.05126441,
      "balance_loss_mlp": 1.02865243,
      "epoch": 0.2312270786989719,
      "flos": 27271066124160.0,
      "grad_norm": 1.6457455402632037,
      "language_loss": 0.86829162,
      "learning_rate": 3.590217097425347e-06,
      "loss": 0.89024162,
      "num_input_tokens_seen": 41059025,
      "step": 1923,
      "time_per_iteration": 2.879423141479492
    },
    {
      "auxiliary_loss_clip": 0.0113397,
      "auxiliary_loss_mlp": 0.00765608,
      "balance_loss_clip": 1.04431295,
      "balance_loss_mlp": 1.00045717,
      "epoch": 0.23134732158961102,
      "flos": 13261344618240.0,
      "grad_norm": 2.358949344843613,
      "language_loss": 0.71418917,
      "learning_rate": 3.589744556110391e-06,
      "loss": 0.73318493,
      "num_input_tokens_seen": 41077015,
      "step": 1924,
      "time_per_iteration": 2.77059268951416
    },
    {
      "auxiliary_loss_clip": 0.01183436,
      "auxiliary_loss_mlp": 0.01037583,
      "balance_loss_clip": 1.05813479,
      "balance_loss_mlp": 1.0263294,
      "epoch": 0.2314675644802501,
      "flos": 36977648250240.0,
      "grad_norm": 1.6695453649526237,
      "language_loss": 0.8437717,
      "learning_rate": 3.58927177363245e-06,
      "loss": 0.86598188,
      "num_input_tokens_seen": 41099840,
      "step": 1925,
      "time_per_iteration": 2.831885814666748
    },
    {
      "auxiliary_loss_clip": 0.0117612,
      "auxiliary_loss_mlp": 0.01035365,
      "balance_loss_clip": 1.05424142,
      "balance_loss_mlp": 1.02306819,
      "epoch": 0.2315878073708892,
      "flos": 23842207779840.0,
      "grad_norm": 2.0619351619430013,
      "language_loss": 0.72745407,
      "learning_rate": 3.5887987500632447e-06,
      "loss": 0.74956888,
      "num_input_tokens_seen": 41117845,
      "step": 1926,
      "time_per_iteration": 2.776681423187256
    },
    {
      "auxiliary_loss_clip": 0.01127787,
      "auxiliary_loss_mlp": 0.01031141,
      "balance_loss_clip": 1.05219102,
      "balance_loss_mlp": 1.02104986,
      "epoch": 0.2317080502615283,
      "flos": 23039424766080.0,
      "grad_norm": 2.267831443150433,
      "language_loss": 0.83878034,
      "learning_rate": 3.5883254854745325e-06,
      "loss": 0.86036962,
      "num_input_tokens_seen": 41136235,
      "step": 1927,
      "time_per_iteration": 2.8959507942199707
    },
    {
      "auxiliary_loss_clip": 0.01124862,
      "auxiliary_loss_mlp": 0.0076528,
      "balance_loss_clip": 1.05325913,
      "balance_loss_mlp": 1.00043547,
      "epoch": 0.23182829315216738,
      "flos": 11254656435840.0,
      "grad_norm": 2.8906038017959497,
      "language_loss": 0.75072205,
      "learning_rate": 3.587851979938107e-06,
      "loss": 0.76962352,
      "num_input_tokens_seen": 41153125,
      "step": 1928,
      "time_per_iteration": 3.719362735748291
    },
    {
      "auxiliary_loss_clip": 0.0119806,
      "auxiliary_loss_mlp": 0.01034989,
      "balance_loss_clip": 1.06138778,
      "balance_loss_mlp": 1.02451038,
      "epoch": 0.23194853604280646,
      "flos": 19828939155840.0,
      "grad_norm": 2.0760553561084563,
      "language_loss": 0.77156746,
      "learning_rate": 3.5873782335257985e-06,
      "loss": 0.79389793,
      "num_input_tokens_seen": 41171290,
      "step": 1929,
      "time_per_iteration": 3.6382346153259277
    },
    {
      "auxiliary_loss_clip": 0.01163122,
      "auxiliary_loss_mlp": 0.00765163,
      "balance_loss_clip": 1.05878305,
      "balance_loss_mlp": 1.00044227,
      "epoch": 0.23206877893344555,
      "flos": 15305020830720.0,
      "grad_norm": 2.0125776179750177,
      "language_loss": 0.78421569,
      "learning_rate": 3.5869042463094744e-06,
      "loss": 0.80349857,
      "num_input_tokens_seen": 41189005,
      "step": 1930,
      "time_per_iteration": 3.7186410427093506
    },
    {
      "auxiliary_loss_clip": 0.01139104,
      "auxiliary_loss_mlp": 0.0104323,
      "balance_loss_clip": 1.04633069,
      "balance_loss_mlp": 1.03152406,
      "epoch": 0.23218902182408466,
      "flos": 22711488572160.0,
      "grad_norm": 2.9133560509605427,
      "language_loss": 0.77379906,
      "learning_rate": 3.586430018361038e-06,
      "loss": 0.79562241,
      "num_input_tokens_seen": 41208775,
      "step": 1931,
      "time_per_iteration": 2.8433401584625244
    },
    {
      "auxiliary_loss_clip": 0.01201435,
      "auxiliary_loss_mlp": 0.01042387,
      "balance_loss_clip": 1.05883384,
      "balance_loss_mlp": 1.03084755,
      "epoch": 0.23230926471472374,
      "flos": 22710734386560.0,
      "grad_norm": 2.3865647455270436,
      "language_loss": 0.76445377,
      "learning_rate": 3.5859555497524283e-06,
      "loss": 0.78689194,
      "num_input_tokens_seen": 41226010,
      "step": 1932,
      "time_per_iteration": 2.7363061904907227
    },
    {
      "auxiliary_loss_clip": 0.01187473,
      "auxiliary_loss_mlp": 0.01036429,
      "balance_loss_clip": 1.06103778,
      "balance_loss_mlp": 1.02634358,
      "epoch": 0.23242950760536282,
      "flos": 20375499479040.0,
      "grad_norm": 1.8708660027366057,
      "language_loss": 0.9235388,
      "learning_rate": 3.5854808405556237e-06,
      "loss": 0.94577783,
      "num_input_tokens_seen": 41245245,
      "step": 1933,
      "time_per_iteration": 2.6802828311920166
    },
    {
      "auxiliary_loss_clip": 0.0115577,
      "auxiliary_loss_mlp": 0.00764931,
      "balance_loss_clip": 1.0531342,
      "balance_loss_mlp": 1.00052476,
      "epoch": 0.23254975049600193,
      "flos": 16908324301440.0,
      "grad_norm": 2.6482421072539393,
      "language_loss": 0.75145501,
      "learning_rate": 3.5850058908426355e-06,
      "loss": 0.77066201,
      "num_input_tokens_seen": 41263795,
      "step": 1934,
      "time_per_iteration": 2.7772469520568848
    },
    {
      "auxiliary_loss_clip": 0.0118678,
      "auxiliary_loss_mlp": 0.01039792,
      "balance_loss_clip": 1.06038916,
      "balance_loss_mlp": 1.02923012,
      "epoch": 0.23266999338664102,
      "flos": 23294821443840.0,
      "grad_norm": 1.9434990635721032,
      "language_loss": 0.8555814,
      "learning_rate": 3.584530700685514e-06,
      "loss": 0.87784714,
      "num_input_tokens_seen": 41284055,
      "step": 1935,
      "time_per_iteration": 2.7373907566070557
    },
    {
      "auxiliary_loss_clip": 0.0118868,
      "auxiliary_loss_mlp": 0.01033968,
      "balance_loss_clip": 1.05745387,
      "balance_loss_mlp": 1.02336383,
      "epoch": 0.2327902362772801,
      "flos": 19569987031680.0,
      "grad_norm": 1.9841370292342715,
      "language_loss": 0.88772452,
      "learning_rate": 3.5840552701563448e-06,
      "loss": 0.90995097,
      "num_input_tokens_seen": 41300255,
      "step": 1936,
      "time_per_iteration": 2.712817668914795
    },
    {
      "auxiliary_loss_clip": 0.01116061,
      "auxiliary_loss_mlp": 0.01035056,
      "balance_loss_clip": 1.04343987,
      "balance_loss_mlp": 1.02512574,
      "epoch": 0.2329104791679192,
      "flos": 16727514215040.0,
      "grad_norm": 2.3329486894795615,
      "language_loss": 0.81929338,
      "learning_rate": 3.5835795993272513e-06,
      "loss": 0.84080458,
      "num_input_tokens_seen": 41318540,
      "step": 1937,
      "time_per_iteration": 2.8993687629699707
    },
    {
      "auxiliary_loss_clip": 0.0117858,
      "auxiliary_loss_mlp": 0.01036608,
      "balance_loss_clip": 1.05836809,
      "balance_loss_mlp": 1.02567101,
      "epoch": 0.2330307220585583,
      "flos": 22163743100160.0,
      "grad_norm": 1.9937654605311756,
      "language_loss": 0.70963335,
      "learning_rate": 3.583103688270391e-06,
      "loss": 0.73178524,
      "num_input_tokens_seen": 41338320,
      "step": 1938,
      "time_per_iteration": 2.941091775894165
    },
    {
      "auxiliary_loss_clip": 0.0118454,
      "auxiliary_loss_mlp": 0.01038652,
      "balance_loss_clip": 1.05648923,
      "balance_loss_mlp": 1.0278573,
      "epoch": 0.23315096494919738,
      "flos": 19317319787520.0,
      "grad_norm": 2.2421722772888826,
      "language_loss": 0.89185524,
      "learning_rate": 3.58262753705796e-06,
      "loss": 0.91408712,
      "num_input_tokens_seen": 41353210,
      "step": 1939,
      "time_per_iteration": 2.712167501449585
    },
    {
      "auxiliary_loss_clip": 0.01081803,
      "auxiliary_loss_mlp": 0.01006251,
      "balance_loss_clip": 1.02324438,
      "balance_loss_mlp": 1.00441539,
      "epoch": 0.23327120783983646,
      "flos": 53031048946560.0,
      "grad_norm": 0.754260457990315,
      "language_loss": 0.55510247,
      "learning_rate": 3.5821511457621902e-06,
      "loss": 0.57598299,
      "num_input_tokens_seen": 41410510,
      "step": 1940,
      "time_per_iteration": 3.2642579078674316
    },
    {
      "auxiliary_loss_clip": 0.01180973,
      "auxiliary_loss_mlp": 0.00766512,
      "balance_loss_clip": 1.05304432,
      "balance_loss_mlp": 1.00055003,
      "epoch": 0.23339145073047557,
      "flos": 17126984344320.0,
      "grad_norm": 2.8636161402470184,
      "language_loss": 0.80701613,
      "learning_rate": 3.5816745144553497e-06,
      "loss": 0.826491,
      "num_input_tokens_seen": 41425830,
      "step": 1941,
      "time_per_iteration": 2.792553186416626
    },
    {
      "auxiliary_loss_clip": 0.01143739,
      "auxiliary_loss_mlp": 0.0103291,
      "balance_loss_clip": 1.04671597,
      "balance_loss_mlp": 1.02220511,
      "epoch": 0.23351169362111465,
      "flos": 13078918419840.0,
      "grad_norm": 2.0110869892068024,
      "language_loss": 0.75139719,
      "learning_rate": 3.5811976432097424e-06,
      "loss": 0.77316368,
      "num_input_tokens_seen": 41443500,
      "step": 1942,
      "time_per_iteration": 2.7732841968536377
    },
    {
      "auxiliary_loss_clip": 0.0116304,
      "auxiliary_loss_mlp": 0.01041656,
      "balance_loss_clip": 1.05218959,
      "balance_loss_mlp": 1.0311178,
      "epoch": 0.23363193651175373,
      "flos": 15851257931520.0,
      "grad_norm": 2.0660855840003287,
      "language_loss": 0.84482145,
      "learning_rate": 3.58072053209771e-06,
      "loss": 0.86686838,
      "num_input_tokens_seen": 41460055,
      "step": 1943,
      "time_per_iteration": 2.7874300479888916
    },
    {
      "auxiliary_loss_clip": 0.01145772,
      "auxiliary_loss_mlp": 0.01035191,
      "balance_loss_clip": 1.05283523,
      "balance_loss_mlp": 1.02375889,
      "epoch": 0.23375217940239285,
      "flos": 21025769345280.0,
      "grad_norm": 2.363808100184675,
      "language_loss": 0.79116976,
      "learning_rate": 3.5802431811916296e-06,
      "loss": 0.81297946,
      "num_input_tokens_seen": 41476665,
      "step": 1944,
      "time_per_iteration": 2.78483247756958
    },
    {
      "auxiliary_loss_clip": 0.01135956,
      "auxiliary_loss_mlp": 0.01032004,
      "balance_loss_clip": 1.04940557,
      "balance_loss_mlp": 1.02209175,
      "epoch": 0.23387242229303193,
      "flos": 20594698225920.0,
      "grad_norm": 1.8559921929776406,
      "language_loss": 0.80613434,
      "learning_rate": 3.579765590563916e-06,
      "loss": 0.82781392,
      "num_input_tokens_seen": 41496065,
      "step": 1945,
      "time_per_iteration": 2.852703332901001
    },
    {
      "auxiliary_loss_clip": 0.01146785,
      "auxiliary_loss_mlp": 0.010366,
      "balance_loss_clip": 1.05278277,
      "balance_loss_mlp": 1.0254364,
      "epoch": 0.233992665183671,
      "flos": 24279491952000.0,
      "grad_norm": 1.9062814379578614,
      "language_loss": 0.8179369,
      "learning_rate": 3.579287760287017e-06,
      "loss": 0.83977079,
      "num_input_tokens_seen": 41516815,
      "step": 1946,
      "time_per_iteration": 2.8443939685821533
    },
    {
      "auxiliary_loss_clip": 0.011102,
      "auxiliary_loss_mlp": 0.01033784,
      "balance_loss_clip": 1.04755282,
      "balance_loss_mlp": 1.02381825,
      "epoch": 0.2341129080743101,
      "flos": 30154621121280.0,
      "grad_norm": 1.9236764131439217,
      "language_loss": 0.73032755,
      "learning_rate": 3.578809690433421e-06,
      "loss": 0.7517674,
      "num_input_tokens_seen": 41538525,
      "step": 1947,
      "time_per_iteration": 2.8967092037200928
    },
    {
      "auxiliary_loss_clip": 0.01111633,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.05011153,
      "balance_loss_mlp": 1.02203536,
      "epoch": 0.2342331509649492,
      "flos": 22784135829120.0,
      "grad_norm": 3.022609480063606,
      "language_loss": 0.81145751,
      "learning_rate": 3.578331381075651e-06,
      "loss": 0.83291292,
      "num_input_tokens_seen": 41559025,
      "step": 1948,
      "time_per_iteration": 2.8545830249786377
    },
    {
      "auxiliary_loss_clip": 0.01183506,
      "auxiliary_loss_mlp": 0.0103826,
      "balance_loss_clip": 1.05743718,
      "balance_loss_mlp": 1.02692914,
      "epoch": 0.2343533938555883,
      "flos": 23623152687360.0,
      "grad_norm": 2.5503819744596217,
      "language_loss": 0.69677186,
      "learning_rate": 3.5778528322862646e-06,
      "loss": 0.71898955,
      "num_input_tokens_seen": 41577845,
      "step": 1949,
      "time_per_iteration": 2.7541885375976562
    },
    {
      "auxiliary_loss_clip": 0.01128623,
      "auxiliary_loss_mlp": 0.01029312,
      "balance_loss_clip": 1.04560542,
      "balance_loss_mlp": 1.01882768,
      "epoch": 0.23447363674622737,
      "flos": 24570332375040.0,
      "grad_norm": 1.5098705468740656,
      "language_loss": 0.86453617,
      "learning_rate": 3.5773740441378585e-06,
      "loss": 0.88611555,
      "num_input_tokens_seen": 41598600,
      "step": 1950,
      "time_per_iteration": 2.813707113265991
    },
    {
      "auxiliary_loss_clip": 0.01129455,
      "auxiliary_loss_mlp": 0.01037083,
      "balance_loss_clip": 1.04702711,
      "balance_loss_mlp": 1.02693248,
      "epoch": 0.23459387963686648,
      "flos": 53140322119680.0,
      "grad_norm": 1.7318078736710316,
      "language_loss": 0.74047184,
      "learning_rate": 3.5768950167030633e-06,
      "loss": 0.76213717,
      "num_input_tokens_seen": 41623300,
      "step": 1951,
      "time_per_iteration": 3.034038782119751
    },
    {
      "auxiliary_loss_clip": 0.01192471,
      "auxiliary_loss_mlp": 0.01032882,
      "balance_loss_clip": 1.05890095,
      "balance_loss_mlp": 1.022892,
      "epoch": 0.23471412252750556,
      "flos": 23951412103680.0,
      "grad_norm": 1.8061378577086142,
      "language_loss": 0.78758538,
      "learning_rate": 3.576415750054548e-06,
      "loss": 0.80983889,
      "num_input_tokens_seen": 41643420,
      "step": 1952,
      "time_per_iteration": 2.685673952102661
    },
    {
      "auxiliary_loss_clip": 0.01144301,
      "auxiliary_loss_mlp": 0.01035344,
      "balance_loss_clip": 1.04826355,
      "balance_loss_mlp": 1.02481174,
      "epoch": 0.23483436541814465,
      "flos": 15706573948800.0,
      "grad_norm": 3.951901909971082,
      "language_loss": 0.85659885,
      "learning_rate": 3.5759362442650172e-06,
      "loss": 0.87839532,
      "num_input_tokens_seen": 41660170,
      "step": 1953,
      "time_per_iteration": 3.850216865539551
    },
    {
      "auxiliary_loss_clip": 0.01142675,
      "auxiliary_loss_mlp": 0.01039908,
      "balance_loss_clip": 1.05506301,
      "balance_loss_mlp": 1.02903032,
      "epoch": 0.23495460830878373,
      "flos": 24936262179840.0,
      "grad_norm": 2.229328292261438,
      "language_loss": 0.85257769,
      "learning_rate": 3.5754564994072113e-06,
      "loss": 0.87440354,
      "num_input_tokens_seen": 41679010,
      "step": 1954,
      "time_per_iteration": 3.7195310592651367
    },
    {
      "auxiliary_loss_clip": 0.01126459,
      "auxiliary_loss_mlp": 0.01035809,
      "balance_loss_clip": 1.04350901,
      "balance_loss_mlp": 1.02477062,
      "epoch": 0.23507485119942284,
      "flos": 30482665056000.0,
      "grad_norm": 3.8377450003642695,
      "language_loss": 0.60026801,
      "learning_rate": 3.5749765155539067e-06,
      "loss": 0.62189066,
      "num_input_tokens_seen": 41699495,
      "step": 1955,
      "time_per_iteration": 2.8603298664093018
    },
    {
      "auxiliary_loss_clip": 0.01184187,
      "auxiliary_loss_mlp": 0.01029112,
      "balance_loss_clip": 1.05385327,
      "balance_loss_mlp": 1.01807928,
      "epoch": 0.23519509409006192,
      "flos": 18329129746560.0,
      "grad_norm": 2.3865602692438497,
      "language_loss": 0.92729443,
      "learning_rate": 3.574496292777917e-06,
      "loss": 0.94942743,
      "num_input_tokens_seen": 41717705,
      "step": 1956,
      "time_per_iteration": 4.583717584609985
    },
    {
      "auxiliary_loss_clip": 0.01155173,
      "auxiliary_loss_mlp": 0.01037667,
      "balance_loss_clip": 1.05341041,
      "balance_loss_mlp": 1.02581763,
      "epoch": 0.235315336980701,
      "flos": 29643217234560.0,
      "grad_norm": 2.2736504268072704,
      "language_loss": 0.71763206,
      "learning_rate": 3.574015831152092e-06,
      "loss": 0.73956048,
      "num_input_tokens_seen": 41738120,
      "step": 1957,
      "time_per_iteration": 2.8490476608276367
    },
    {
      "auxiliary_loss_clip": 0.01192245,
      "auxiliary_loss_mlp": 0.01035942,
      "balance_loss_clip": 1.0603447,
      "balance_loss_mlp": 1.02519488,
      "epoch": 0.23543557987134012,
      "flos": 18551704371840.0,
      "grad_norm": 2.191635551668364,
      "language_loss": 0.83356309,
      "learning_rate": 3.573535130749316e-06,
      "loss": 0.85584497,
      "num_input_tokens_seen": 41756070,
      "step": 1958,
      "time_per_iteration": 2.6944336891174316
    },
    {
      "auxiliary_loss_clip": 0.01163296,
      "auxiliary_loss_mlp": 0.01038264,
      "balance_loss_clip": 1.05599451,
      "balance_loss_mlp": 1.02818489,
      "epoch": 0.2355558227619792,
      "flos": 24679033908480.0,
      "grad_norm": 3.924812030545676,
      "language_loss": 0.73859859,
      "learning_rate": 3.5730541916425127e-06,
      "loss": 0.76061416,
      "num_input_tokens_seen": 41777550,
      "step": 1959,
      "time_per_iteration": 2.803746223449707
    },
    {
      "auxiliary_loss_clip": 0.01151452,
      "auxiliary_loss_mlp": 0.01035298,
      "balance_loss_clip": 1.05006742,
      "balance_loss_mlp": 1.02365732,
      "epoch": 0.23567606565261828,
      "flos": 21944795748480.0,
      "grad_norm": 1.79257424514605,
      "language_loss": 0.85852659,
      "learning_rate": 3.572573013904639e-06,
      "loss": 0.8803941,
      "num_input_tokens_seen": 41797460,
      "step": 1960,
      "time_per_iteration": 2.774048328399658
    },
    {
      "auxiliary_loss_clip": 0.01136597,
      "auxiliary_loss_mlp": 0.01034069,
      "balance_loss_clip": 1.04815483,
      "balance_loss_mlp": 1.0237571,
      "epoch": 0.2357963085432574,
      "flos": 13589352639360.0,
      "grad_norm": 1.8501004167084154,
      "language_loss": 0.9172101,
      "learning_rate": 3.572091597608689e-06,
      "loss": 0.9389168,
      "num_input_tokens_seen": 41815585,
      "step": 1961,
      "time_per_iteration": 2.9771533012390137
    },
    {
      "auxiliary_loss_clip": 0.01198443,
      "auxiliary_loss_mlp": 0.01037713,
      "balance_loss_clip": 1.05616188,
      "balance_loss_mlp": 1.02680576,
      "epoch": 0.23591655143389648,
      "flos": 22088689632000.0,
      "grad_norm": 2.7412197152883833,
      "language_loss": 0.73392987,
      "learning_rate": 3.571609942827694e-06,
      "loss": 0.75629145,
      "num_input_tokens_seen": 41834700,
      "step": 1962,
      "time_per_iteration": 2.7169415950775146
    },
    {
      "auxiliary_loss_clip": 0.01136825,
      "auxiliary_loss_mlp": 0.0103631,
      "balance_loss_clip": 1.04739726,
      "balance_loss_mlp": 1.02623677,
      "epoch": 0.23603679432453556,
      "flos": 17017349057280.0,
      "grad_norm": 1.7580447916184205,
      "language_loss": 0.88312447,
      "learning_rate": 3.57112804963472e-06,
      "loss": 0.90485585,
      "num_input_tokens_seen": 41852915,
      "step": 1963,
      "time_per_iteration": 2.748574733734131
    },
    {
      "auxiliary_loss_clip": 0.01134632,
      "auxiliary_loss_mlp": 0.01032606,
      "balance_loss_clip": 1.05635583,
      "balance_loss_mlp": 1.02207351,
      "epoch": 0.23615703721517464,
      "flos": 19171307001600.0,
      "grad_norm": 1.779499673729491,
      "language_loss": 0.76454186,
      "learning_rate": 3.57064591810287e-06,
      "loss": 0.78621423,
      "num_input_tokens_seen": 41870415,
      "step": 1964,
      "time_per_iteration": 2.748159170150757
    },
    {
      "auxiliary_loss_clip": 0.01167822,
      "auxiliary_loss_mlp": 0.01029573,
      "balance_loss_clip": 1.053514,
      "balance_loss_mlp": 1.02023935,
      "epoch": 0.23627728010581375,
      "flos": 19098803399040.0,
      "grad_norm": 3.984640231824102,
      "language_loss": 0.80167401,
      "learning_rate": 3.570163548305284e-06,
      "loss": 0.82364798,
      "num_input_tokens_seen": 41889345,
      "step": 1965,
      "time_per_iteration": 2.6891915798187256
    },
    {
      "auxiliary_loss_clip": 0.01193428,
      "auxiliary_loss_mlp": 0.01041803,
      "balance_loss_clip": 1.05488098,
      "balance_loss_mlp": 1.03038883,
      "epoch": 0.23639752299645284,
      "flos": 14282213057280.0,
      "grad_norm": 2.4252244924337196,
      "language_loss": 0.69859642,
      "learning_rate": 3.569680940315135e-06,
      "loss": 0.72094876,
      "num_input_tokens_seen": 41905745,
      "step": 1966,
      "time_per_iteration": 2.6151270866394043
    },
    {
      "auxiliary_loss_clip": 0.01187142,
      "auxiliary_loss_mlp": 0.01043302,
      "balance_loss_clip": 1.05923986,
      "balance_loss_mlp": 1.03230464,
      "epoch": 0.23651776588709192,
      "flos": 22893411980160.0,
      "grad_norm": 1.8809989378181076,
      "language_loss": 0.82034874,
      "learning_rate": 3.5691980942056356e-06,
      "loss": 0.84265321,
      "num_input_tokens_seen": 41925115,
      "step": 1967,
      "time_per_iteration": 2.734827995300293
    },
    {
      "auxiliary_loss_clip": 0.0109939,
      "auxiliary_loss_mlp": 0.01034157,
      "balance_loss_clip": 1.04492927,
      "balance_loss_mlp": 1.02265978,
      "epoch": 0.23663800877773103,
      "flos": 18624531196800.0,
      "grad_norm": 1.95700718190351,
      "language_loss": 0.79208052,
      "learning_rate": 3.5687150100500332e-06,
      "loss": 0.813416,
      "num_input_tokens_seen": 41944815,
      "step": 1968,
      "time_per_iteration": 2.804368734359741
    },
    {
      "auxiliary_loss_clip": 0.01127104,
      "auxiliary_loss_mlp": 0.01041894,
      "balance_loss_clip": 1.04707181,
      "balance_loss_mlp": 1.03074169,
      "epoch": 0.2367582516683701,
      "flos": 25555828896000.0,
      "grad_norm": 1.6324672256986206,
      "language_loss": 0.74553776,
      "learning_rate": 3.568231687921611e-06,
      "loss": 0.76722771,
      "num_input_tokens_seen": 41964990,
      "step": 1969,
      "time_per_iteration": 2.807912826538086
    },
    {
      "auxiliary_loss_clip": 0.01109265,
      "auxiliary_loss_mlp": 0.01035319,
      "balance_loss_clip": 1.04807127,
      "balance_loss_mlp": 1.02487004,
      "epoch": 0.2368784945590092,
      "flos": 23295072839040.0,
      "grad_norm": 1.5757849118511535,
      "language_loss": 0.80686128,
      "learning_rate": 3.5677481278936883e-06,
      "loss": 0.82830715,
      "num_input_tokens_seen": 41984570,
      "step": 1970,
      "time_per_iteration": 2.846386671066284
    },
    {
      "auxiliary_loss_clip": 0.01037842,
      "auxiliary_loss_mlp": 0.01003379,
      "balance_loss_clip": 1.02264798,
      "balance_loss_mlp": 1.00171018,
      "epoch": 0.23699873744964828,
      "flos": 69859291875840.0,
      "grad_norm": 0.8295482687154213,
      "language_loss": 0.57836735,
      "learning_rate": 3.5672643300396214e-06,
      "loss": 0.59877956,
      "num_input_tokens_seen": 42053715,
      "step": 1971,
      "time_per_iteration": 3.4799039363861084
    },
    {
      "auxiliary_loss_clip": 0.01170118,
      "auxiliary_loss_mlp": 0.00765116,
      "balance_loss_clip": 1.05488825,
      "balance_loss_mlp": 1.00042629,
      "epoch": 0.2371189803402874,
      "flos": 21835052720640.0,
      "grad_norm": 2.4647158024023725,
      "language_loss": 0.67386603,
      "learning_rate": 3.566780294432802e-06,
      "loss": 0.69321835,
      "num_input_tokens_seen": 42070890,
      "step": 1972,
      "time_per_iteration": 2.877135992050171
    },
    {
      "auxiliary_loss_clip": 0.01196002,
      "auxiliary_loss_mlp": 0.01032042,
      "balance_loss_clip": 1.05671048,
      "balance_loss_mlp": 1.02130127,
      "epoch": 0.23723922323092647,
      "flos": 21908490076800.0,
      "grad_norm": 4.4556911251641615,
      "language_loss": 0.74275702,
      "learning_rate": 3.566296021146657e-06,
      "loss": 0.76503742,
      "num_input_tokens_seen": 42090270,
      "step": 1973,
      "time_per_iteration": 2.6989712715148926
    },
    {
      "auxiliary_loss_clip": 0.01144154,
      "auxiliary_loss_mlp": 0.01033423,
      "balance_loss_clip": 1.05088043,
      "balance_loss_mlp": 1.02137041,
      "epoch": 0.23735946612156555,
      "flos": 32708803380480.0,
      "grad_norm": 1.661347457011444,
      "language_loss": 0.73377377,
      "learning_rate": 3.565811510254652e-06,
      "loss": 0.75554955,
      "num_input_tokens_seen": 42111150,
      "step": 1974,
      "time_per_iteration": 2.8197035789489746
    },
    {
      "auxiliary_loss_clip": 0.01043058,
      "auxiliary_loss_mlp": 0.01002847,
      "balance_loss_clip": 1.01794004,
      "balance_loss_mlp": 1.00119019,
      "epoch": 0.23747970901220466,
      "flos": 70546944821760.0,
      "grad_norm": 0.835956857852852,
      "language_loss": 0.58255869,
      "learning_rate": 3.5653267618302845e-06,
      "loss": 0.60301775,
      "num_input_tokens_seen": 42178730,
      "step": 1975,
      "time_per_iteration": 3.367081642150879
    },
    {
      "auxiliary_loss_clip": 0.01158429,
      "auxiliary_loss_mlp": 0.01039902,
      "balance_loss_clip": 1.04874527,
      "balance_loss_mlp": 1.02879143,
      "epoch": 0.23759995190284375,
      "flos": 20849807594880.0,
      "grad_norm": 1.8575435350243774,
      "language_loss": 0.85844028,
      "learning_rate": 3.564841775947093e-06,
      "loss": 0.88042355,
      "num_input_tokens_seen": 42199620,
      "step": 1976,
      "time_per_iteration": 2.767172336578369
    },
    {
      "auxiliary_loss_clip": 0.01194348,
      "auxiliary_loss_mlp": 0.01038851,
      "balance_loss_clip": 1.05977643,
      "balance_loss_mlp": 1.02820587,
      "epoch": 0.23772019479348283,
      "flos": 32921645420160.0,
      "grad_norm": 2.329874645766392,
      "language_loss": 0.7612614,
      "learning_rate": 3.5643565526786475e-06,
      "loss": 0.78359342,
      "num_input_tokens_seen": 42219560,
      "step": 1977,
      "time_per_iteration": 3.015263557434082
    },
    {
      "auxiliary_loss_clip": 0.01153489,
      "auxiliary_loss_mlp": 0.01032973,
      "balance_loss_clip": 1.05346274,
      "balance_loss_mlp": 1.02081323,
      "epoch": 0.2378404376841219,
      "flos": 32342765834880.0,
      "grad_norm": 1.6302593466162252,
      "language_loss": 0.77160662,
      "learning_rate": 3.5638710920985574e-06,
      "loss": 0.79347122,
      "num_input_tokens_seen": 42241020,
      "step": 1978,
      "time_per_iteration": 2.846900224685669
    },
    {
      "auxiliary_loss_clip": 0.01195331,
      "auxiliary_loss_mlp": 0.01037623,
      "balance_loss_clip": 1.0571053,
      "balance_loss_mlp": 1.02543402,
      "epoch": 0.23796068057476102,
      "flos": 22997624313600.0,
      "grad_norm": 2.113339395060178,
      "language_loss": 0.82004529,
      "learning_rate": 3.5633853942804655e-06,
      "loss": 0.84237486,
      "num_input_tokens_seen": 42259345,
      "step": 1979,
      "time_per_iteration": 3.9196910858154297
    },
    {
      "auxiliary_loss_clip": 0.01142806,
      "auxiliary_loss_mlp": 0.01039378,
      "balance_loss_clip": 1.04658675,
      "balance_loss_mlp": 1.02723622,
      "epoch": 0.2380809234654001,
      "flos": 13480938414720.0,
      "grad_norm": 2.1390646536285405,
      "language_loss": 0.76839757,
      "learning_rate": 3.5628994592980527e-06,
      "loss": 0.79021943,
      "num_input_tokens_seen": 42277250,
      "step": 1980,
      "time_per_iteration": 2.759484052658081
    },
    {
      "auxiliary_loss_clip": 0.01180679,
      "auxiliary_loss_mlp": 0.01043331,
      "balance_loss_clip": 1.05447865,
      "balance_loss_mlp": 1.03180385,
      "epoch": 0.2382011663560392,
      "flos": 16871803148160.0,
      "grad_norm": 2.8138045756893786,
      "language_loss": 0.701594,
      "learning_rate": 3.562413287225034e-06,
      "loss": 0.72383416,
      "num_input_tokens_seen": 42295360,
      "step": 1981,
      "time_per_iteration": 3.6612460613250732
    },
    {
      "auxiliary_loss_clip": 0.01144772,
      "auxiliary_loss_mlp": 0.01041156,
      "balance_loss_clip": 1.05130386,
      "balance_loss_mlp": 1.02874684,
      "epoch": 0.2383214092466783,
      "flos": 18441135331200.0,
      "grad_norm": 2.16365914419778,
      "language_loss": 0.89243269,
      "learning_rate": 3.5619268781351623e-06,
      "loss": 0.91429198,
      "num_input_tokens_seen": 42313430,
      "step": 1982,
      "time_per_iteration": 4.538982629776001
    },
    {
      "auxiliary_loss_clip": 0.01160145,
      "auxiliary_loss_mlp": 0.01038917,
      "balance_loss_clip": 1.05242634,
      "balance_loss_mlp": 1.0275327,
      "epoch": 0.23844165213731738,
      "flos": 19755717281280.0,
      "grad_norm": 2.956762303318106,
      "language_loss": 0.76923668,
      "learning_rate": 3.5614402321022256e-06,
      "loss": 0.79122728,
      "num_input_tokens_seen": 42331260,
      "step": 1983,
      "time_per_iteration": 2.7488532066345215
    },
    {
      "auxiliary_loss_clip": 0.01176392,
      "auxiliary_loss_mlp": 0.01035225,
      "balance_loss_clip": 1.05682278,
      "balance_loss_mlp": 1.02483022,
      "epoch": 0.23856189502795647,
      "flos": 23367360960000.0,
      "grad_norm": 1.8705982658104277,
      "language_loss": 0.87294436,
      "learning_rate": 3.5609533492000463e-06,
      "loss": 0.89506054,
      "num_input_tokens_seen": 42350150,
      "step": 1984,
      "time_per_iteration": 2.7816789150238037
    },
    {
      "auxiliary_loss_clip": 0.01182322,
      "auxiliary_loss_mlp": 0.01039849,
      "balance_loss_clip": 1.05663598,
      "balance_loss_mlp": 1.02859521,
      "epoch": 0.23868213791859555,
      "flos": 23475056912640.0,
      "grad_norm": 2.1598741776655106,
      "language_loss": 0.7836566,
      "learning_rate": 3.560466229502485e-06,
      "loss": 0.80587828,
      "num_input_tokens_seen": 42369495,
      "step": 1985,
      "time_per_iteration": 2.6613948345184326
    },
    {
      "auxiliary_loss_clip": 0.01133545,
      "auxiliary_loss_mlp": 0.00765941,
      "balance_loss_clip": 1.05426157,
      "balance_loss_mlp": 1.00038958,
      "epoch": 0.23880238080923466,
      "flos": 16617340224000.0,
      "grad_norm": 2.150733812168654,
      "language_loss": 0.89983571,
      "learning_rate": 3.5599788730834384e-06,
      "loss": 0.91883057,
      "num_input_tokens_seen": 42387455,
      "step": 1986,
      "time_per_iteration": 2.8071677684783936
    },
    {
      "auxiliary_loss_clip": 0.01180723,
      "auxiliary_loss_mlp": 0.01036073,
      "balance_loss_clip": 1.05406237,
      "balance_loss_mlp": 1.02506995,
      "epoch": 0.23892262369987374,
      "flos": 17348409734400.0,
      "grad_norm": 2.5730901561277015,
      "language_loss": 0.78822416,
      "learning_rate": 3.559491280016836e-06,
      "loss": 0.81039214,
      "num_input_tokens_seen": 42405400,
      "step": 1987,
      "time_per_iteration": 2.695791721343994
    },
    {
      "auxiliary_loss_clip": 0.01173475,
      "auxiliary_loss_mlp": 0.01036083,
      "balance_loss_clip": 1.05414486,
      "balance_loss_mlp": 1.02443647,
      "epoch": 0.23904286659051283,
      "flos": 22309899540480.0,
      "grad_norm": 1.8664517243152035,
      "language_loss": 0.71005249,
      "learning_rate": 3.5590034503766465e-06,
      "loss": 0.73214805,
      "num_input_tokens_seen": 42425065,
      "step": 1988,
      "time_per_iteration": 2.73650860786438
    },
    {
      "auxiliary_loss_clip": 0.0114194,
      "auxiliary_loss_mlp": 0.01035975,
      "balance_loss_clip": 1.05349159,
      "balance_loss_mlp": 1.02524066,
      "epoch": 0.23916310948115194,
      "flos": 21178246579200.0,
      "grad_norm": 2.102818370618407,
      "language_loss": 0.81633848,
      "learning_rate": 3.558515384236874e-06,
      "loss": 0.83811766,
      "num_input_tokens_seen": 42442495,
      "step": 1989,
      "time_per_iteration": 2.6948142051696777
    },
    {
      "auxiliary_loss_clip": 0.01212352,
      "auxiliary_loss_mlp": 0.01033262,
      "balance_loss_clip": 1.06144643,
      "balance_loss_mlp": 1.021734,
      "epoch": 0.23928335237179102,
      "flos": 14137349506560.0,
      "grad_norm": 1.951458226672493,
      "language_loss": 0.83574605,
      "learning_rate": 3.558027081671556e-06,
      "loss": 0.85820222,
      "num_input_tokens_seen": 42459480,
      "step": 1990,
      "time_per_iteration": 2.6461715698242188
    },
    {
      "auxiliary_loss_clip": 0.01148857,
      "auxiliary_loss_mlp": 0.01035326,
      "balance_loss_clip": 1.05337524,
      "balance_loss_mlp": 1.02331531,
      "epoch": 0.2394035952624301,
      "flos": 23769596436480.0,
      "grad_norm": 2.8325040267232344,
      "language_loss": 0.69019425,
      "learning_rate": 3.557538542754769e-06,
      "loss": 0.71203601,
      "num_input_tokens_seen": 42479175,
      "step": 1991,
      "time_per_iteration": 2.8573009967803955
    },
    {
      "auxiliary_loss_clip": 0.01193494,
      "auxiliary_loss_mlp": 0.01037523,
      "balance_loss_clip": 1.05691803,
      "balance_loss_mlp": 1.02563822,
      "epoch": 0.2395238381530692,
      "flos": 24206198250240.0,
      "grad_norm": 1.8146011134168742,
      "language_loss": 0.66967112,
      "learning_rate": 3.557049767560623e-06,
      "loss": 0.69198132,
      "num_input_tokens_seen": 42498090,
      "step": 1992,
      "time_per_iteration": 2.6643919944763184
    },
    {
      "auxiliary_loss_clip": 0.01149143,
      "auxiliary_loss_mlp": 0.01035751,
      "balance_loss_clip": 1.05311227,
      "balance_loss_mlp": 1.02427101,
      "epoch": 0.2396440810437083,
      "flos": 25295763450240.0,
      "grad_norm": 2.0754319911708516,
      "language_loss": 0.859074,
      "learning_rate": 3.5565607561632655e-06,
      "loss": 0.88092291,
      "num_input_tokens_seen": 42516930,
      "step": 1993,
      "time_per_iteration": 2.7587029933929443
    },
    {
      "auxiliary_loss_clip": 0.01190276,
      "auxiliary_loss_mlp": 0.01040591,
      "balance_loss_clip": 1.05427694,
      "balance_loss_mlp": 1.02818775,
      "epoch": 0.23976432393434738,
      "flos": 28543093436160.0,
      "grad_norm": 2.1014092603252985,
      "language_loss": 0.79388547,
      "learning_rate": 3.5560715086368787e-06,
      "loss": 0.81619418,
      "num_input_tokens_seen": 42534800,
      "step": 1994,
      "time_per_iteration": 2.7373404502868652
    },
    {
      "auxiliary_loss_clip": 0.01158451,
      "auxiliary_loss_mlp": 0.0103631,
      "balance_loss_clip": 1.04955959,
      "balance_loss_mlp": 1.02486014,
      "epoch": 0.23988456682498646,
      "flos": 19494358945920.0,
      "grad_norm": 1.8610056836935107,
      "language_loss": 0.81866205,
      "learning_rate": 3.5555820250556816e-06,
      "loss": 0.84060967,
      "num_input_tokens_seen": 42552000,
      "step": 1995,
      "time_per_iteration": 2.695013999938965
    },
    {
      "auxiliary_loss_clip": 0.01160061,
      "auxiliary_loss_mlp": 0.01038068,
      "balance_loss_clip": 1.05515289,
      "balance_loss_mlp": 1.02554476,
      "epoch": 0.24000480971562557,
      "flos": 20266331068800.0,
      "grad_norm": 2.3597108911492226,
      "language_loss": 0.69168884,
      "learning_rate": 3.5550923054939278e-06,
      "loss": 0.71367013,
      "num_input_tokens_seen": 42571455,
      "step": 1996,
      "time_per_iteration": 2.7548370361328125
    },
    {
      "auxiliary_loss_clip": 0.011623,
      "auxiliary_loss_mlp": 0.01036368,
      "balance_loss_clip": 1.05441058,
      "balance_loss_mlp": 1.02509117,
      "epoch": 0.24012505260626466,
      "flos": 25443176866560.0,
      "grad_norm": 2.0040834230744924,
      "language_loss": 0.74482119,
      "learning_rate": 3.5546023500259083e-06,
      "loss": 0.76680791,
      "num_input_tokens_seen": 42592550,
      "step": 1997,
      "time_per_iteration": 2.835874319076538
    },
    {
      "auxiliary_loss_clip": 0.01132476,
      "auxiliary_loss_mlp": 0.01034021,
      "balance_loss_clip": 1.04899716,
      "balance_loss_mlp": 1.02231479,
      "epoch": 0.24024529549690374,
      "flos": 15553342529280.0,
      "grad_norm": 2.04419216699465,
      "language_loss": 0.80713964,
      "learning_rate": 3.5541121587259477e-06,
      "loss": 0.82880455,
      "num_input_tokens_seen": 42610385,
      "step": 1998,
      "time_per_iteration": 2.720419406890869
    },
    {
      "auxiliary_loss_clip": 0.0108453,
      "auxiliary_loss_mlp": 0.0100649,
      "balance_loss_clip": 1.02821863,
      "balance_loss_mlp": 1.00461793,
      "epoch": 0.24036553838754285,
      "flos": 57122351867520.0,
      "grad_norm": 0.8336420815007461,
      "language_loss": 0.57886118,
      "learning_rate": 3.553621731668408e-06,
      "loss": 0.59977138,
      "num_input_tokens_seen": 42673595,
      "step": 1999,
      "time_per_iteration": 3.3336074352264404
    },
    {
      "auxiliary_loss_clip": 0.01122873,
      "auxiliary_loss_mlp": 0.01031724,
      "balance_loss_clip": 1.04482305,
      "balance_loss_mlp": 1.02032733,
      "epoch": 0.24048578127818193,
      "flos": 24969946158720.0,
      "grad_norm": 1.857415396048282,
      "language_loss": 0.8311317,
      "learning_rate": 3.553131068927688e-06,
      "loss": 0.85267758,
      "num_input_tokens_seen": 42692000,
      "step": 2000,
      "time_per_iteration": 2.811185359954834
    },
    {
      "auxiliary_loss_clip": 0.01137371,
      "auxiliary_loss_mlp": 0.01038694,
      "balance_loss_clip": 1.04982817,
      "balance_loss_mlp": 1.02800083,
      "epoch": 0.24060602416882101,
      "flos": 23330947547520.0,
      "grad_norm": 3.418798403634787,
      "language_loss": 0.8047092,
      "learning_rate": 3.552640170578219e-06,
      "loss": 0.8264699,
      "num_input_tokens_seen": 42712250,
      "step": 2001,
      "time_per_iteration": 2.832458972930908
    },
    {
      "auxiliary_loss_clip": 0.01209652,
      "auxiliary_loss_mlp": 0.01033214,
      "balance_loss_clip": 1.05965805,
      "balance_loss_mlp": 1.02256846,
      "epoch": 0.2407262670594601,
      "flos": 14173260128640.0,
      "grad_norm": 1.933529565427797,
      "language_loss": 0.77731252,
      "learning_rate": 3.5521490366944703e-06,
      "loss": 0.79974127,
      "num_input_tokens_seen": 42729900,
      "step": 2002,
      "time_per_iteration": 2.666938066482544
    },
    {
      "auxiliary_loss_clip": 0.01133374,
      "auxiliary_loss_mlp": 0.01038421,
      "balance_loss_clip": 1.04703248,
      "balance_loss_mlp": 1.02657771,
      "epoch": 0.2408465099500992,
      "flos": 13663113217920.0,
      "grad_norm": 2.031850793732826,
      "language_loss": 0.79998124,
      "learning_rate": 3.5516576673509474e-06,
      "loss": 0.8216992,
      "num_input_tokens_seen": 42747900,
      "step": 2003,
      "time_per_iteration": 2.787015438079834
    },
    {
      "auxiliary_loss_clip": 0.01149752,
      "auxiliary_loss_mlp": 0.01036934,
      "balance_loss_clip": 1.05467248,
      "balance_loss_mlp": 1.02569842,
      "epoch": 0.2409667528407383,
      "flos": 31248029076480.0,
      "grad_norm": 2.386897839425311,
      "language_loss": 0.86516368,
      "learning_rate": 3.5511660626221896e-06,
      "loss": 0.88703054,
      "num_input_tokens_seen": 42768540,
      "step": 2004,
      "time_per_iteration": 2.9800775051116943
    },
    {
      "auxiliary_loss_clip": 0.01157316,
      "auxiliary_loss_mlp": 0.01039819,
      "balance_loss_clip": 1.05244517,
      "balance_loss_mlp": 1.02794552,
      "epoch": 0.24108699573137737,
      "flos": 22199941031040.0,
      "grad_norm": 2.2274480537502086,
      "language_loss": 0.88930947,
      "learning_rate": 3.5506742225827744e-06,
      "loss": 0.91128087,
      "num_input_tokens_seen": 42785395,
      "step": 2005,
      "time_per_iteration": 3.9070701599121094
    },
    {
      "auxiliary_loss_clip": 0.0117549,
      "auxiliary_loss_mlp": 0.01037853,
      "balance_loss_clip": 1.05717516,
      "balance_loss_mlp": 1.02639723,
      "epoch": 0.24120723862201648,
      "flos": 26103035664000.0,
      "grad_norm": 2.2547042154991384,
      "language_loss": 0.90422386,
      "learning_rate": 3.5501821473073116e-06,
      "loss": 0.92635733,
      "num_input_tokens_seen": 42801980,
      "step": 2006,
      "time_per_iteration": 3.759795665740967
    },
    {
      "auxiliary_loss_clip": 0.01129823,
      "auxiliary_loss_mlp": 0.01037811,
      "balance_loss_clip": 1.04914188,
      "balance_loss_mlp": 1.02624178,
      "epoch": 0.24132748151265557,
      "flos": 18624926246400.0,
      "grad_norm": 2.5729972172662565,
      "language_loss": 0.86655635,
      "learning_rate": 3.54968983687045e-06,
      "loss": 0.88823271,
      "num_input_tokens_seen": 42818850,
      "step": 2007,
      "time_per_iteration": 3.7256088256835938
    },
    {
      "auxiliary_loss_clip": 0.01211199,
      "auxiliary_loss_mlp": 0.00765739,
      "balance_loss_clip": 1.05908263,
      "balance_loss_mlp": 1.00052238,
      "epoch": 0.24144772440329465,
      "flos": 15267673664640.0,
      "grad_norm": 2.676572964979271,
      "language_loss": 0.89353758,
      "learning_rate": 3.549197291346872e-06,
      "loss": 0.91330695,
      "num_input_tokens_seen": 42835375,
      "step": 2008,
      "time_per_iteration": 3.508429527282715
    },
    {
      "auxiliary_loss_clip": 0.01125345,
      "auxiliary_loss_mlp": 0.01036377,
      "balance_loss_clip": 1.04828095,
      "balance_loss_mlp": 1.02494454,
      "epoch": 0.24156796729393373,
      "flos": 24024274842240.0,
      "grad_norm": 2.237520599574218,
      "language_loss": 0.79370892,
      "learning_rate": 3.548704510811297e-06,
      "loss": 0.81532609,
      "num_input_tokens_seen": 42854570,
      "step": 2009,
      "time_per_iteration": 2.860386371612549
    },
    {
      "auxiliary_loss_clip": 0.01195392,
      "auxiliary_loss_mlp": 0.01033673,
      "balance_loss_clip": 1.05795538,
      "balance_loss_mlp": 1.022825,
      "epoch": 0.24168821018457284,
      "flos": 26286790665600.0,
      "grad_norm": 2.4768502998162996,
      "language_loss": 0.74784899,
      "learning_rate": 3.5482114953384787e-06,
      "loss": 0.77013969,
      "num_input_tokens_seen": 42873800,
      "step": 2010,
      "time_per_iteration": 2.771315813064575
    },
    {
      "auxiliary_loss_clip": 0.01123787,
      "auxiliary_loss_mlp": 0.01035658,
      "balance_loss_clip": 1.04948664,
      "balance_loss_mlp": 1.02514935,
      "epoch": 0.24180845307521193,
      "flos": 18223193560320.0,
      "grad_norm": 2.0904233045963068,
      "language_loss": 0.84578097,
      "learning_rate": 3.5477182450032077e-06,
      "loss": 0.86737537,
      "num_input_tokens_seen": 42892400,
      "step": 2011,
      "time_per_iteration": 2.7982237339019775
    },
    {
      "auxiliary_loss_clip": 0.01214,
      "auxiliary_loss_mlp": 0.01039805,
      "balance_loss_clip": 1.06244016,
      "balance_loss_mlp": 1.02901077,
      "epoch": 0.241928695965851,
      "flos": 20449260057600.0,
      "grad_norm": 2.197108704833637,
      "language_loss": 0.83172572,
      "learning_rate": 3.5472247598803097e-06,
      "loss": 0.85426366,
      "num_input_tokens_seen": 42911745,
      "step": 2012,
      "time_per_iteration": 2.616293430328369
    },
    {
      "auxiliary_loss_clip": 0.01177103,
      "auxiliary_loss_mlp": 0.01033787,
      "balance_loss_clip": 1.05603814,
      "balance_loss_mlp": 1.02136517,
      "epoch": 0.24204893885649012,
      "flos": 25556475340800.0,
      "grad_norm": 2.2271535513864356,
      "language_loss": 0.85697234,
      "learning_rate": 3.546731040044645e-06,
      "loss": 0.87908119,
      "num_input_tokens_seen": 42926915,
      "step": 2013,
      "time_per_iteration": 2.741132974624634
    },
    {
      "auxiliary_loss_clip": 0.01147059,
      "auxiliary_loss_mlp": 0.01037208,
      "balance_loss_clip": 1.05156016,
      "balance_loss_mlp": 1.02568603,
      "epoch": 0.2421691817471292,
      "flos": 30660207004800.0,
      "grad_norm": 1.7597019814767014,
      "language_loss": 0.75020903,
      "learning_rate": 3.546237085571112e-06,
      "loss": 0.77205169,
      "num_input_tokens_seen": 42945350,
      "step": 2014,
      "time_per_iteration": 2.8546488285064697
    },
    {
      "auxiliary_loss_clip": 0.01193269,
      "auxiliary_loss_mlp": 0.01046781,
      "balance_loss_clip": 1.0576725,
      "balance_loss_mlp": 1.03586149,
      "epoch": 0.24228942463776829,
      "flos": 21945011230080.0,
      "grad_norm": 2.2706634751245467,
      "language_loss": 0.72292095,
      "learning_rate": 3.5457428965346425e-06,
      "loss": 0.74532151,
      "num_input_tokens_seen": 42964290,
      "step": 2015,
      "time_per_iteration": 2.6926357746124268
    },
    {
      "auxiliary_loss_clip": 0.01151983,
      "auxiliary_loss_mlp": 0.01034632,
      "balance_loss_clip": 1.05167663,
      "balance_loss_mlp": 1.02384377,
      "epoch": 0.2424096675284074,
      "flos": 33984493879680.0,
      "grad_norm": 1.7453104217402644,
      "language_loss": 0.74724931,
      "learning_rate": 3.545248473010205e-06,
      "loss": 0.76911545,
      "num_input_tokens_seen": 42987095,
      "step": 2016,
      "time_per_iteration": 2.8073372840881348
    },
    {
      "auxiliary_loss_clip": 0.01204385,
      "auxiliary_loss_mlp": 0.01037894,
      "balance_loss_clip": 1.06108093,
      "balance_loss_mlp": 1.02535939,
      "epoch": 0.24252991041904648,
      "flos": 21653416621440.0,
      "grad_norm": 3.409093215365441,
      "language_loss": 0.87829167,
      "learning_rate": 3.544753815072802e-06,
      "loss": 0.90071446,
      "num_input_tokens_seen": 43005750,
      "step": 2017,
      "time_per_iteration": 2.6860032081604004
    },
    {
      "auxiliary_loss_clip": 0.01210329,
      "auxiliary_loss_mlp": 0.01030684,
      "balance_loss_clip": 1.06053722,
      "balance_loss_mlp": 1.01955557,
      "epoch": 0.24265015330968556,
      "flos": 21870065502720.0,
      "grad_norm": 1.888625375926245,
      "language_loss": 0.8802411,
      "learning_rate": 3.544258922797474e-06,
      "loss": 0.90265125,
      "num_input_tokens_seen": 43023870,
      "step": 2018,
      "time_per_iteration": 2.647522449493408
    },
    {
      "auxiliary_loss_clip": 0.01170266,
      "auxiliary_loss_mlp": 0.01033284,
      "balance_loss_clip": 1.04921675,
      "balance_loss_mlp": 1.0220902,
      "epoch": 0.24277039620032465,
      "flos": 25628260671360.0,
      "grad_norm": 1.9057963709925487,
      "language_loss": 0.77988052,
      "learning_rate": 3.543763796259295e-06,
      "loss": 0.80191606,
      "num_input_tokens_seen": 43043825,
      "step": 2019,
      "time_per_iteration": 2.7484378814697266
    },
    {
      "auxiliary_loss_clip": 0.01208514,
      "auxiliary_loss_mlp": 0.01033505,
      "balance_loss_clip": 1.05634093,
      "balance_loss_mlp": 1.02223372,
      "epoch": 0.24289063909096376,
      "flos": 26286575184000.0,
      "grad_norm": 1.8318803814544298,
      "language_loss": 0.9110378,
      "learning_rate": 3.5432684355333754e-06,
      "loss": 0.93345791,
      "num_input_tokens_seen": 43062480,
      "step": 2020,
      "time_per_iteration": 2.72304630279541
    },
    {
      "auxiliary_loss_clip": 0.01197743,
      "auxiliary_loss_mlp": 0.01033458,
      "balance_loss_clip": 1.05799854,
      "balance_loss_mlp": 1.0222584,
      "epoch": 0.24301088198160284,
      "flos": 25075056332160.0,
      "grad_norm": 2.3217429057445074,
      "language_loss": 0.76801372,
      "learning_rate": 3.5427728406948613e-06,
      "loss": 0.7903257,
      "num_input_tokens_seen": 43081595,
      "step": 2021,
      "time_per_iteration": 2.684828996658325
    },
    {
      "auxiliary_loss_clip": 0.01046762,
      "auxiliary_loss_mlp": 0.01002875,
      "balance_loss_clip": 1.02351999,
      "balance_loss_mlp": 1.0011704,
      "epoch": 0.24313112487224192,
      "flos": 69900948673920.0,
      "grad_norm": 0.7990727013995524,
      "language_loss": 0.57860792,
      "learning_rate": 3.542277011818934e-06,
      "loss": 0.59910434,
      "num_input_tokens_seen": 43145430,
      "step": 2022,
      "time_per_iteration": 3.6484920978546143
    },
    {
      "auxiliary_loss_clip": 0.01178199,
      "auxiliary_loss_mlp": 0.00765065,
      "balance_loss_clip": 1.05831957,
      "balance_loss_mlp": 1.0005908,
      "epoch": 0.24325136776288103,
      "flos": 40662334235520.0,
      "grad_norm": 1.9846293687142877,
      "language_loss": 0.74197555,
      "learning_rate": 3.5417809489808104e-06,
      "loss": 0.76140821,
      "num_input_tokens_seen": 43167040,
      "step": 2023,
      "time_per_iteration": 3.0973589420318604
    },
    {
      "auxiliary_loss_clip": 0.01160576,
      "auxiliary_loss_mlp": 0.0103548,
      "balance_loss_clip": 1.05033684,
      "balance_loss_mlp": 1.02456594,
      "epoch": 0.24337161065352012,
      "flos": 25046400257280.0,
      "grad_norm": 1.923823965372565,
      "language_loss": 0.72539061,
      "learning_rate": 3.5412846522557422e-06,
      "loss": 0.74735117,
      "num_input_tokens_seen": 43187930,
      "step": 2024,
      "time_per_iteration": 2.818063497543335
    },
    {
      "auxiliary_loss_clip": 0.01158904,
      "auxiliary_loss_mlp": 0.00765343,
      "balance_loss_clip": 1.05965209,
      "balance_loss_mlp": 1.00044453,
      "epoch": 0.2434918535441592,
      "flos": 18661160090880.0,
      "grad_norm": 2.2415568760548505,
      "language_loss": 0.74249429,
      "learning_rate": 3.540788121719018e-06,
      "loss": 0.76173675,
      "num_input_tokens_seen": 43206350,
      "step": 2025,
      "time_per_iteration": 2.705803155899048
    },
    {
      "auxiliary_loss_clip": 0.01156216,
      "auxiliary_loss_mlp": 0.01031888,
      "balance_loss_clip": 1.05373144,
      "balance_loss_mlp": 1.0212487,
      "epoch": 0.24361209643479828,
      "flos": 23915142345600.0,
      "grad_norm": 2.3735765531690642,
      "language_loss": 0.8185581,
      "learning_rate": 3.5402913574459604e-06,
      "loss": 0.84043908,
      "num_input_tokens_seen": 43226255,
      "step": 2026,
      "time_per_iteration": 2.7497010231018066
    },
    {
      "auxiliary_loss_clip": 0.01141801,
      "auxiliary_loss_mlp": 0.01029324,
      "balance_loss_clip": 1.05018795,
      "balance_loss_mlp": 1.01926303,
      "epoch": 0.2437323393254374,
      "flos": 28657505232000.0,
      "grad_norm": 1.8032438157947468,
      "language_loss": 0.86254007,
      "learning_rate": 3.5397943595119297e-06,
      "loss": 0.8842513,
      "num_input_tokens_seen": 43247675,
      "step": 2027,
      "time_per_iteration": 2.899824857711792
    },
    {
      "auxiliary_loss_clip": 0.01131489,
      "auxiliary_loss_mlp": 0.0103834,
      "balance_loss_clip": 1.04981899,
      "balance_loss_mlp": 1.02691388,
      "epoch": 0.24385258221607647,
      "flos": 23550325862400.0,
      "grad_norm": 4.094705417002457,
      "language_loss": 0.77589148,
      "learning_rate": 3.5392971279923177e-06,
      "loss": 0.79758978,
      "num_input_tokens_seen": 43265895,
      "step": 2028,
      "time_per_iteration": 2.827955961227417
    },
    {
      "auxiliary_loss_clip": 0.0118684,
      "auxiliary_loss_mlp": 0.01039394,
      "balance_loss_clip": 1.05450726,
      "balance_loss_mlp": 1.02819395,
      "epoch": 0.24397282510671556,
      "flos": 25336091445120.0,
      "grad_norm": 2.0239869124489367,
      "language_loss": 0.83225167,
      "learning_rate": 3.5387996629625557e-06,
      "loss": 0.854514,
      "num_input_tokens_seen": 43283485,
      "step": 2029,
      "time_per_iteration": 2.784990072250366
    },
    {
      "auxiliary_loss_clip": 0.01093077,
      "auxiliary_loss_mlp": 0.01002914,
      "balance_loss_clip": 1.03099334,
      "balance_loss_mlp": 1.00110233,
      "epoch": 0.24409306799735467,
      "flos": 65187421430400.0,
      "grad_norm": 0.8014656554947234,
      "language_loss": 0.54999328,
      "learning_rate": 3.5383019644981083e-06,
      "loss": 0.57095319,
      "num_input_tokens_seen": 43347180,
      "step": 2030,
      "time_per_iteration": 3.3082480430603027
    },
    {
      "auxiliary_loss_clip": 0.01159117,
      "auxiliary_loss_mlp": 0.01049754,
      "balance_loss_clip": 1.05421078,
      "balance_loss_mlp": 1.03792858,
      "epoch": 0.24421331088799375,
      "flos": 19537093152000.0,
      "grad_norm": 2.5420873493213754,
      "language_loss": 0.72886068,
      "learning_rate": 3.5378040326744763e-06,
      "loss": 0.75094938,
      "num_input_tokens_seen": 43366665,
      "step": 2031,
      "time_per_iteration": 3.6642119884490967
    },
    {
      "auxiliary_loss_clip": 0.01139485,
      "auxiliary_loss_mlp": 0.01034467,
      "balance_loss_clip": 1.04924285,
      "balance_loss_mlp": 1.0231421,
      "epoch": 0.24433355377863283,
      "flos": 21068575378560.0,
      "grad_norm": 2.0771281356254523,
      "language_loss": 0.85446751,
      "learning_rate": 3.5373058675671946e-06,
      "loss": 0.87620699,
      "num_input_tokens_seen": 43384670,
      "step": 2032,
      "time_per_iteration": 3.950477361679077
    },
    {
      "auxiliary_loss_clip": 0.01199237,
      "auxiliary_loss_mlp": 0.01036482,
      "balance_loss_clip": 1.06119633,
      "balance_loss_mlp": 1.02557397,
      "epoch": 0.24445379666927192,
      "flos": 22637189289600.0,
      "grad_norm": 2.4345389333144305,
      "language_loss": 0.71995842,
      "learning_rate": 3.536807469251836e-06,
      "loss": 0.74231565,
      "num_input_tokens_seen": 43403825,
      "step": 2033,
      "time_per_iteration": 3.582909345626831
    },
    {
      "auxiliary_loss_clip": 0.01158337,
      "auxiliary_loss_mlp": 0.01041883,
      "balance_loss_clip": 1.05363262,
      "balance_loss_mlp": 1.03095722,
      "epoch": 0.24457403955991103,
      "flos": 21251612108160.0,
      "grad_norm": 1.8991116486862245,
      "language_loss": 0.8243829,
      "learning_rate": 3.5363088378040055e-06,
      "loss": 0.84638512,
      "num_input_tokens_seen": 43422715,
      "step": 2034,
      "time_per_iteration": 3.7009494304656982
    },
    {
      "auxiliary_loss_clip": 0.01039228,
      "auxiliary_loss_mlp": 0.00757955,
      "balance_loss_clip": 1.0235157,
      "balance_loss_mlp": 1.00050545,
      "epoch": 0.2446942824505501,
      "flos": 66997820764800.0,
      "grad_norm": 0.759845941682328,
      "language_loss": 0.64371461,
      "learning_rate": 3.5358099732993463e-06,
      "loss": 0.66168636,
      "num_input_tokens_seen": 43481825,
      "step": 2035,
      "time_per_iteration": 3.435791492462158
    },
    {
      "auxiliary_loss_clip": 0.01150798,
      "auxiliary_loss_mlp": 0.01032921,
      "balance_loss_clip": 1.05051458,
      "balance_loss_mlp": 1.02131653,
      "epoch": 0.2448145253411892,
      "flos": 20411122792320.0,
      "grad_norm": 2.1775509155921617,
      "language_loss": 0.89519858,
      "learning_rate": 3.535310875813535e-06,
      "loss": 0.91703582,
      "num_input_tokens_seen": 43500220,
      "step": 2036,
      "time_per_iteration": 3.0357844829559326
    },
    {
      "auxiliary_loss_clip": 0.01163352,
      "auxiliary_loss_mlp": 0.01037199,
      "balance_loss_clip": 1.05283237,
      "balance_loss_mlp": 1.02695942,
      "epoch": 0.2449347682318283,
      "flos": 28804739080320.0,
      "grad_norm": 1.8632070417448394,
      "language_loss": 0.81775033,
      "learning_rate": 3.5348115454222843e-06,
      "loss": 0.83975583,
      "num_input_tokens_seen": 43522805,
      "step": 2037,
      "time_per_iteration": 2.7948455810546875
    },
    {
      "auxiliary_loss_clip": 0.01193178,
      "auxiliary_loss_mlp": 0.01037611,
      "balance_loss_clip": 1.05826581,
      "balance_loss_mlp": 1.02636909,
      "epoch": 0.2450550111224674,
      "flos": 22528990546560.0,
      "grad_norm": 1.8562037431016916,
      "language_loss": 0.86622596,
      "learning_rate": 3.5343119822013425e-06,
      "loss": 0.88853383,
      "num_input_tokens_seen": 43541915,
      "step": 2038,
      "time_per_iteration": 2.726956605911255
    },
    {
      "auxiliary_loss_clip": 0.01151545,
      "auxiliary_loss_mlp": 0.01032995,
      "balance_loss_clip": 1.05157757,
      "balance_loss_mlp": 1.02095485,
      "epoch": 0.24517525401310647,
      "flos": 21759137326080.0,
      "grad_norm": 1.776968737651013,
      "language_loss": 0.77834284,
      "learning_rate": 3.533812186226493e-06,
      "loss": 0.80018818,
      "num_input_tokens_seen": 43562625,
      "step": 2039,
      "time_per_iteration": 2.737461805343628
    },
    {
      "auxiliary_loss_clip": 0.01205392,
      "auxiliary_loss_mlp": 0.01033505,
      "balance_loss_clip": 1.05750728,
      "balance_loss_mlp": 1.0237304,
      "epoch": 0.24529549690374555,
      "flos": 25043311687680.0,
      "grad_norm": 1.9079236569928517,
      "language_loss": 0.75672233,
      "learning_rate": 3.5333121575735545e-06,
      "loss": 0.77911133,
      "num_input_tokens_seen": 43582265,
      "step": 2040,
      "time_per_iteration": 2.69529128074646
    },
    {
      "auxiliary_loss_clip": 0.01154012,
      "auxiliary_loss_mlp": 0.01029857,
      "balance_loss_clip": 1.05278826,
      "balance_loss_mlp": 1.0190804,
      "epoch": 0.24541573979438466,
      "flos": 32123638915200.0,
      "grad_norm": 2.0580194651166046,
      "language_loss": 0.75436205,
      "learning_rate": 3.532811896318381e-06,
      "loss": 0.77620077,
      "num_input_tokens_seen": 43604335,
      "step": 2041,
      "time_per_iteration": 2.8047404289245605
    },
    {
      "auxiliary_loss_clip": 0.01195623,
      "auxiliary_loss_mlp": 0.01029658,
      "balance_loss_clip": 1.06002712,
      "balance_loss_mlp": 1.01823163,
      "epoch": 0.24553598268502375,
      "flos": 31357556622720.0,
      "grad_norm": 2.4437489757687585,
      "language_loss": 0.81435782,
      "learning_rate": 3.5323114025368615e-06,
      "loss": 0.83661062,
      "num_input_tokens_seen": 43619400,
      "step": 2042,
      "time_per_iteration": 2.7758495807647705
    },
    {
      "auxiliary_loss_clip": 0.01186237,
      "auxiliary_loss_mlp": 0.01030651,
      "balance_loss_clip": 1.05444205,
      "balance_loss_mlp": 1.02034545,
      "epoch": 0.24565622557566283,
      "flos": 14027462824320.0,
      "grad_norm": 2.1752634903205568,
      "language_loss": 0.81916189,
      "learning_rate": 3.53181067630492e-06,
      "loss": 0.84133077,
      "num_input_tokens_seen": 43636870,
      "step": 2043,
      "time_per_iteration": 2.619115114212036
    },
    {
      "auxiliary_loss_clip": 0.01169246,
      "auxiliary_loss_mlp": 0.01039433,
      "balance_loss_clip": 1.05329275,
      "balance_loss_mlp": 1.0281558,
      "epoch": 0.24577646846630194,
      "flos": 16581465515520.0,
      "grad_norm": 1.9220554526911287,
      "language_loss": 0.75973678,
      "learning_rate": 3.5313097176985175e-06,
      "loss": 0.78182358,
      "num_input_tokens_seen": 43655180,
      "step": 2044,
      "time_per_iteration": 2.7110812664031982
    },
    {
      "auxiliary_loss_clip": 0.01157166,
      "auxiliary_loss_mlp": 0.01035412,
      "balance_loss_clip": 1.05527139,
      "balance_loss_mlp": 1.02477908,
      "epoch": 0.24589671135694102,
      "flos": 18807424272000.0,
      "grad_norm": 1.8562207576564966,
      "language_loss": 0.81273198,
      "learning_rate": 3.5308085267936482e-06,
      "loss": 0.83465779,
      "num_input_tokens_seen": 43672895,
      "step": 2045,
      "time_per_iteration": 2.679760456085205
    },
    {
      "auxiliary_loss_clip": 0.01153028,
      "auxiliary_loss_mlp": 0.01036979,
      "balance_loss_clip": 1.05476868,
      "balance_loss_mlp": 1.02736521,
      "epoch": 0.2460169542475801,
      "flos": 19938538529280.0,
      "grad_norm": 1.7164303362973472,
      "language_loss": 0.89523691,
      "learning_rate": 3.530307103666342e-06,
      "loss": 0.91713697,
      "num_input_tokens_seen": 43691975,
      "step": 2046,
      "time_per_iteration": 2.7593631744384766
    },
    {
      "auxiliary_loss_clip": 0.01171986,
      "auxiliary_loss_mlp": 0.0103277,
      "balance_loss_clip": 1.05535221,
      "balance_loss_mlp": 1.021523,
      "epoch": 0.24613719713821922,
      "flos": 24171221381760.0,
      "grad_norm": 2.067453831155588,
      "language_loss": 0.80197507,
      "learning_rate": 3.5298054483926658e-06,
      "loss": 0.82402259,
      "num_input_tokens_seen": 43712670,
      "step": 2047,
      "time_per_iteration": 2.7378814220428467
    },
    {
      "auxiliary_loss_clip": 0.01168374,
      "auxiliary_loss_mlp": 0.01037945,
      "balance_loss_clip": 1.06024444,
      "balance_loss_mlp": 1.02718067,
      "epoch": 0.2462574400288583,
      "flos": 30221055325440.0,
      "grad_norm": 2.0293497646430443,
      "language_loss": 0.82531416,
      "learning_rate": 3.5293035610487187e-06,
      "loss": 0.8473773,
      "num_input_tokens_seen": 43732035,
      "step": 2048,
      "time_per_iteration": 2.8137500286102295
    },
    {
      "auxiliary_loss_clip": 0.01070059,
      "auxiliary_loss_mlp": 0.01002391,
      "balance_loss_clip": 1.0246706,
      "balance_loss_mlp": 1.0008297,
      "epoch": 0.24637768291949738,
      "flos": 68943030819840.0,
      "grad_norm": 0.7364855541845254,
      "language_loss": 0.62014419,
      "learning_rate": 3.5288014417106374e-06,
      "loss": 0.64086866,
      "num_input_tokens_seen": 43798055,
      "step": 2049,
      "time_per_iteration": 3.336472749710083
    },
    {
      "auxiliary_loss_clip": 0.01146431,
      "auxiliary_loss_mlp": 0.01032177,
      "balance_loss_clip": 1.05201936,
      "balance_loss_mlp": 1.02141821,
      "epoch": 0.24649792581013646,
      "flos": 34383999922560.0,
      "grad_norm": 1.8196831199570562,
      "language_loss": 0.75320911,
      "learning_rate": 3.528299090454593e-06,
      "loss": 0.77499521,
      "num_input_tokens_seen": 43818590,
      "step": 2050,
      "time_per_iteration": 2.956871509552002
    },
    {
      "auxiliary_loss_clip": 0.01169619,
      "auxiliary_loss_mlp": 0.01039969,
      "balance_loss_clip": 1.05315185,
      "balance_loss_mlp": 1.02860284,
      "epoch": 0.24661816870077558,
      "flos": 19680448331520.0,
      "grad_norm": 2.510408468281843,
      "language_loss": 0.83197606,
      "learning_rate": 3.527796507356792e-06,
      "loss": 0.85407192,
      "num_input_tokens_seen": 43832480,
      "step": 2051,
      "time_per_iteration": 2.740192174911499
    },
    {
      "auxiliary_loss_clip": 0.01177002,
      "auxiliary_loss_mlp": 0.01034057,
      "balance_loss_clip": 1.05477166,
      "balance_loss_mlp": 1.02375746,
      "epoch": 0.24673841159141466,
      "flos": 20002279213440.0,
      "grad_norm": 3.0285795556837947,
      "language_loss": 0.90587801,
      "learning_rate": 3.527293692493475e-06,
      "loss": 0.92798859,
      "num_input_tokens_seen": 43848345,
      "step": 2052,
      "time_per_iteration": 2.657773494720459
    },
    {
      "auxiliary_loss_clip": 0.01135619,
      "auxiliary_loss_mlp": 0.007654,
      "balance_loss_clip": 1.05051577,
      "balance_loss_mlp": 1.00026846,
      "epoch": 0.24685865448205374,
      "flos": 21646593037440.0,
      "grad_norm": 2.7337138555363776,
      "language_loss": 0.73456371,
      "learning_rate": 3.52679064594092e-06,
      "loss": 0.75357389,
      "num_input_tokens_seen": 43865685,
      "step": 2053,
      "time_per_iteration": 2.722748041152954
    },
    {
      "auxiliary_loss_clip": 0.01184006,
      "auxiliary_loss_mlp": 0.01041032,
      "balance_loss_clip": 1.06183815,
      "balance_loss_mlp": 1.03092337,
      "epoch": 0.24697889737269285,
      "flos": 17960470508160.0,
      "grad_norm": 2.2211887991727943,
      "language_loss": 0.7498014,
      "learning_rate": 3.5262873677754375e-06,
      "loss": 0.77205169,
      "num_input_tokens_seen": 43883690,
      "step": 2054,
      "time_per_iteration": 2.6993982791900635
    },
    {
      "auxiliary_loss_clip": 0.01170686,
      "auxiliary_loss_mlp": 0.01040606,
      "balance_loss_clip": 1.05422115,
      "balance_loss_mlp": 1.03027058,
      "epoch": 0.24709914026333193,
      "flos": 27344611221120.0,
      "grad_norm": 1.7205870365207094,
      "language_loss": 0.80339044,
      "learning_rate": 3.5257838580733745e-06,
      "loss": 0.82550341,
      "num_input_tokens_seen": 43903295,
      "step": 2055,
      "time_per_iteration": 2.8349170684814453
    },
    {
      "auxiliary_loss_clip": 0.01119588,
      "auxiliary_loss_mlp": 0.01035349,
      "balance_loss_clip": 1.04798102,
      "balance_loss_mlp": 1.02530015,
      "epoch": 0.24721938315397102,
      "flos": 19275519335040.0,
      "grad_norm": 5.828973230855534,
      "language_loss": 0.87378848,
      "learning_rate": 3.5252801169111138e-06,
      "loss": 0.89533782,
      "num_input_tokens_seen": 43920960,
      "step": 2056,
      "time_per_iteration": 3.711160182952881
    },
    {
      "auxiliary_loss_clip": 0.01151533,
      "auxiliary_loss_mlp": 0.01039395,
      "balance_loss_clip": 1.05910575,
      "balance_loss_mlp": 1.02872634,
      "epoch": 0.2473396260446101,
      "flos": 23185796688000.0,
      "grad_norm": 3.6760284938465095,
      "language_loss": 0.80009395,
      "learning_rate": 3.524776144365072e-06,
      "loss": 0.82200325,
      "num_input_tokens_seen": 43939415,
      "step": 2057,
      "time_per_iteration": 4.208636283874512
    },
    {
      "auxiliary_loss_clip": 0.01159322,
      "auxiliary_loss_mlp": 0.01032581,
      "balance_loss_clip": 1.05712676,
      "balance_loss_mlp": 1.02246642,
      "epoch": 0.2474598689352492,
      "flos": 21142443697920.0,
      "grad_norm": 1.5636558634266735,
      "language_loss": 0.79301703,
      "learning_rate": 3.5242719405117016e-06,
      "loss": 0.8149361,
      "num_input_tokens_seen": 43959220,
      "step": 2058,
      "time_per_iteration": 2.785003185272217
    },
    {
      "auxiliary_loss_clip": 0.01144479,
      "auxiliary_loss_mlp": 0.01032164,
      "balance_loss_clip": 1.04980445,
      "balance_loss_mlp": 1.02213204,
      "epoch": 0.2475801118258883,
      "flos": 21648352803840.0,
      "grad_norm": 2.624635137419916,
      "language_loss": 0.75095224,
      "learning_rate": 3.5237675054274893e-06,
      "loss": 0.77271867,
      "num_input_tokens_seen": 43978420,
      "step": 2059,
      "time_per_iteration": 3.629668951034546
    },
    {
      "auxiliary_loss_clip": 0.01198259,
      "auxiliary_loss_mlp": 0.01031733,
      "balance_loss_clip": 1.06150389,
      "balance_loss_mlp": 1.02088547,
      "epoch": 0.24770035471652738,
      "flos": 22674500542080.0,
      "grad_norm": 1.9576464350277742,
      "language_loss": 0.80401504,
      "learning_rate": 3.5232628391889584e-06,
      "loss": 0.82631493,
      "num_input_tokens_seen": 43996710,
      "step": 2060,
      "time_per_iteration": 3.6512932777404785
    },
    {
      "auxiliary_loss_clip": 0.01130327,
      "auxiliary_loss_mlp": 0.01038778,
      "balance_loss_clip": 1.04792893,
      "balance_loss_mlp": 1.02839494,
      "epoch": 0.2478205976071665,
      "flos": 22163814927360.0,
      "grad_norm": 2.726555413493,
      "language_loss": 0.64520139,
      "learning_rate": 3.522757941872666e-06,
      "loss": 0.66689241,
      "num_input_tokens_seen": 44014865,
      "step": 2061,
      "time_per_iteration": 2.744201183319092
    },
    {
      "auxiliary_loss_clip": 0.01143834,
      "auxiliary_loss_mlp": 0.01037733,
      "balance_loss_clip": 1.05766892,
      "balance_loss_mlp": 1.02689695,
      "epoch": 0.24794084049780557,
      "flos": 24973106555520.0,
      "grad_norm": 1.9606980836870551,
      "language_loss": 0.82479274,
      "learning_rate": 3.5222528135552042e-06,
      "loss": 0.84660852,
      "num_input_tokens_seen": 44036325,
      "step": 2062,
      "time_per_iteration": 2.835028886795044
    },
    {
      "auxiliary_loss_clip": 0.01197036,
      "auxiliary_loss_mlp": 0.01031571,
      "balance_loss_clip": 1.06366014,
      "balance_loss_mlp": 1.02130091,
      "epoch": 0.24806108338844465,
      "flos": 18296379521280.0,
      "grad_norm": 1.857752641590369,
      "language_loss": 0.80430394,
      "learning_rate": 3.521747454313201e-06,
      "loss": 0.82659,
      "num_input_tokens_seen": 44055005,
      "step": 2063,
      "time_per_iteration": 2.658966302871704
    },
    {
      "auxiliary_loss_clip": 0.01131933,
      "auxiliary_loss_mlp": 0.01028085,
      "balance_loss_clip": 1.05365992,
      "balance_loss_mlp": 1.01701641,
      "epoch": 0.24818132627908374,
      "flos": 19282163351040.0,
      "grad_norm": 2.036737882110295,
      "language_loss": 0.67220831,
      "learning_rate": 3.521241864223319e-06,
      "loss": 0.6938085,
      "num_input_tokens_seen": 44073965,
      "step": 2064,
      "time_per_iteration": 3.018651008605957
    },
    {
      "auxiliary_loss_clip": 0.01090071,
      "auxiliary_loss_mlp": 0.01004319,
      "balance_loss_clip": 1.02618182,
      "balance_loss_mlp": 1.00259078,
      "epoch": 0.24830156916972285,
      "flos": 70285837881600.0,
      "grad_norm": 0.8240252148696553,
      "language_loss": 0.61952698,
      "learning_rate": 3.5207360433622552e-06,
      "loss": 0.64047086,
      "num_input_tokens_seen": 44135965,
      "step": 2065,
      "time_per_iteration": 3.4406280517578125
    },
    {
      "auxiliary_loss_clip": 0.01119863,
      "auxiliary_loss_mlp": 0.01031367,
      "balance_loss_clip": 1.05019689,
      "balance_loss_mlp": 1.02141881,
      "epoch": 0.24842181206036193,
      "flos": 40409128287360.0,
      "grad_norm": 1.6718475726172815,
      "language_loss": 0.74895966,
      "learning_rate": 3.5202299918067437e-06,
      "loss": 0.77047193,
      "num_input_tokens_seen": 44159560,
      "step": 2066,
      "time_per_iteration": 2.9696030616760254
    },
    {
      "auxiliary_loss_clip": 0.01190341,
      "auxiliary_loss_mlp": 0.01040181,
      "balance_loss_clip": 1.0549196,
      "balance_loss_mlp": 1.02928483,
      "epoch": 0.248542054951001,
      "flos": 20082432412800.0,
      "grad_norm": 2.288750410597905,
      "language_loss": 0.69562542,
      "learning_rate": 3.519723709633551e-06,
      "loss": 0.71793061,
      "num_input_tokens_seen": 44178320,
      "step": 2067,
      "time_per_iteration": 2.699223041534424
    },
    {
      "auxiliary_loss_clip": 0.01157822,
      "auxiliary_loss_mlp": 0.01035604,
      "balance_loss_clip": 1.05415499,
      "balance_loss_mlp": 1.02572775,
      "epoch": 0.24866229784164012,
      "flos": 23513948363520.0,
      "grad_norm": 1.9571459294139644,
      "language_loss": 0.83854496,
      "learning_rate": 3.519217196919479e-06,
      "loss": 0.86047924,
      "num_input_tokens_seen": 44197305,
      "step": 2068,
      "time_per_iteration": 2.8285064697265625
    },
    {
      "auxiliary_loss_clip": 0.01163704,
      "auxiliary_loss_mlp": 0.01041911,
      "balance_loss_clip": 1.05431342,
      "balance_loss_mlp": 1.03092051,
      "epoch": 0.2487825407322792,
      "flos": 19865101173120.0,
      "grad_norm": 2.1387169157603134,
      "language_loss": 0.73847455,
      "learning_rate": 3.518710453741367e-06,
      "loss": 0.76053071,
      "num_input_tokens_seen": 44216505,
      "step": 2069,
      "time_per_iteration": 2.7208614349365234
    },
    {
      "auxiliary_loss_clip": 0.01183731,
      "auxiliary_loss_mlp": 0.01030374,
      "balance_loss_clip": 1.05459237,
      "balance_loss_mlp": 1.02045012,
      "epoch": 0.2489027836229183,
      "flos": 22017622573440.0,
      "grad_norm": 2.31569690021248,
      "language_loss": 0.68161845,
      "learning_rate": 3.518203480176086e-06,
      "loss": 0.70375949,
      "num_input_tokens_seen": 44235435,
      "step": 2070,
      "time_per_iteration": 2.765014410018921
    },
    {
      "auxiliary_loss_clip": 0.01074771,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.03532434,
      "balance_loss_mlp": 1.02166235,
      "epoch": 0.2490230265135574,
      "flos": 23294354567040.0,
      "grad_norm": 2.010662496726524,
      "language_loss": 0.80595839,
      "learning_rate": 3.517696276300545e-06,
      "loss": 0.82702386,
      "num_input_tokens_seen": 44256975,
      "step": 2071,
      "time_per_iteration": 3.1234538555145264
    },
    {
      "auxiliary_loss_clip": 0.01118954,
      "auxiliary_loss_mlp": 0.01037262,
      "balance_loss_clip": 1.05417418,
      "balance_loss_mlp": 1.02602077,
      "epoch": 0.24914326940419648,
      "flos": 19826784339840.0,
      "grad_norm": 2.6330088023096105,
      "language_loss": 0.6915133,
      "learning_rate": 3.517188842191685e-06,
      "loss": 0.7130754,
      "num_input_tokens_seen": 44275125,
      "step": 2072,
      "time_per_iteration": 4.192270994186401
    },
    {
      "auxiliary_loss_clip": 0.01125896,
      "auxiliary_loss_mlp": 0.01035816,
      "balance_loss_clip": 1.0504185,
      "balance_loss_mlp": 1.02522445,
      "epoch": 0.24926351229483557,
      "flos": 20229271211520.0,
      "grad_norm": 2.5380664457835964,
      "language_loss": 0.73881721,
      "learning_rate": 3.5166811779264837e-06,
      "loss": 0.76043433,
      "num_input_tokens_seen": 44295445,
      "step": 2073,
      "time_per_iteration": 3.4285717010498047
    },
    {
      "auxiliary_loss_clip": 0.01173322,
      "auxiliary_loss_mlp": 0.0103726,
      "balance_loss_clip": 1.05167258,
      "balance_loss_mlp": 1.02688837,
      "epoch": 0.24938375518547465,
      "flos": 23294570048640.0,
      "grad_norm": 1.9195941331546955,
      "language_loss": 0.78042424,
      "learning_rate": 3.5161732835819545e-06,
      "loss": 0.80253005,
      "num_input_tokens_seen": 44314755,
      "step": 2074,
      "time_per_iteration": 2.721787929534912
    },
    {
      "auxiliary_loss_clip": 0.01159915,
      "auxiliary_loss_mlp": 0.00764951,
      "balance_loss_clip": 1.06018591,
      "balance_loss_mlp": 1.00028002,
      "epoch": 0.24950399807611376,
      "flos": 17311673099520.0,
      "grad_norm": 2.1816624812789844,
      "language_loss": 0.83082408,
      "learning_rate": 3.515665159235143e-06,
      "loss": 0.8500728,
      "num_input_tokens_seen": 44333640,
      "step": 2075,
      "time_per_iteration": 2.8586654663085938
    },
    {
      "auxiliary_loss_clip": 0.0114007,
      "auxiliary_loss_mlp": 0.01028353,
      "balance_loss_clip": 1.05134606,
      "balance_loss_mlp": 1.01923954,
      "epoch": 0.24962424096675284,
      "flos": 19024863252480.0,
      "grad_norm": 1.6354509646727258,
      "language_loss": 0.74965268,
      "learning_rate": 3.5151568049631318e-06,
      "loss": 0.77133691,
      "num_input_tokens_seen": 44352355,
      "step": 2076,
      "time_per_iteration": 2.789933919906616
    },
    {
      "auxiliary_loss_clip": 0.01176641,
      "auxiliary_loss_mlp": 0.00764431,
      "balance_loss_clip": 1.05951393,
      "balance_loss_mlp": 1.00025594,
      "epoch": 0.24974448385739192,
      "flos": 33398790710400.0,
      "grad_norm": 1.8101256982235125,
      "language_loss": 0.80302346,
      "learning_rate": 3.5146482208430385e-06,
      "loss": 0.82243419,
      "num_input_tokens_seen": 44374185,
      "step": 2077,
      "time_per_iteration": 2.7921438217163086
    },
    {
      "auxiliary_loss_clip": 0.01158309,
      "auxiliary_loss_mlp": 0.01035809,
      "balance_loss_clip": 1.05304992,
      "balance_loss_mlp": 1.02390563,
      "epoch": 0.24986472674803104,
      "flos": 30007279532160.0,
      "grad_norm": 2.552545253965136,
      "language_loss": 0.6778084,
      "learning_rate": 3.514139406952014e-06,
      "loss": 0.69974953,
      "num_input_tokens_seen": 44396210,
      "step": 2078,
      "time_per_iteration": 2.7377235889434814
    },
    {
      "auxiliary_loss_clip": 0.0113313,
      "auxiliary_loss_mlp": 0.01031225,
      "balance_loss_clip": 1.05158627,
      "balance_loss_mlp": 1.020913,
      "epoch": 0.24998496963867012,
      "flos": 26613074833920.0,
      "grad_norm": 2.011090545184727,
      "language_loss": 0.83681178,
      "learning_rate": 3.5136303633672454e-06,
      "loss": 0.8584553,
      "num_input_tokens_seen": 44416340,
      "step": 2079,
      "time_per_iteration": 2.8542420864105225
    },
    {
      "auxiliary_loss_clip": 0.01152763,
      "auxiliary_loss_mlp": 0.01032102,
      "balance_loss_clip": 1.05383849,
      "balance_loss_mlp": 1.02230859,
      "epoch": 0.25010521252930923,
      "flos": 23553989049600.0,
      "grad_norm": 1.7728027183428572,
      "language_loss": 0.74485457,
      "learning_rate": 3.5131210901659544e-06,
      "loss": 0.76670319,
      "num_input_tokens_seen": 44438095,
      "step": 2080,
      "time_per_iteration": 2.8065638542175293
    },
    {
      "auxiliary_loss_clip": 0.01197399,
      "auxiliary_loss_mlp": 0.01035626,
      "balance_loss_clip": 1.05793178,
      "balance_loss_mlp": 1.0254277,
      "epoch": 0.2502254554199483,
      "flos": 23441193365760.0,
      "grad_norm": 2.346459000531074,
      "language_loss": 0.82055408,
      "learning_rate": 3.5126115874253967e-06,
      "loss": 0.84288436,
      "num_input_tokens_seen": 44457650,
      "step": 2081,
      "time_per_iteration": 2.676683187484741
    },
    {
      "auxiliary_loss_clip": 0.01138096,
      "auxiliary_loss_mlp": 0.01034765,
      "balance_loss_clip": 1.04938745,
      "balance_loss_mlp": 1.02437639,
      "epoch": 0.2503456983105874,
      "flos": 28761681651840.0,
      "grad_norm": 2.490476096203782,
      "language_loss": 0.80992419,
      "learning_rate": 3.5121018552228644e-06,
      "loss": 0.83165276,
      "num_input_tokens_seen": 44476155,
      "step": 2082,
      "time_per_iteration": 3.7767333984375
    },
    {
      "auxiliary_loss_clip": 0.01156989,
      "auxiliary_loss_mlp": 0.01035298,
      "balance_loss_clip": 1.0578413,
      "balance_loss_mlp": 1.02447426,
      "epoch": 0.2504659412012265,
      "flos": 18770256673920.0,
      "grad_norm": 1.9840393208111418,
      "language_loss": 0.76170915,
      "learning_rate": 3.5115918936356827e-06,
      "loss": 0.78363204,
      "num_input_tokens_seen": 44492910,
      "step": 2083,
      "time_per_iteration": 3.7355401515960693
    },
    {
      "auxiliary_loss_clip": 0.0115354,
      "auxiliary_loss_mlp": 0.0103545,
      "balance_loss_clip": 1.05533111,
      "balance_loss_mlp": 1.02526331,
      "epoch": 0.25058618409186556,
      "flos": 16873383346560.0,
      "grad_norm": 2.305490498167349,
      "language_loss": 0.78726721,
      "learning_rate": 3.5110817027412123e-06,
      "loss": 0.80915713,
      "num_input_tokens_seen": 44512000,
      "step": 2084,
      "time_per_iteration": 3.633739471435547
    },
    {
      "auxiliary_loss_clip": 0.01125883,
      "auxiliary_loss_mlp": 0.01030033,
      "balance_loss_clip": 1.05068529,
      "balance_loss_mlp": 1.01957214,
      "epoch": 0.25070642698250467,
      "flos": 24425540651520.0,
      "grad_norm": 2.2685143946082884,
      "language_loss": 0.69001526,
      "learning_rate": 3.5105712826168493e-06,
      "loss": 0.71157444,
      "num_input_tokens_seen": 44531650,
      "step": 2085,
      "time_per_iteration": 3.804171085357666
    },
    {
      "auxiliary_loss_clip": 0.01189437,
      "auxiliary_loss_mlp": 0.01037008,
      "balance_loss_clip": 1.0591675,
      "balance_loss_mlp": 1.02644014,
      "epoch": 0.2508266698731437,
      "flos": 20260944028800.0,
      "grad_norm": 1.9084006713832595,
      "language_loss": 0.7073651,
      "learning_rate": 3.5100606333400235e-06,
      "loss": 0.72962958,
      "num_input_tokens_seen": 44548785,
      "step": 2086,
      "time_per_iteration": 2.646765947341919
    },
    {
      "auxiliary_loss_clip": 0.01158448,
      "auxiliary_loss_mlp": 0.0104245,
      "balance_loss_clip": 1.05862582,
      "balance_loss_mlp": 1.02986109,
      "epoch": 0.25094691276378284,
      "flos": 19245318975360.0,
      "grad_norm": 1.9967871298245192,
      "language_loss": 0.77074099,
      "learning_rate": 3.5095497549882006e-06,
      "loss": 0.79275,
      "num_input_tokens_seen": 44567230,
      "step": 2087,
      "time_per_iteration": 2.7812180519104004
    },
    {
      "auxiliary_loss_clip": 0.01126609,
      "auxiliary_loss_mlp": 0.01030085,
      "balance_loss_clip": 1.06035542,
      "balance_loss_mlp": 1.01920676,
      "epoch": 0.25106715565442195,
      "flos": 26943237671040.0,
      "grad_norm": 2.327546806865252,
      "language_loss": 0.72582471,
      "learning_rate": 3.50903864763888e-06,
      "loss": 0.7473917,
      "num_input_tokens_seen": 44588020,
      "step": 2088,
      "time_per_iteration": 2.94746470451355
    },
    {
      "auxiliary_loss_clip": 0.01183107,
      "auxiliary_loss_mlp": 0.01032,
      "balance_loss_clip": 1.05944657,
      "balance_loss_mlp": 1.02134824,
      "epoch": 0.251187398545061,
      "flos": 48359570572800.0,
      "grad_norm": 2.58334666338619,
      "language_loss": 0.76205182,
      "learning_rate": 3.5085273113695965e-06,
      "loss": 0.78420281,
      "num_input_tokens_seen": 44612590,
      "step": 2089,
      "time_per_iteration": 3.2469213008880615
    },
    {
      "auxiliary_loss_clip": 0.01184359,
      "auxiliary_loss_mlp": 0.01037475,
      "balance_loss_clip": 1.05836904,
      "balance_loss_mlp": 1.02713919,
      "epoch": 0.2513076414357001,
      "flos": 27016100409600.0,
      "grad_norm": 2.8628762999363806,
      "language_loss": 0.78540182,
      "learning_rate": 3.508015746257919e-06,
      "loss": 0.80762017,
      "num_input_tokens_seen": 44631630,
      "step": 2090,
      "time_per_iteration": 2.8011605739593506
    },
    {
      "auxiliary_loss_clip": 0.01103229,
      "auxiliary_loss_mlp": 0.00765511,
      "balance_loss_clip": 1.04892468,
      "balance_loss_mlp": 1.00022578,
      "epoch": 0.2514278843263392,
      "flos": 19463619882240.0,
      "grad_norm": 1.969532244775705,
      "language_loss": 0.82992512,
      "learning_rate": 3.5075039523814518e-06,
      "loss": 0.84861255,
      "num_input_tokens_seen": 44650820,
      "step": 2091,
      "time_per_iteration": 2.965684413909912
    },
    {
      "auxiliary_loss_clip": 0.01144076,
      "auxiliary_loss_mlp": 0.01034879,
      "balance_loss_clip": 1.05123389,
      "balance_loss_mlp": 1.02390563,
      "epoch": 0.2515481272169783,
      "flos": 16866092885760.0,
      "grad_norm": 2.023491739304695,
      "language_loss": 0.82052201,
      "learning_rate": 3.506991929817834e-06,
      "loss": 0.84231156,
      "num_input_tokens_seen": 44667540,
      "step": 2092,
      "time_per_iteration": 3.3593342304229736
    },
    {
      "auxiliary_loss_clip": 0.01169629,
      "auxiliary_loss_mlp": 0.01039978,
      "balance_loss_clip": 1.05544567,
      "balance_loss_mlp": 1.02948749,
      "epoch": 0.2516683701076174,
      "flos": 23732464752000.0,
      "grad_norm": 1.7715223560258173,
      "language_loss": 0.82743084,
      "learning_rate": 3.506479678644738e-06,
      "loss": 0.84952688,
      "num_input_tokens_seen": 44687935,
      "step": 2093,
      "time_per_iteration": 2.783951997756958
    },
    {
      "auxiliary_loss_clip": 0.01181691,
      "auxiliary_loss_mlp": 0.0103592,
      "balance_loss_clip": 1.05485213,
      "balance_loss_mlp": 1.02569151,
      "epoch": 0.2517886129982565,
      "flos": 27635954434560.0,
      "grad_norm": 2.5971226466767967,
      "language_loss": 0.74248683,
      "learning_rate": 3.505967198939873e-06,
      "loss": 0.76466298,
      "num_input_tokens_seen": 44704975,
      "step": 2094,
      "time_per_iteration": 2.722266435623169
    },
    {
      "auxiliary_loss_clip": 0.01187626,
      "auxiliary_loss_mlp": 0.01034842,
      "balance_loss_clip": 1.05551088,
      "balance_loss_mlp": 1.02488232,
      "epoch": 0.25190885588889556,
      "flos": 38104596529920.0,
      "grad_norm": 1.8815809305291924,
      "language_loss": 0.78266823,
      "learning_rate": 3.5054544907809813e-06,
      "loss": 0.8048929,
      "num_input_tokens_seen": 44725475,
      "step": 2095,
      "time_per_iteration": 2.8457040786743164
    },
    {
      "auxiliary_loss_clip": 0.01195306,
      "auxiliary_loss_mlp": 0.01037191,
      "balance_loss_clip": 1.06247044,
      "balance_loss_mlp": 1.02623594,
      "epoch": 0.25202909877953467,
      "flos": 22269894768000.0,
      "grad_norm": 2.090474597741467,
      "language_loss": 0.80422705,
      "learning_rate": 3.50494155424584e-06,
      "loss": 0.82655203,
      "num_input_tokens_seen": 44744380,
      "step": 2096,
      "time_per_iteration": 2.6769134998321533
    },
    {
      "auxiliary_loss_clip": 0.01154816,
      "auxiliary_loss_mlp": 0.01029562,
      "balance_loss_clip": 1.05094671,
      "balance_loss_mlp": 1.01897061,
      "epoch": 0.2521493416701738,
      "flos": 21761759018880.0,
      "grad_norm": 6.000297487881914,
      "language_loss": 0.83188426,
      "learning_rate": 3.504428389412262e-06,
      "loss": 0.853728,
      "num_input_tokens_seen": 44765190,
      "step": 2097,
      "time_per_iteration": 2.756481647491455
    },
    {
      "auxiliary_loss_clip": 0.01157388,
      "auxiliary_loss_mlp": 0.01037406,
      "balance_loss_clip": 1.05368853,
      "balance_loss_mlp": 1.02639091,
      "epoch": 0.25226958456081283,
      "flos": 27746738956800.0,
      "grad_norm": 2.2115749705851977,
      "language_loss": 0.72791058,
      "learning_rate": 3.5039149963580927e-06,
      "loss": 0.7498585,
      "num_input_tokens_seen": 44785210,
      "step": 2098,
      "time_per_iteration": 2.718316078186035
    },
    {
      "auxiliary_loss_clip": 0.01172508,
      "auxiliary_loss_mlp": 0.01033637,
      "balance_loss_clip": 1.05364132,
      "balance_loss_mlp": 1.02342081,
      "epoch": 0.25238982745145194,
      "flos": 30732171903360.0,
      "grad_norm": 2.1024596011830283,
      "language_loss": 0.7009716,
      "learning_rate": 3.503401375161215e-06,
      "loss": 0.72303307,
      "num_input_tokens_seen": 44804955,
      "step": 2099,
      "time_per_iteration": 2.813511848449707
    },
    {
      "auxiliary_loss_clip": 0.01132189,
      "auxiliary_loss_mlp": 0.0103352,
      "balance_loss_clip": 1.04900503,
      "balance_loss_mlp": 1.02312529,
      "epoch": 0.252510070342091,
      "flos": 20266331068800.0,
      "grad_norm": 1.788184025870816,
      "language_loss": 0.83867079,
      "learning_rate": 3.502887525899544e-06,
      "loss": 0.86032784,
      "num_input_tokens_seen": 44823935,
      "step": 2100,
      "time_per_iteration": 2.80049729347229
    },
    {
      "auxiliary_loss_clip": 0.01164425,
      "auxiliary_loss_mlp": 0.0103814,
      "balance_loss_clip": 1.05418921,
      "balance_loss_mlp": 1.02735722,
      "epoch": 0.2526303132327301,
      "flos": 22747399194240.0,
      "grad_norm": 1.932608758577853,
      "language_loss": 0.82860857,
      "learning_rate": 3.50237344865103e-06,
      "loss": 0.85063422,
      "num_input_tokens_seen": 44844935,
      "step": 2101,
      "time_per_iteration": 2.731322765350342
    },
    {
      "auxiliary_loss_clip": 0.01182897,
      "auxiliary_loss_mlp": 0.01036037,
      "balance_loss_clip": 1.06053007,
      "balance_loss_mlp": 1.02562451,
      "epoch": 0.2527505561233692,
      "flos": 30263466309120.0,
      "grad_norm": 3.3240206115984976,
      "language_loss": 0.76135945,
      "learning_rate": 3.501859143493658e-06,
      "loss": 0.78354883,
      "num_input_tokens_seen": 44865565,
      "step": 2102,
      "time_per_iteration": 2.8522756099700928
    },
    {
      "auxiliary_loss_clip": 0.01076814,
      "auxiliary_loss_mlp": 0.01004591,
      "balance_loss_clip": 1.02699804,
      "balance_loss_mlp": 1.0030297,
      "epoch": 0.2528707990140083,
      "flos": 58492917164160.0,
      "grad_norm": 0.9211190650212024,
      "language_loss": 0.60517955,
      "learning_rate": 3.5013446105054488e-06,
      "loss": 0.62599361,
      "num_input_tokens_seen": 44918485,
      "step": 2103,
      "time_per_iteration": 3.0368785858154297
    },
    {
      "auxiliary_loss_clip": 0.01156832,
      "auxiliary_loss_mlp": 0.01035667,
      "balance_loss_clip": 1.04875517,
      "balance_loss_mlp": 1.02596903,
      "epoch": 0.2529910419046474,
      "flos": 24645134448000.0,
      "grad_norm": 1.6477545156886504,
      "language_loss": 0.74517131,
      "learning_rate": 3.5008298497644555e-06,
      "loss": 0.76709634,
      "num_input_tokens_seen": 44937530,
      "step": 2104,
      "time_per_iteration": 2.7598378658294678
    },
    {
      "auxiliary_loss_clip": 0.01155994,
      "auxiliary_loss_mlp": 0.00764253,
      "balance_loss_clip": 1.05241787,
      "balance_loss_mlp": 1.0001924,
      "epoch": 0.2531112847952865,
      "flos": 23842135952640.0,
      "grad_norm": 1.8807764173457175,
      "language_loss": 0.87779188,
      "learning_rate": 3.500314861348767e-06,
      "loss": 0.89699447,
      "num_input_tokens_seen": 44958165,
      "step": 2105,
      "time_per_iteration": 2.744337558746338
    },
    {
      "auxiliary_loss_clip": 0.01175762,
      "auxiliary_loss_mlp": 0.01036022,
      "balance_loss_clip": 1.0552392,
      "balance_loss_mlp": 1.02568078,
      "epoch": 0.25323152768592555,
      "flos": 16143822207360.0,
      "grad_norm": 2.088298810847765,
      "language_loss": 0.77113152,
      "learning_rate": 3.499799645336507e-06,
      "loss": 0.79324937,
      "num_input_tokens_seen": 44975060,
      "step": 2106,
      "time_per_iteration": 2.7067208290100098
    },
    {
      "auxiliary_loss_clip": 0.01100827,
      "auxiliary_loss_mlp": 0.01032397,
      "balance_loss_clip": 1.0489434,
      "balance_loss_mlp": 1.0228368,
      "epoch": 0.25335177057656466,
      "flos": 28405161210240.0,
      "grad_norm": 1.6220883119832763,
      "language_loss": 0.86625385,
      "learning_rate": 3.4992842018058336e-06,
      "loss": 0.887586,
      "num_input_tokens_seen": 44997960,
      "step": 2107,
      "time_per_iteration": 3.9397335052490234
    },
    {
      "auxiliary_loss_clip": 0.01160193,
      "auxiliary_loss_mlp": 0.01034835,
      "balance_loss_clip": 1.05860376,
      "balance_loss_mlp": 1.02476764,
      "epoch": 0.25347201346720377,
      "flos": 18799666934400.0,
      "grad_norm": 1.9854994091038756,
      "language_loss": 0.88571858,
      "learning_rate": 3.4987685308349384e-06,
      "loss": 0.90766889,
      "num_input_tokens_seen": 45015690,
      "step": 2108,
      "time_per_iteration": 4.005795478820801
    },
    {
      "auxiliary_loss_clip": 0.01194843,
      "auxiliary_loss_mlp": 0.0103336,
      "balance_loss_clip": 1.05729103,
      "balance_loss_mlp": 1.02340627,
      "epoch": 0.2535922563578428,
      "flos": 15815490963840.0,
      "grad_norm": 2.1411074770358876,
      "language_loss": 0.61017931,
      "learning_rate": 3.4982526325020497e-06,
      "loss": 0.63246137,
      "num_input_tokens_seen": 45032660,
      "step": 2109,
      "time_per_iteration": 2.6498348712921143
    },
    {
      "auxiliary_loss_clip": 0.01137551,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.04641795,
      "balance_loss_mlp": 1.01887059,
      "epoch": 0.25371249924848194,
      "flos": 16318922031360.0,
      "grad_norm": 2.5866159776767237,
      "language_loss": 0.82388663,
      "learning_rate": 3.4977365068854273e-06,
      "loss": 0.84555542,
      "num_input_tokens_seen": 45048280,
      "step": 2110,
      "time_per_iteration": 3.643758535385132
    },
    {
      "auxiliary_loss_clip": 0.01194895,
      "auxiliary_loss_mlp": 0.01035919,
      "balance_loss_clip": 1.05742133,
      "balance_loss_mlp": 1.02539277,
      "epoch": 0.25383274213912105,
      "flos": 21761615364480.0,
      "grad_norm": 1.7210087613761726,
      "language_loss": 0.73536098,
      "learning_rate": 3.4972201540633676e-06,
      "loss": 0.75766909,
      "num_input_tokens_seen": 45067635,
      "step": 2111,
      "time_per_iteration": 2.6638166904449463
    },
    {
      "auxiliary_loss_clip": 0.01148804,
      "auxiliary_loss_mlp": 0.00764636,
      "balance_loss_clip": 1.04799008,
      "balance_loss_mlp": 1.00020599,
      "epoch": 0.2539529850297601,
      "flos": 21396870708480.0,
      "grad_norm": 1.7768388525975798,
      "language_loss": 0.85337842,
      "learning_rate": 3.4967035741142008e-06,
      "loss": 0.87251282,
      "num_input_tokens_seen": 45086455,
      "step": 2112,
      "time_per_iteration": 3.742396831512451
    },
    {
      "auxiliary_loss_clip": 0.01189411,
      "auxiliary_loss_mlp": 0.0103019,
      "balance_loss_clip": 1.05766821,
      "balance_loss_mlp": 1.02063537,
      "epoch": 0.2540732279203992,
      "flos": 25228467319680.0,
      "grad_norm": 1.7625483138213665,
      "language_loss": 0.81819558,
      "learning_rate": 3.4961867671162917e-06,
      "loss": 0.84039158,
      "num_input_tokens_seen": 45106385,
      "step": 2113,
      "time_per_iteration": 2.704379081726074
    },
    {
      "auxiliary_loss_clip": 0.01117619,
      "auxiliary_loss_mlp": 0.0103121,
      "balance_loss_clip": 1.04885793,
      "balance_loss_mlp": 1.02036774,
      "epoch": 0.2541934708110383,
      "flos": 19427386037760.0,
      "grad_norm": 3.469549651039737,
      "language_loss": 0.7720055,
      "learning_rate": 3.4956697331480402e-06,
      "loss": 0.79349381,
      "num_input_tokens_seen": 45124955,
      "step": 2114,
      "time_per_iteration": 2.802192211151123
    },
    {
      "auxiliary_loss_clip": 0.01166442,
      "auxiliary_loss_mlp": 0.01037413,
      "balance_loss_clip": 1.05259871,
      "balance_loss_mlp": 1.02713072,
      "epoch": 0.2543137137016774,
      "flos": 23949436855680.0,
      "grad_norm": 2.4111335930785134,
      "language_loss": 0.80063832,
      "learning_rate": 3.495152472287879e-06,
      "loss": 0.8226769,
      "num_input_tokens_seen": 45145665,
      "step": 2115,
      "time_per_iteration": 2.777148485183716
    },
    {
      "auxiliary_loss_clip": 0.0117794,
      "auxiliary_loss_mlp": 0.01032109,
      "balance_loss_clip": 1.05280864,
      "balance_loss_mlp": 1.02157676,
      "epoch": 0.2544339565923165,
      "flos": 25593283802880.0,
      "grad_norm": 1.929349997280376,
      "language_loss": 0.73883605,
      "learning_rate": 3.4946349846142766e-06,
      "loss": 0.76093656,
      "num_input_tokens_seen": 45164805,
      "step": 2116,
      "time_per_iteration": 2.7473111152648926
    },
    {
      "auxiliary_loss_clip": 0.01176399,
      "auxiliary_loss_mlp": 0.01029188,
      "balance_loss_clip": 1.05234051,
      "balance_loss_mlp": 1.01952577,
      "epoch": 0.25455419948295555,
      "flos": 21689470897920.0,
      "grad_norm": 2.1622659142569303,
      "language_loss": 0.75949979,
      "learning_rate": 3.4941172702057353e-06,
      "loss": 0.78155565,
      "num_input_tokens_seen": 45184865,
      "step": 2117,
      "time_per_iteration": 2.6743178367614746
    },
    {
      "auxiliary_loss_clip": 0.01171023,
      "auxiliary_loss_mlp": 0.01029543,
      "balance_loss_clip": 1.05576336,
      "balance_loss_mlp": 1.01983368,
      "epoch": 0.25467444237359466,
      "flos": 26250341339520.0,
      "grad_norm": 2.0066166988817904,
      "language_loss": 0.80399871,
      "learning_rate": 3.4935993291407924e-06,
      "loss": 0.82600439,
      "num_input_tokens_seen": 45203690,
      "step": 2118,
      "time_per_iteration": 2.7740283012390137
    },
    {
      "auxiliary_loss_clip": 0.01163368,
      "auxiliary_loss_mlp": 0.01032348,
      "balance_loss_clip": 1.05514324,
      "balance_loss_mlp": 1.02197695,
      "epoch": 0.25479468526423377,
      "flos": 26979686997120.0,
      "grad_norm": 2.0970630815425597,
      "language_loss": 0.70817065,
      "learning_rate": 3.4930811614980183e-06,
      "loss": 0.73012781,
      "num_input_tokens_seen": 45225385,
      "step": 2119,
      "time_per_iteration": 2.7485411167144775
    },
    {
      "auxiliary_loss_clip": 0.01118816,
      "auxiliary_loss_mlp": 0.01036087,
      "balance_loss_clip": 1.05091166,
      "balance_loss_mlp": 1.02606118,
      "epoch": 0.2549149281548728,
      "flos": 23475811098240.0,
      "grad_norm": 1.990933892860569,
      "language_loss": 0.7917316,
      "learning_rate": 3.4925627673560198e-06,
      "loss": 0.81328058,
      "num_input_tokens_seen": 45246045,
      "step": 2120,
      "time_per_iteration": 2.832386016845703
    },
    {
      "auxiliary_loss_clip": 0.01171844,
      "auxiliary_loss_mlp": 0.01030598,
      "balance_loss_clip": 1.05627227,
      "balance_loss_mlp": 1.02101362,
      "epoch": 0.25503517104551193,
      "flos": 25812302981760.0,
      "grad_norm": 2.0292731743032477,
      "language_loss": 0.88394177,
      "learning_rate": 3.4920441467934357e-06,
      "loss": 0.90596622,
      "num_input_tokens_seen": 45266560,
      "step": 2121,
      "time_per_iteration": 2.7687442302703857
    },
    {
      "auxiliary_loss_clip": 0.01202614,
      "auxiliary_loss_mlp": 0.01034925,
      "balance_loss_clip": 1.05740619,
      "balance_loss_mlp": 1.02514958,
      "epoch": 0.25515541393615104,
      "flos": 26645106787200.0,
      "grad_norm": 2.413446246046313,
      "language_loss": 0.82729542,
      "learning_rate": 3.491525299888941e-06,
      "loss": 0.84967077,
      "num_input_tokens_seen": 45285405,
      "step": 2122,
      "time_per_iteration": 2.743124008178711
    },
    {
      "auxiliary_loss_clip": 0.01023983,
      "auxiliary_loss_mlp": 0.01001976,
      "balance_loss_clip": 1.02285433,
      "balance_loss_mlp": 1.00023556,
      "epoch": 0.2552756568267901,
      "flos": 65955945847680.0,
      "grad_norm": 0.8846093497243885,
      "language_loss": 0.62614542,
      "learning_rate": 3.491006226721244e-06,
      "loss": 0.64640504,
      "num_input_tokens_seen": 45349615,
      "step": 2123,
      "time_per_iteration": 3.847766637802124
    },
    {
      "auxiliary_loss_clip": 0.01162703,
      "auxiliary_loss_mlp": 0.01035311,
      "balance_loss_clip": 1.05522072,
      "balance_loss_mlp": 1.02482641,
      "epoch": 0.2553958997174292,
      "flos": 17931096161280.0,
      "grad_norm": 2.151275635220423,
      "language_loss": 0.77676749,
      "learning_rate": 3.4904869273690882e-06,
      "loss": 0.79874766,
      "num_input_tokens_seen": 45367505,
      "step": 2124,
      "time_per_iteration": 3.1046557426452637
    },
    {
      "auxiliary_loss_clip": 0.0117297,
      "auxiliary_loss_mlp": 0.01032608,
      "balance_loss_clip": 1.05342352,
      "balance_loss_mlp": 1.02254117,
      "epoch": 0.2555161426080683,
      "flos": 23367791923200.0,
      "grad_norm": 1.9007032801777497,
      "language_loss": 0.8875674,
      "learning_rate": 3.489967401911251e-06,
      "loss": 0.90962315,
      "num_input_tokens_seen": 45386805,
      "step": 2125,
      "time_per_iteration": 2.715838670730591
    },
    {
      "auxiliary_loss_clip": 0.01161543,
      "auxiliary_loss_mlp": 0.01034447,
      "balance_loss_clip": 1.05332863,
      "balance_loss_mlp": 1.02373052,
      "epoch": 0.2556363854987074,
      "flos": 40625130723840.0,
      "grad_norm": 1.6292905365094703,
      "language_loss": 0.69306356,
      "learning_rate": 3.4894476504265428e-06,
      "loss": 0.71502346,
      "num_input_tokens_seen": 45411045,
      "step": 2126,
      "time_per_iteration": 2.8380868434906006
    },
    {
      "auxiliary_loss_clip": 0.01072038,
      "auxiliary_loss_mlp": 0.01002367,
      "balance_loss_clip": 1.02432752,
      "balance_loss_mlp": 1.00067413,
      "epoch": 0.2557566283893465,
      "flos": 68019443389440.0,
      "grad_norm": 0.7521751347028983,
      "language_loss": 0.54508579,
      "learning_rate": 3.4889276729938104e-06,
      "loss": 0.56582981,
      "num_input_tokens_seen": 45469575,
      "step": 2127,
      "time_per_iteration": 3.1803109645843506
    },
    {
      "auxiliary_loss_clip": 0.01205723,
      "auxiliary_loss_mlp": 0.01034428,
      "balance_loss_clip": 1.05816031,
      "balance_loss_mlp": 1.02425885,
      "epoch": 0.2558768712799856,
      "flos": 22635645004800.0,
      "grad_norm": 2.6237619598546154,
      "language_loss": 0.80192739,
      "learning_rate": 3.488407469691934e-06,
      "loss": 0.8243289,
      "num_input_tokens_seen": 45490270,
      "step": 2128,
      "time_per_iteration": 2.607384204864502
    },
    {
      "auxiliary_loss_clip": 0.01098409,
      "auxiliary_loss_mlp": 0.01043899,
      "balance_loss_clip": 1.04676247,
      "balance_loss_mlp": 1.03206158,
      "epoch": 0.25599711417062465,
      "flos": 26396354125440.0,
      "grad_norm": 1.9077063324478059,
      "language_loss": 0.80816996,
      "learning_rate": 3.487887040599828e-06,
      "loss": 0.82959306,
      "num_input_tokens_seen": 45510070,
      "step": 2129,
      "time_per_iteration": 2.9032680988311768
    },
    {
      "auxiliary_loss_clip": 0.01138561,
      "auxiliary_loss_mlp": 0.01029535,
      "balance_loss_clip": 1.0546,
      "balance_loss_mlp": 1.01998639,
      "epoch": 0.25611735706126376,
      "flos": 22852042490880.0,
      "grad_norm": 2.514673675203185,
      "language_loss": 0.76092327,
      "learning_rate": 3.4873663857964407e-06,
      "loss": 0.78260422,
      "num_input_tokens_seen": 45527285,
      "step": 2130,
      "time_per_iteration": 2.792269468307495
    },
    {
      "auxiliary_loss_clip": 0.01204362,
      "auxiliary_loss_mlp": 0.00764014,
      "balance_loss_clip": 1.05774438,
      "balance_loss_mlp": 1.00019979,
      "epoch": 0.2562375999519028,
      "flos": 23367863750400.0,
      "grad_norm": 1.8440210753392077,
      "language_loss": 0.668594,
      "learning_rate": 3.4868455053607556e-06,
      "loss": 0.68827772,
      "num_input_tokens_seen": 45546900,
      "step": 2131,
      "time_per_iteration": 2.6818788051605225
    },
    {
      "auxiliary_loss_clip": 0.0117454,
      "auxiliary_loss_mlp": 0.01041282,
      "balance_loss_clip": 1.0515306,
      "balance_loss_mlp": 1.03001094,
      "epoch": 0.2563578428425419,
      "flos": 22856962654080.0,
      "grad_norm": 4.733005806762767,
      "language_loss": 0.71460164,
      "learning_rate": 3.486324399371789e-06,
      "loss": 0.7367599,
      "num_input_tokens_seen": 45566200,
      "step": 2132,
      "time_per_iteration": 2.699817180633545
    },
    {
      "auxiliary_loss_clip": 0.01165393,
      "auxiliary_loss_mlp": 0.0103304,
      "balance_loss_clip": 1.05679131,
      "balance_loss_mlp": 1.02320504,
      "epoch": 0.25647808573318104,
      "flos": 21653883498240.0,
      "grad_norm": 2.0445871042029355,
      "language_loss": 0.78781301,
      "learning_rate": 3.485803067908593e-06,
      "loss": 0.80979735,
      "num_input_tokens_seen": 45585710,
      "step": 2133,
      "time_per_iteration": 4.398736000061035
    },
    {
      "auxiliary_loss_clip": 0.01163714,
      "auxiliary_loss_mlp": 0.01029373,
      "balance_loss_clip": 1.05255246,
      "balance_loss_mlp": 1.01884079,
      "epoch": 0.2565983286238201,
      "flos": 33730569659520.0,
      "grad_norm": 1.7789964623410492,
      "language_loss": 0.79737985,
      "learning_rate": 3.485281511050253e-06,
      "loss": 0.81931072,
      "num_input_tokens_seen": 45607845,
      "step": 2134,
      "time_per_iteration": 3.967287063598633
    },
    {
      "auxiliary_loss_clip": 0.01174258,
      "auxiliary_loss_mlp": 0.01033027,
      "balance_loss_clip": 1.05387175,
      "balance_loss_mlp": 1.02358592,
      "epoch": 0.2567185715144592,
      "flos": 16216002587520.0,
      "grad_norm": 2.3868787861102407,
      "language_loss": 0.8967036,
      "learning_rate": 3.484759728875889e-06,
      "loss": 0.91877651,
      "num_input_tokens_seen": 45623210,
      "step": 2135,
      "time_per_iteration": 2.763861894607544
    },
    {
      "auxiliary_loss_clip": 0.0115532,
      "auxiliary_loss_mlp": 0.00763977,
      "balance_loss_clip": 1.0538851,
      "balance_loss_mlp": 1.00019431,
      "epoch": 0.2568388144050983,
      "flos": 17458475984640.0,
      "grad_norm": 1.7778240291886245,
      "language_loss": 0.80848581,
      "learning_rate": 3.4842377214646543e-06,
      "loss": 0.8276788,
      "num_input_tokens_seen": 45641505,
      "step": 2136,
      "time_per_iteration": 3.658735752105713
    },
    {
      "auxiliary_loss_clip": 0.01174998,
      "auxiliary_loss_mlp": 0.01030382,
      "balance_loss_clip": 1.06040847,
      "balance_loss_mlp": 1.02086282,
      "epoch": 0.25695905729573737,
      "flos": 20887442069760.0,
      "grad_norm": 1.7736303176752455,
      "language_loss": 0.66392481,
      "learning_rate": 3.483715488895737e-06,
      "loss": 0.68597859,
      "num_input_tokens_seen": 45661835,
      "step": 2137,
      "time_per_iteration": 3.7986907958984375
    },
    {
      "auxiliary_loss_clip": 0.01173329,
      "auxiliary_loss_mlp": 0.01029845,
      "balance_loss_clip": 1.05481911,
      "balance_loss_mlp": 1.01970029,
      "epoch": 0.2570793001863765,
      "flos": 24717278914560.0,
      "grad_norm": 1.8866970450020775,
      "language_loss": 0.78913313,
      "learning_rate": 3.48319303124836e-06,
      "loss": 0.81116486,
      "num_input_tokens_seen": 45682215,
      "step": 2138,
      "time_per_iteration": 2.8153932094573975
    },
    {
      "auxiliary_loss_clip": 0.01182537,
      "auxiliary_loss_mlp": 0.01032506,
      "balance_loss_clip": 1.05419064,
      "balance_loss_mlp": 1.02225375,
      "epoch": 0.2571995430770156,
      "flos": 26906896085760.0,
      "grad_norm": 2.3542602708852503,
      "language_loss": 0.67497921,
      "learning_rate": 3.4826703486017798e-06,
      "loss": 0.69712967,
      "num_input_tokens_seen": 45701840,
      "step": 2139,
      "time_per_iteration": 2.7422773838043213
    },
    {
      "auxiliary_loss_clip": 0.01153612,
      "auxiliary_loss_mlp": 0.01037499,
      "balance_loss_clip": 1.05572224,
      "balance_loss_mlp": 1.02724099,
      "epoch": 0.25731978596765465,
      "flos": 19792561656960.0,
      "grad_norm": 1.6879005549689547,
      "language_loss": 0.76865417,
      "learning_rate": 3.4821474410352867e-06,
      "loss": 0.79056525,
      "num_input_tokens_seen": 45720500,
      "step": 2140,
      "time_per_iteration": 2.739124298095703
    },
    {
      "auxiliary_loss_clip": 0.01076811,
      "auxiliary_loss_mlp": 0.01001714,
      "balance_loss_clip": 1.02614951,
      "balance_loss_mlp": 1.00022984,
      "epoch": 0.25744002885829376,
      "flos": 70564970471040.0,
      "grad_norm": 0.8975491762237257,
      "language_loss": 0.62638247,
      "learning_rate": 3.481624308628205e-06,
      "loss": 0.64716774,
      "num_input_tokens_seen": 45781870,
      "step": 2141,
      "time_per_iteration": 3.5241892337799072
    },
    {
      "auxiliary_loss_clip": 0.01127274,
      "auxiliary_loss_mlp": 0.01033928,
      "balance_loss_clip": 1.0502553,
      "balance_loss_mlp": 1.02363479,
      "epoch": 0.25756027174893287,
      "flos": 18038181582720.0,
      "grad_norm": 3.1796488442592343,
      "language_loss": 1.00251245,
      "learning_rate": 3.481100951459893e-06,
      "loss": 1.02412462,
      "num_input_tokens_seen": 45794890,
      "step": 2142,
      "time_per_iteration": 3.142296314239502
    },
    {
      "auxiliary_loss_clip": 0.01159311,
      "auxiliary_loss_mlp": 0.01030306,
      "balance_loss_clip": 1.05178785,
      "balance_loss_mlp": 1.02063203,
      "epoch": 0.2576805146395719,
      "flos": 22674069578880.0,
      "grad_norm": 1.5930501616937678,
      "language_loss": 0.7877357,
      "learning_rate": 3.4805773696097453e-06,
      "loss": 0.80963182,
      "num_input_tokens_seen": 45815780,
      "step": 2143,
      "time_per_iteration": 2.739164113998413
    },
    {
      "auxiliary_loss_clip": 0.01153399,
      "auxiliary_loss_mlp": 0.01032864,
      "balance_loss_clip": 1.05728257,
      "balance_loss_mlp": 1.02321398,
      "epoch": 0.25780075753021103,
      "flos": 16472225278080.0,
      "grad_norm": 2.057457412509148,
      "language_loss": 0.8792215,
      "learning_rate": 3.4800535631571874e-06,
      "loss": 0.90108418,
      "num_input_tokens_seen": 45831310,
      "step": 2144,
      "time_per_iteration": 2.726072072982788
    },
    {
      "auxiliary_loss_clip": 0.011718,
      "auxiliary_loss_mlp": 0.01041021,
      "balance_loss_clip": 1.05543268,
      "balance_loss_mlp": 1.03023863,
      "epoch": 0.25792100042085014,
      "flos": 22820297846400.0,
      "grad_norm": 2.521372503662837,
      "language_loss": 0.76429999,
      "learning_rate": 3.4795295321816804e-06,
      "loss": 0.78642821,
      "num_input_tokens_seen": 45850135,
      "step": 2145,
      "time_per_iteration": 2.767190456390381
    },
    {
      "auxiliary_loss_clip": 0.01119686,
      "auxiliary_loss_mlp": 0.01031824,
      "balance_loss_clip": 1.04623532,
      "balance_loss_mlp": 1.02185178,
      "epoch": 0.2580412433114892,
      "flos": 18697286194560.0,
      "grad_norm": 2.2459901753698457,
      "language_loss": 0.91119945,
      "learning_rate": 3.47900527676272e-06,
      "loss": 0.93271458,
      "num_input_tokens_seen": 45868470,
      "step": 2146,
      "time_per_iteration": 2.9181129932403564
    },
    {
      "auxiliary_loss_clip": 0.01157065,
      "auxiliary_loss_mlp": 0.0103515,
      "balance_loss_clip": 1.0547905,
      "balance_loss_mlp": 1.02445674,
      "epoch": 0.2581614862021283,
      "flos": 14283146810880.0,
      "grad_norm": 1.765576503581701,
      "language_loss": 0.8825711,
      "learning_rate": 3.478480796979835e-06,
      "loss": 0.90449321,
      "num_input_tokens_seen": 45886355,
      "step": 2147,
      "time_per_iteration": 3.1346940994262695
    },
    {
      "auxiliary_loss_clip": 0.01086373,
      "auxiliary_loss_mlp": 0.01034771,
      "balance_loss_clip": 1.0453651,
      "balance_loss_mlp": 1.02444196,
      "epoch": 0.25828172909276736,
      "flos": 29498281856640.0,
      "grad_norm": 1.5598748485673377,
      "language_loss": 0.7809546,
      "learning_rate": 3.4779560929125894e-06,
      "loss": 0.80216604,
      "num_input_tokens_seen": 45907900,
      "step": 2148,
      "time_per_iteration": 3.0558178424835205
    },
    {
      "auxiliary_loss_clip": 0.01086888,
      "auxiliary_loss_mlp": 0.01001654,
      "balance_loss_clip": 1.02630556,
      "balance_loss_mlp": 1.00007999,
      "epoch": 0.2584019719834065,
      "flos": 67114387376640.0,
      "grad_norm": 0.6644278080325944,
      "language_loss": 0.56901437,
      "learning_rate": 3.4774311646405783e-06,
      "loss": 0.58989978,
      "num_input_tokens_seen": 45977805,
      "step": 2149,
      "time_per_iteration": 3.739992141723633
    },
    {
      "auxiliary_loss_clip": 0.0116327,
      "auxiliary_loss_mlp": 0.00764163,
      "balance_loss_clip": 1.05543685,
      "balance_loss_mlp": 1.00017393,
      "epoch": 0.2585222148740456,
      "flos": 22893555634560.0,
      "grad_norm": 1.9076754250386132,
      "language_loss": 0.83357662,
      "learning_rate": 3.476906012243435e-06,
      "loss": 0.85285091,
      "num_input_tokens_seen": 45996715,
      "step": 2150,
      "time_per_iteration": 2.797743082046509
    },
    {
      "auxiliary_loss_clip": 0.01151381,
      "auxiliary_loss_mlp": 0.01036597,
      "balance_loss_clip": 1.05247521,
      "balance_loss_mlp": 1.02684569,
      "epoch": 0.25864245776468464,
      "flos": 28909202808960.0,
      "grad_norm": 1.6761624650468303,
      "language_loss": 0.81307304,
      "learning_rate": 3.476380635800824e-06,
      "loss": 0.83495277,
      "num_input_tokens_seen": 46017915,
      "step": 2151,
      "time_per_iteration": 2.862705945968628
    },
    {
      "auxiliary_loss_clip": 0.01124492,
      "auxiliary_loss_mlp": 0.01031934,
      "balance_loss_clip": 1.04994833,
      "balance_loss_mlp": 1.02204549,
      "epoch": 0.25876270065532375,
      "flos": 14793185980800.0,
      "grad_norm": 2.315470316012382,
      "language_loss": 0.85731965,
      "learning_rate": 3.475855035392444e-06,
      "loss": 0.87888396,
      "num_input_tokens_seen": 46033235,
      "step": 2152,
      "time_per_iteration": 2.8485937118530273
    },
    {
      "auxiliary_loss_clip": 0.01084241,
      "auxiliary_loss_mlp": 0.00764764,
      "balance_loss_clip": 1.03906989,
      "balance_loss_mlp": 1.00018108,
      "epoch": 0.25888294354596286,
      "flos": 60467821810560.0,
      "grad_norm": 1.7531072518668132,
      "language_loss": 0.71337831,
      "learning_rate": 3.475329211098029e-06,
      "loss": 0.73186839,
      "num_input_tokens_seen": 46056390,
      "step": 2153,
      "time_per_iteration": 3.2490391731262207
    },
    {
      "auxiliary_loss_clip": 0.01134293,
      "auxiliary_loss_mlp": 0.01028201,
      "balance_loss_clip": 1.05126023,
      "balance_loss_mlp": 1.01912355,
      "epoch": 0.2590031864366019,
      "flos": 27851166771840.0,
      "grad_norm": 1.6237395630902083,
      "language_loss": 0.82175016,
      "learning_rate": 3.4748031629973453e-06,
      "loss": 0.84337509,
      "num_input_tokens_seen": 46077120,
      "step": 2154,
      "time_per_iteration": 3.6351840496063232
    },
    {
      "auxiliary_loss_clip": 0.01089882,
      "auxiliary_loss_mlp": 0.01005366,
      "balance_loss_clip": 1.04638863,
      "balance_loss_mlp": 1.00362527,
      "epoch": 0.25912342932724103,
      "flos": 62422444206720.0,
      "grad_norm": 0.9154949356858783,
      "language_loss": 0.56652468,
      "learning_rate": 3.4742768911701944e-06,
      "loss": 0.58747721,
      "num_input_tokens_seen": 46139815,
      "step": 2155,
      "time_per_iteration": 3.497990608215332
    },
    {
      "auxiliary_loss_clip": 0.01185488,
      "auxiliary_loss_mlp": 0.0103307,
      "balance_loss_clip": 1.06080759,
      "balance_loss_mlp": 1.0219717,
      "epoch": 0.25924367221788014,
      "flos": 12378839368320.0,
      "grad_norm": 2.3233607634390765,
      "language_loss": 0.70601618,
      "learning_rate": 3.4737503956964113e-06,
      "loss": 0.72820181,
      "num_input_tokens_seen": 46152120,
      "step": 2156,
      "time_per_iteration": 2.704946756362915
    },
    {
      "auxiliary_loss_clip": 0.01179751,
      "auxiliary_loss_mlp": 0.01039014,
      "balance_loss_clip": 1.05569804,
      "balance_loss_mlp": 1.02762389,
      "epoch": 0.2593639151085192,
      "flos": 14575208296320.0,
      "grad_norm": 2.0515572917105707,
      "language_loss": 0.67394525,
      "learning_rate": 3.473223676655865e-06,
      "loss": 0.6961329,
      "num_input_tokens_seen": 46170120,
      "step": 2157,
      "time_per_iteration": 2.698179006576538
    },
    {
      "auxiliary_loss_clip": 0.01180562,
      "auxiliary_loss_mlp": 0.01043745,
      "balance_loss_clip": 1.06299317,
      "balance_loss_mlp": 1.03254533,
      "epoch": 0.2594841579991583,
      "flos": 15230937029760.0,
      "grad_norm": 4.090497966526451,
      "language_loss": 0.79783022,
      "learning_rate": 3.472696734128459e-06,
      "loss": 0.82007331,
      "num_input_tokens_seen": 46187985,
      "step": 2158,
      "time_per_iteration": 3.626131772994995
    },
    {
      "auxiliary_loss_clip": 0.01158775,
      "auxiliary_loss_mlp": 0.01033872,
      "balance_loss_clip": 1.0564456,
      "balance_loss_mlp": 1.02438331,
      "epoch": 0.2596044008897974,
      "flos": 23623583650560.0,
      "grad_norm": 1.9632749517999948,
      "language_loss": 0.75679016,
      "learning_rate": 3.4721695681941286e-06,
      "loss": 0.77871668,
      "num_input_tokens_seen": 46207025,
      "step": 2159,
      "time_per_iteration": 2.8525428771972656
    },
    {
      "auxiliary_loss_clip": 0.01121734,
      "auxiliary_loss_mlp": 0.01030448,
      "balance_loss_clip": 1.04963255,
      "balance_loss_mlp": 1.02035081,
      "epoch": 0.25972464378043647,
      "flos": 13772281628160.0,
      "grad_norm": 2.058280269976242,
      "language_loss": 0.82317257,
      "learning_rate": 3.471642178932845e-06,
      "loss": 0.84469438,
      "num_input_tokens_seen": 46225670,
      "step": 2160,
      "time_per_iteration": 4.1131062507629395
    },
    {
      "auxiliary_loss_clip": 0.01176639,
      "auxiliary_loss_mlp": 0.01034807,
      "balance_loss_clip": 1.05547547,
      "balance_loss_mlp": 1.02452576,
      "epoch": 0.2598448866710756,
      "flos": 19573578391680.0,
      "grad_norm": 2.0009699065351003,
      "language_loss": 0.8988474,
      "learning_rate": 3.471114566424613e-06,
      "loss": 0.92096186,
      "num_input_tokens_seen": 46244130,
      "step": 2161,
      "time_per_iteration": 3.841566324234009
    },
    {
      "auxiliary_loss_clip": 0.01152408,
      "auxiliary_loss_mlp": 0.00764735,
      "balance_loss_clip": 1.05376911,
      "balance_loss_mlp": 1.00017691,
      "epoch": 0.25996512956171464,
      "flos": 21653237053440.0,
      "grad_norm": 2.1148412459336257,
      "language_loss": 0.75801343,
      "learning_rate": 3.4705867307494715e-06,
      "loss": 0.77718484,
      "num_input_tokens_seen": 46263200,
      "step": 2162,
      "time_per_iteration": 3.718700885772705
    },
    {
      "auxiliary_loss_clip": 0.01179528,
      "auxiliary_loss_mlp": 0.01033252,
      "balance_loss_clip": 1.05685568,
      "balance_loss_mlp": 1.02249384,
      "epoch": 0.26008537245235375,
      "flos": 18223480869120.0,
      "grad_norm": 2.0830217527001977,
      "language_loss": 0.84837526,
      "learning_rate": 3.470058671987492e-06,
      "loss": 0.87050307,
      "num_input_tokens_seen": 46281465,
      "step": 2163,
      "time_per_iteration": 2.673424482345581
    },
    {
      "auxiliary_loss_clip": 0.01190434,
      "auxiliary_loss_mlp": 0.01040876,
      "balance_loss_clip": 1.05991936,
      "balance_loss_mlp": 1.02974808,
      "epoch": 0.26020561534299286,
      "flos": 24645385843200.0,
      "grad_norm": 1.9569754560700496,
      "language_loss": 0.84971666,
      "learning_rate": 3.4695303902187805e-06,
      "loss": 0.87202978,
      "num_input_tokens_seen": 46301020,
      "step": 2164,
      "time_per_iteration": 2.7375688552856445
    },
    {
      "auxiliary_loss_clip": 0.01195242,
      "auxiliary_loss_mlp": 0.01040612,
      "balance_loss_clip": 1.05906355,
      "balance_loss_mlp": 1.02969849,
      "epoch": 0.2603258582336319,
      "flos": 25773662926080.0,
      "grad_norm": 1.9812391133593623,
      "language_loss": 0.78723937,
      "learning_rate": 3.469001885523478e-06,
      "loss": 0.80959791,
      "num_input_tokens_seen": 46321740,
      "step": 2165,
      "time_per_iteration": 2.7445068359375
    },
    {
      "auxiliary_loss_clip": 0.01130226,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.05221295,
      "balance_loss_mlp": 1.02129912,
      "epoch": 0.260446101124271,
      "flos": 28766314506240.0,
      "grad_norm": 1.7554630146063146,
      "language_loss": 0.80859226,
      "learning_rate": 3.4684731579817568e-06,
      "loss": 0.8302002,
      "num_input_tokens_seen": 46342730,
      "step": 2166,
      "time_per_iteration": 2.862365961074829
    },
    {
      "auxiliary_loss_clip": 0.01142153,
      "auxiliary_loss_mlp": 0.0103496,
      "balance_loss_clip": 1.05311918,
      "balance_loss_mlp": 1.02546537,
      "epoch": 0.26056634401491013,
      "flos": 25666757072640.0,
      "grad_norm": 1.578237672923456,
      "language_loss": 0.76590884,
      "learning_rate": 3.4679442076738247e-06,
      "loss": 0.78767997,
      "num_input_tokens_seen": 46362445,
      "step": 2167,
      "time_per_iteration": 2.801440954208374
    },
    {
      "auxiliary_loss_clip": 0.01178529,
      "auxiliary_loss_mlp": 0.00764722,
      "balance_loss_clip": 1.05643749,
      "balance_loss_mlp": 1.00021696,
      "epoch": 0.2606865869055492,
      "flos": 27052765217280.0,
      "grad_norm": 1.9080714866343347,
      "language_loss": 0.83473051,
      "learning_rate": 3.4674150346799245e-06,
      "loss": 0.85416305,
      "num_input_tokens_seen": 46382145,
      "step": 2168,
      "time_per_iteration": 2.726412534713745
    },
    {
      "auxiliary_loss_clip": 0.01114775,
      "auxiliary_loss_mlp": 0.01031606,
      "balance_loss_clip": 1.04663038,
      "balance_loss_mlp": 1.0208354,
      "epoch": 0.2608068297961883,
      "flos": 17712615686400.0,
      "grad_norm": 2.0253330286340754,
      "language_loss": 0.79859304,
      "learning_rate": 3.4668856390803295e-06,
      "loss": 0.82005686,
      "num_input_tokens_seen": 46400025,
      "step": 2169,
      "time_per_iteration": 2.7768726348876953
    },
    {
      "auxiliary_loss_clip": 0.01123541,
      "auxiliary_loss_mlp": 0.01027304,
      "balance_loss_clip": 1.04722381,
      "balance_loss_mlp": 1.01792216,
      "epoch": 0.2609270726868274,
      "flos": 18551632544640.0,
      "grad_norm": 1.98123064412714,
      "language_loss": 0.89605129,
      "learning_rate": 3.4663560209553495e-06,
      "loss": 0.91755968,
      "num_input_tokens_seen": 46418090,
      "step": 2170,
      "time_per_iteration": 2.801992177963257
    },
    {
      "auxiliary_loss_clip": 0.01203857,
      "auxiliary_loss_mlp": 0.01032373,
      "balance_loss_clip": 1.0593605,
      "balance_loss_mlp": 1.0228126,
      "epoch": 0.26104731557746647,
      "flos": 21835699165440.0,
      "grad_norm": 1.7707949504621363,
      "language_loss": 0.79465771,
      "learning_rate": 3.4658261803853267e-06,
      "loss": 0.81702006,
      "num_input_tokens_seen": 46436015,
      "step": 2171,
      "time_per_iteration": 2.621964693069458
    },
    {
      "auxiliary_loss_clip": 0.01135756,
      "auxiliary_loss_mlp": 0.01037332,
      "balance_loss_clip": 1.04685688,
      "balance_loss_mlp": 1.02671671,
      "epoch": 0.2611675584681056,
      "flos": 21689650465920.0,
      "grad_norm": 2.7779417214236637,
      "language_loss": 0.8112154,
      "learning_rate": 3.4652961174506383e-06,
      "loss": 0.8329463,
      "num_input_tokens_seen": 46455885,
      "step": 2172,
      "time_per_iteration": 2.879605770111084
    },
    {
      "auxiliary_loss_clip": 0.01037655,
      "auxiliary_loss_mlp": 0.01003535,
      "balance_loss_clip": 1.01793408,
      "balance_loss_mlp": 1.00192547,
      "epoch": 0.2612878013587447,
      "flos": 71862101389440.0,
      "grad_norm": 0.9672730305688927,
      "language_loss": 0.58114976,
      "learning_rate": 3.464765832231694e-06,
      "loss": 0.60156167,
      "num_input_tokens_seen": 46510050,
      "step": 2173,
      "time_per_iteration": 3.5312585830688477
    },
    {
      "auxiliary_loss_clip": 0.01193661,
      "auxiliary_loss_mlp": 0.01038051,
      "balance_loss_clip": 1.05745518,
      "balance_loss_mlp": 1.02794158,
      "epoch": 0.26140804424938374,
      "flos": 20227511445120.0,
      "grad_norm": 2.5592460528911687,
      "language_loss": 0.70753002,
      "learning_rate": 3.4642353248089373e-06,
      "loss": 0.72984713,
      "num_input_tokens_seen": 46528810,
      "step": 2174,
      "time_per_iteration": 2.990753412246704
    },
    {
      "auxiliary_loss_clip": 0.01178598,
      "auxiliary_loss_mlp": 0.01034284,
      "balance_loss_clip": 1.05899072,
      "balance_loss_mlp": 1.02268529,
      "epoch": 0.26152828714002285,
      "flos": 25557085872000.0,
      "grad_norm": 2.036448518438675,
      "language_loss": 0.80071092,
      "learning_rate": 3.463704595262846e-06,
      "loss": 0.82283974,
      "num_input_tokens_seen": 46549690,
      "step": 2175,
      "time_per_iteration": 2.7343146800994873
    },
    {
      "auxiliary_loss_clip": 0.01130851,
      "auxiliary_loss_mlp": 0.01036209,
      "balance_loss_clip": 1.05194807,
      "balance_loss_mlp": 1.02651143,
      "epoch": 0.26164853003066196,
      "flos": 25446516831360.0,
      "grad_norm": 2.888008541032961,
      "language_loss": 0.70803237,
      "learning_rate": 3.463173643673931e-06,
      "loss": 0.72970295,
      "num_input_tokens_seen": 46572215,
      "step": 2176,
      "time_per_iteration": 2.9077155590057373
    },
    {
      "auxiliary_loss_clip": 0.01041693,
      "auxiliary_loss_mlp": 0.0075808,
      "balance_loss_clip": 1.01447535,
      "balance_loss_mlp": 1.00096357,
      "epoch": 0.261768772921301,
      "flos": 53944580568960.0,
      "grad_norm": 0.8961849807774818,
      "language_loss": 0.63520283,
      "learning_rate": 3.4626424701227387e-06,
      "loss": 0.65320057,
      "num_input_tokens_seen": 46627275,
      "step": 2177,
      "time_per_iteration": 3.272279739379883
    },
    {
      "auxiliary_loss_clip": 0.01091857,
      "auxiliary_loss_mlp": 0.01001534,
      "balance_loss_clip": 1.02162206,
      "balance_loss_mlp": 0.99988919,
      "epoch": 0.26188901581194013,
      "flos": 70687606481280.0,
      "grad_norm": 0.8193326317901974,
      "language_loss": 0.55787951,
      "learning_rate": 3.4621110746898452e-06,
      "loss": 0.57881343,
      "num_input_tokens_seen": 46695135,
      "step": 2178,
      "time_per_iteration": 3.2890186309814453
    },
    {
      "auxiliary_loss_clip": 0.01186825,
      "auxiliary_loss_mlp": 0.01034958,
      "balance_loss_clip": 1.0595696,
      "balance_loss_mlp": 1.02523065,
      "epoch": 0.2620092587025792,
      "flos": 21069580959360.0,
      "grad_norm": 1.9137155868475253,
      "language_loss": 0.74965465,
      "learning_rate": 3.4615794574558654e-06,
      "loss": 0.77187246,
      "num_input_tokens_seen": 46714145,
      "step": 2179,
      "time_per_iteration": 2.663081169128418
    },
    {
      "auxiliary_loss_clip": 0.01132115,
      "auxiliary_loss_mlp": 0.01035311,
      "balance_loss_clip": 1.04785538,
      "balance_loss_mlp": 1.02581561,
      "epoch": 0.2621295015932183,
      "flos": 18369601395840.0,
      "grad_norm": 2.1300498058473427,
      "language_loss": 0.84206426,
      "learning_rate": 3.4610476185014436e-06,
      "loss": 0.86373854,
      "num_input_tokens_seen": 46731405,
      "step": 2180,
      "time_per_iteration": 2.733334541320801
    },
    {
      "auxiliary_loss_clip": 0.01166901,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.05517769,
      "balance_loss_mlp": 1.02346396,
      "epoch": 0.2622497444838574,
      "flos": 23659997063040.0,
      "grad_norm": 1.7074980322548337,
      "language_loss": 0.79408121,
      "learning_rate": 3.4605155579072597e-06,
      "loss": 0.81609857,
      "num_input_tokens_seen": 46751260,
      "step": 2181,
      "time_per_iteration": 2.753816604614258
    },
    {
      "auxiliary_loss_clip": 0.01188889,
      "auxiliary_loss_mlp": 0.0103001,
      "balance_loss_clip": 1.05695069,
      "balance_loss_mlp": 1.02030659,
      "epoch": 0.26236998737449646,
      "flos": 22123810154880.0,
      "grad_norm": 1.7609988498826765,
      "language_loss": 0.71280235,
      "learning_rate": 3.459983275754027e-06,
      "loss": 0.73499143,
      "num_input_tokens_seen": 46770155,
      "step": 2182,
      "time_per_iteration": 2.6256484985351562
    },
    {
      "auxiliary_loss_clip": 0.01157062,
      "auxiliary_loss_mlp": 0.0102941,
      "balance_loss_clip": 1.05012298,
      "balance_loss_mlp": 1.01868093,
      "epoch": 0.26249023026513557,
      "flos": 17895185539200.0,
      "grad_norm": 2.6115009203443136,
      "language_loss": 0.79923201,
      "learning_rate": 3.4594507721224918e-06,
      "loss": 0.82109678,
      "num_input_tokens_seen": 46788805,
      "step": 2183,
      "time_per_iteration": 2.702057361602783
    },
    {
      "auxiliary_loss_clip": 0.01101898,
      "auxiliary_loss_mlp": 0.01034987,
      "balance_loss_clip": 1.04446363,
      "balance_loss_mlp": 1.02494955,
      "epoch": 0.2626104731557747,
      "flos": 18332936588160.0,
      "grad_norm": 1.905977138289329,
      "language_loss": 0.82080519,
      "learning_rate": 3.4589180470934353e-06,
      "loss": 0.84217411,
      "num_input_tokens_seen": 46808670,
      "step": 2184,
      "time_per_iteration": 3.8893980979919434
    },
    {
      "auxiliary_loss_clip": 0.01185717,
      "auxiliary_loss_mlp": 0.00765486,
      "balance_loss_clip": 1.05356193,
      "balance_loss_mlp": 1.00019634,
      "epoch": 0.26273071604641374,
      "flos": 19317714837120.0,
      "grad_norm": 1.7649496799238058,
      "language_loss": 0.7680819,
      "learning_rate": 3.4583851007476713e-06,
      "loss": 0.78759396,
      "num_input_tokens_seen": 46827140,
      "step": 2185,
      "time_per_iteration": 2.7250537872314453
    },
    {
      "auxiliary_loss_clip": 0.01195205,
      "auxiliary_loss_mlp": 0.01035454,
      "balance_loss_clip": 1.0579772,
      "balance_loss_mlp": 1.02483845,
      "epoch": 0.26285095893705285,
      "flos": 18327477720960.0,
      "grad_norm": 4.156478954254319,
      "language_loss": 0.68444288,
      "learning_rate": 3.4578519331660464e-06,
      "loss": 0.70674944,
      "num_input_tokens_seen": 46844135,
      "step": 2186,
      "time_per_iteration": 4.010732889175415
    },
    {
      "auxiliary_loss_clip": 0.01176776,
      "auxiliary_loss_mlp": 0.01034552,
      "balance_loss_clip": 1.05735993,
      "balance_loss_mlp": 1.02487195,
      "epoch": 0.26297120182769196,
      "flos": 20193827466240.0,
      "grad_norm": 1.880189129155291,
      "language_loss": 0.82016617,
      "learning_rate": 3.4573185444294426e-06,
      "loss": 0.84227949,
      "num_input_tokens_seen": 46862500,
      "step": 2187,
      "time_per_iteration": 3.580376625061035
    },
    {
      "auxiliary_loss_clip": 0.0115883,
      "auxiliary_loss_mlp": 0.01032378,
      "balance_loss_clip": 1.05597031,
      "balance_loss_mlp": 1.0224539,
      "epoch": 0.263091444718331,
      "flos": 22418421505920.0,
      "grad_norm": 1.6330192440055589,
      "language_loss": 0.78806031,
      "learning_rate": 3.456784934618774e-06,
      "loss": 0.80997241,
      "num_input_tokens_seen": 46883665,
      "step": 2188,
      "time_per_iteration": 3.6256840229034424
    },
    {
      "auxiliary_loss_clip": 0.01169626,
      "auxiliary_loss_mlp": 0.01030161,
      "balance_loss_clip": 1.0553062,
      "balance_loss_mlp": 1.0198251,
      "epoch": 0.2632116876089701,
      "flos": 19024827338880.0,
      "grad_norm": 1.907192603322275,
      "language_loss": 0.80034733,
      "learning_rate": 3.4562511038149897e-06,
      "loss": 0.8223452,
      "num_input_tokens_seen": 46899160,
      "step": 2189,
      "time_per_iteration": 2.742124557495117
    },
    {
      "auxiliary_loss_clip": 0.01045574,
      "auxiliary_loss_mlp": 0.01001623,
      "balance_loss_clip": 1.01980019,
      "balance_loss_mlp": 1.00007308,
      "epoch": 0.26333193049960923,
      "flos": 67308054531840.0,
      "grad_norm": 0.8622613890357039,
      "language_loss": 0.57792926,
      "learning_rate": 3.4557170520990705e-06,
      "loss": 0.59840119,
      "num_input_tokens_seen": 46959835,
      "step": 2190,
      "time_per_iteration": 3.475649118423462
    },
    {
      "auxiliary_loss_clip": 0.01172281,
      "auxiliary_loss_mlp": 0.01036533,
      "balance_loss_clip": 1.05407882,
      "balance_loss_mlp": 1.02709126,
      "epoch": 0.2634521733902483,
      "flos": 25048806468480.0,
      "grad_norm": 1.5289888753557035,
      "language_loss": 0.86385822,
      "learning_rate": 3.4551827795520324e-06,
      "loss": 0.88594639,
      "num_input_tokens_seen": 46982720,
      "step": 2191,
      "time_per_iteration": 3.0718605518341064
    },
    {
      "auxiliary_loss_clip": 0.01130675,
      "auxiliary_loss_mlp": 0.01038134,
      "balance_loss_clip": 1.04853892,
      "balance_loss_mlp": 1.02820981,
      "epoch": 0.2635724162808874,
      "flos": 20594985534720.0,
      "grad_norm": 1.9408739085301394,
      "language_loss": 0.8512671,
      "learning_rate": 3.4546482862549226e-06,
      "loss": 0.8729552,
      "num_input_tokens_seen": 47003035,
      "step": 2192,
      "time_per_iteration": 2.8501670360565186
    },
    {
      "auxiliary_loss_clip": 0.01161891,
      "auxiliary_loss_mlp": 0.01038871,
      "balance_loss_clip": 1.05687642,
      "balance_loss_mlp": 1.02862525,
      "epoch": 0.2636926591715265,
      "flos": 19244636616960.0,
      "grad_norm": 2.8025384469617407,
      "language_loss": 0.78808272,
      "learning_rate": 3.4541135722888253e-06,
      "loss": 0.81009036,
      "num_input_tokens_seen": 47019625,
      "step": 2193,
      "time_per_iteration": 2.8066294193267822
    },
    {
      "auxiliary_loss_clip": 0.01187763,
      "auxiliary_loss_mlp": 0.01033218,
      "balance_loss_clip": 1.05665898,
      "balance_loss_mlp": 1.02318025,
      "epoch": 0.26381290206216557,
      "flos": 28804882734720.0,
      "grad_norm": 1.6963376786822013,
      "language_loss": 0.80778933,
      "learning_rate": 3.453578637734854e-06,
      "loss": 0.82999909,
      "num_input_tokens_seen": 47040815,
      "step": 2194,
      "time_per_iteration": 2.744016170501709
    },
    {
      "auxiliary_loss_clip": 0.0120531,
      "auxiliary_loss_mlp": 0.01042026,
      "balance_loss_clip": 1.0608232,
      "balance_loss_mlp": 1.03163075,
      "epoch": 0.2639331449528047,
      "flos": 25008909436800.0,
      "grad_norm": 1.587175627429301,
      "language_loss": 0.78316057,
      "learning_rate": 3.4530434826741605e-06,
      "loss": 0.8056339,
      "num_input_tokens_seen": 47061755,
      "step": 2195,
      "time_per_iteration": 2.665670871734619
    },
    {
      "auxiliary_loss_clip": 0.01136015,
      "auxiliary_loss_mlp": 0.00764176,
      "balance_loss_clip": 1.05000925,
      "balance_loss_mlp": 1.00012589,
      "epoch": 0.26405338784344373,
      "flos": 46535775465600.0,
      "grad_norm": 1.6999025078417405,
      "language_loss": 0.69140208,
      "learning_rate": 3.452508107187926e-06,
      "loss": 0.71040404,
      "num_input_tokens_seen": 47085130,
      "step": 2196,
      "time_per_iteration": 3.003899097442627
    },
    {
      "auxiliary_loss_clip": 0.01193622,
      "auxiliary_loss_mlp": 0.01036059,
      "balance_loss_clip": 1.06293559,
      "balance_loss_mlp": 1.02533674,
      "epoch": 0.26417363073408284,
      "flos": 21179467641600.0,
      "grad_norm": 1.9481896286913014,
      "language_loss": 0.77198768,
      "learning_rate": 3.451972511357366e-06,
      "loss": 0.79428446,
      "num_input_tokens_seen": 47104675,
      "step": 2197,
      "time_per_iteration": 2.716717004776001
    },
    {
      "auxiliary_loss_clip": 0.01154767,
      "auxiliary_loss_mlp": 0.01035118,
      "balance_loss_clip": 1.05138576,
      "balance_loss_mlp": 1.025599,
      "epoch": 0.26429387362472195,
      "flos": 22674751937280.0,
      "grad_norm": 1.678918559005747,
      "language_loss": 0.85107255,
      "learning_rate": 3.45143669526373e-06,
      "loss": 0.87297136,
      "num_input_tokens_seen": 47124435,
      "step": 2198,
      "time_per_iteration": 2.744459867477417
    },
    {
      "auxiliary_loss_clip": 0.01073948,
      "auxiliary_loss_mlp": 0.01003143,
      "balance_loss_clip": 1.01460147,
      "balance_loss_mlp": 1.00159359,
      "epoch": 0.264414116515361,
      "flos": 67180534272000.0,
      "grad_norm": 0.7849837716067074,
      "language_loss": 0.63205737,
      "learning_rate": 3.450900658988302e-06,
      "loss": 0.65282834,
      "num_input_tokens_seen": 47185985,
      "step": 2199,
      "time_per_iteration": 3.239539861679077
    },
    {
      "auxiliary_loss_clip": 0.0120784,
      "auxiliary_loss_mlp": 0.01036632,
      "balance_loss_clip": 1.06167948,
      "balance_loss_mlp": 1.02660632,
      "epoch": 0.2645343594060001,
      "flos": 25664709997440.0,
      "grad_norm": 2.056434917179322,
      "language_loss": 0.77996778,
      "learning_rate": 3.450364402612397e-06,
      "loss": 0.80241251,
      "num_input_tokens_seen": 47203140,
      "step": 2200,
      "time_per_iteration": 2.7010011672973633
    },
    {
      "auxiliary_loss_clip": 0.0115382,
      "auxiliary_loss_mlp": 0.01036707,
      "balance_loss_clip": 1.05086899,
      "balance_loss_mlp": 1.02640164,
      "epoch": 0.26465460229663923,
      "flos": 22491822948480.0,
      "grad_norm": 1.8610841126763473,
      "language_loss": 0.84010983,
      "learning_rate": 3.449827926217366e-06,
      "loss": 0.86201519,
      "num_input_tokens_seen": 47222575,
      "step": 2201,
      "time_per_iteration": 2.8877615928649902
    },
    {
      "auxiliary_loss_clip": 0.01176291,
      "auxiliary_loss_mlp": 0.01034747,
      "balance_loss_clip": 1.05241966,
      "balance_loss_mlp": 1.02457798,
      "epoch": 0.2647748451872783,
      "flos": 29388036038400.0,
      "grad_norm": 2.359869543369659,
      "language_loss": 0.80520499,
      "learning_rate": 3.449291229884591e-06,
      "loss": 0.82731539,
      "num_input_tokens_seen": 47243815,
      "step": 2202,
      "time_per_iteration": 2.8963277339935303
    },
    {
      "auxiliary_loss_clip": 0.01098803,
      "auxiliary_loss_mlp": 0.01032541,
      "balance_loss_clip": 1.05080104,
      "balance_loss_mlp": 1.02168691,
      "epoch": 0.2648950880779174,
      "flos": 26797799502720.0,
      "grad_norm": 1.7619674733072366,
      "language_loss": 0.86816716,
      "learning_rate": 3.4487543136954887e-06,
      "loss": 0.88948059,
      "num_input_tokens_seen": 47263435,
      "step": 2203,
      "time_per_iteration": 2.8691790103912354
    },
    {
      "auxiliary_loss_clip": 0.01134595,
      "auxiliary_loss_mlp": 0.01032085,
      "balance_loss_clip": 1.05043316,
      "balance_loss_mlp": 1.02138615,
      "epoch": 0.2650153309685565,
      "flos": 28841008838400.0,
      "grad_norm": 1.654732368686101,
      "language_loss": 0.9098202,
      "learning_rate": 3.448217177731509e-06,
      "loss": 0.93148696,
      "num_input_tokens_seen": 47283920,
      "step": 2204,
      "time_per_iteration": 2.85127329826355
    },
    {
      "auxiliary_loss_clip": 0.01097888,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.04646432,
      "balance_loss_mlp": 1.02328038,
      "epoch": 0.26513557385919556,
      "flos": 20303247271680.0,
      "grad_norm": 1.9810420439766399,
      "language_loss": 0.77938056,
      "learning_rate": 3.4476798220741348e-06,
      "loss": 0.80068845,
      "num_input_tokens_seen": 47302800,
      "step": 2205,
      "time_per_iteration": 2.895852565765381
    },
    {
      "auxiliary_loss_clip": 0.01204503,
      "auxiliary_loss_mlp": 0.01027277,
      "balance_loss_clip": 1.0611459,
      "balance_loss_mlp": 1.01792455,
      "epoch": 0.26525581674983467,
      "flos": 17676274101120.0,
      "grad_norm": 1.5034781490680025,
      "language_loss": 0.78234738,
      "learning_rate": 3.4471422468048826e-06,
      "loss": 0.80466521,
      "num_input_tokens_seen": 47321525,
      "step": 2206,
      "time_per_iteration": 2.7071049213409424
    },
    {
      "auxiliary_loss_clip": 0.01160784,
      "auxiliary_loss_mlp": 0.01036158,
      "balance_loss_clip": 1.05617619,
      "balance_loss_mlp": 1.02477968,
      "epoch": 0.2653760596404738,
      "flos": 26833746038400.0,
      "grad_norm": 2.633378922710899,
      "language_loss": 0.7310183,
      "learning_rate": 3.4466044520053022e-06,
      "loss": 0.75298768,
      "num_input_tokens_seen": 47340530,
      "step": 2207,
      "time_per_iteration": 2.9310860633850098
    },
    {
      "auxiliary_loss_clip": 0.01188277,
      "auxiliary_loss_mlp": 0.00763993,
      "balance_loss_clip": 1.05869734,
      "balance_loss_mlp": 1.00018835,
      "epoch": 0.26549630253111284,
      "flos": 22782160581120.0,
      "grad_norm": 1.99487536511571,
      "language_loss": 0.60256684,
      "learning_rate": 3.446066437756977e-06,
      "loss": 0.62208956,
      "num_input_tokens_seen": 47359735,
      "step": 2208,
      "time_per_iteration": 2.741499423980713
    },
    {
      "auxiliary_loss_clip": 0.01159187,
      "auxiliary_loss_mlp": 0.01035113,
      "balance_loss_clip": 1.05380821,
      "balance_loss_mlp": 1.02494478,
      "epoch": 0.26561654542175195,
      "flos": 23550002640000.0,
      "grad_norm": 2.3908111417340696,
      "language_loss": 0.75011206,
      "learning_rate": 3.4455282041415224e-06,
      "loss": 0.77205503,
      "num_input_tokens_seen": 47378945,
      "step": 2209,
      "time_per_iteration": 2.7694857120513916
    },
    {
      "auxiliary_loss_clip": 0.01130698,
      "auxiliary_loss_mlp": 0.01033067,
      "balance_loss_clip": 1.04437733,
      "balance_loss_mlp": 1.02333999,
      "epoch": 0.265736788312391,
      "flos": 26906680604160.0,
      "grad_norm": 4.086384132498449,
      "language_loss": 0.88169169,
      "learning_rate": 3.4449897512405894e-06,
      "loss": 0.90332937,
      "num_input_tokens_seen": 47398095,
      "step": 2210,
      "time_per_iteration": 4.07525897026062
    },
    {
      "auxiliary_loss_clip": 0.0114945,
      "auxiliary_loss_mlp": 0.01033373,
      "balance_loss_clip": 1.05514562,
      "balance_loss_mlp": 1.02356839,
      "epoch": 0.2658570312030301,
      "flos": 23477139901440.0,
      "grad_norm": 2.087426082158178,
      "language_loss": 0.75518525,
      "learning_rate": 3.444451079135859e-06,
      "loss": 0.77701348,
      "num_input_tokens_seen": 47417605,
      "step": 2211,
      "time_per_iteration": 3.752501964569092
    },
    {
      "auxiliary_loss_clip": 0.01167642,
      "auxiliary_loss_mlp": 0.01036173,
      "balance_loss_clip": 1.0506345,
      "balance_loss_mlp": 1.02570605,
      "epoch": 0.2659772740936692,
      "flos": 21866402315520.0,
      "grad_norm": 2.117381906915871,
      "language_loss": 0.73889995,
      "learning_rate": 3.4439121879090493e-06,
      "loss": 0.76093811,
      "num_input_tokens_seen": 47435385,
      "step": 2212,
      "time_per_iteration": 2.7870938777923584
    },
    {
      "auxiliary_loss_clip": 0.01177173,
      "auxiliary_loss_mlp": 0.01035478,
      "balance_loss_clip": 1.05731416,
      "balance_loss_mlp": 1.02584028,
      "epoch": 0.2660975169843083,
      "flos": 19793100360960.0,
      "grad_norm": 1.8717778419550684,
      "language_loss": 0.83164346,
      "learning_rate": 3.4433730776419082e-06,
      "loss": 0.85376996,
      "num_input_tokens_seen": 47454310,
      "step": 2213,
      "time_per_iteration": 3.580636739730835
    },
    {
      "auxiliary_loss_clip": 0.01127095,
      "auxiliary_loss_mlp": 0.01036375,
      "balance_loss_clip": 1.04353678,
      "balance_loss_mlp": 1.0260098,
      "epoch": 0.2662177598749474,
      "flos": 29018981750400.0,
      "grad_norm": 2.3355398844046085,
      "language_loss": 0.80392635,
      "learning_rate": 3.4428337484162183e-06,
      "loss": 0.82556099,
      "num_input_tokens_seen": 47475120,
      "step": 2214,
      "time_per_iteration": 3.755660057067871
    },
    {
      "auxiliary_loss_clip": 0.01162843,
      "auxiliary_loss_mlp": 0.01030342,
      "balance_loss_clip": 1.05417395,
      "balance_loss_mlp": 1.0197866,
      "epoch": 0.2663380027655865,
      "flos": 21762549118080.0,
      "grad_norm": 1.973136230936309,
      "language_loss": 0.84853423,
      "learning_rate": 3.442294200313797e-06,
      "loss": 0.87046617,
      "num_input_tokens_seen": 47493150,
      "step": 2215,
      "time_per_iteration": 2.884361743927002
    },
    {
      "auxiliary_loss_clip": 0.01058994,
      "auxiliary_loss_mlp": 0.010024,
      "balance_loss_clip": 1.01414967,
      "balance_loss_mlp": 1.00076735,
      "epoch": 0.26645824565622556,
      "flos": 66980333819520.0,
      "grad_norm": 0.7709664771895247,
      "language_loss": 0.52716023,
      "learning_rate": 3.4417544334164916e-06,
      "loss": 0.5477742,
      "num_input_tokens_seen": 47557295,
      "step": 2216,
      "time_per_iteration": 3.3424155712127686
    },
    {
      "auxiliary_loss_clip": 0.01121069,
      "auxiliary_loss_mlp": 0.01035721,
      "balance_loss_clip": 1.04796517,
      "balance_loss_mlp": 1.02605367,
      "epoch": 0.26657848854686467,
      "flos": 25264198373760.0,
      "grad_norm": 3.6606803441946716,
      "language_loss": 0.77427572,
      "learning_rate": 3.4412144478061854e-06,
      "loss": 0.7958436,
      "num_input_tokens_seen": 47579705,
      "step": 2217,
      "time_per_iteration": 2.8884966373443604
    },
    {
      "auxiliary_loss_clip": 0.0118062,
      "auxiliary_loss_mlp": 0.01038965,
      "balance_loss_clip": 1.05720747,
      "balance_loss_mlp": 1.02795053,
      "epoch": 0.2666987314375038,
      "flos": 23696769611520.0,
      "grad_norm": 2.432072090256995,
      "language_loss": 0.75499189,
      "learning_rate": 3.4406742435647925e-06,
      "loss": 0.77718771,
      "num_input_tokens_seen": 47599770,
      "step": 2218,
      "time_per_iteration": 2.7333521842956543
    },
    {
      "auxiliary_loss_clip": 0.01191453,
      "auxiliary_loss_mlp": 0.01033271,
      "balance_loss_clip": 1.06049669,
      "balance_loss_mlp": 1.02332902,
      "epoch": 0.26681897432814283,
      "flos": 27048958375680.0,
      "grad_norm": 1.9593256685361515,
      "language_loss": 0.78989089,
      "learning_rate": 3.440133820774263e-06,
      "loss": 0.81213814,
      "num_input_tokens_seen": 47619580,
      "step": 2219,
      "time_per_iteration": 2.71809720993042
    },
    {
      "auxiliary_loss_clip": 0.01138763,
      "auxiliary_loss_mlp": 0.01039296,
      "balance_loss_clip": 1.04566813,
      "balance_loss_mlp": 1.02826321,
      "epoch": 0.26693921721878194,
      "flos": 28985944216320.0,
      "grad_norm": 2.089295059747091,
      "language_loss": 0.82126969,
      "learning_rate": 3.439593179516578e-06,
      "loss": 0.84305024,
      "num_input_tokens_seen": 47639490,
      "step": 2220,
      "time_per_iteration": 2.7919375896453857
    },
    {
      "auxiliary_loss_clip": 0.01155245,
      "auxiliary_loss_mlp": 0.01039112,
      "balance_loss_clip": 1.0531888,
      "balance_loss_mlp": 1.02737582,
      "epoch": 0.26705946010942105,
      "flos": 21507834798720.0,
      "grad_norm": 1.7986534609015525,
      "language_loss": 0.81322199,
      "learning_rate": 3.4390523198737524e-06,
      "loss": 0.83516556,
      "num_input_tokens_seen": 47658650,
      "step": 2221,
      "time_per_iteration": 2.7334370613098145
    },
    {
      "auxiliary_loss_clip": 0.01206344,
      "auxiliary_loss_mlp": 0.01035815,
      "balance_loss_clip": 1.06085968,
      "balance_loss_mlp": 1.02457333,
      "epoch": 0.2671797030000601,
      "flos": 21471277731840.0,
      "grad_norm": 3.713593575590001,
      "language_loss": 0.73794192,
      "learning_rate": 3.4385112419278333e-06,
      "loss": 0.76036346,
      "num_input_tokens_seen": 47679875,
      "step": 2222,
      "time_per_iteration": 2.7067692279815674
    },
    {
      "auxiliary_loss_clip": 0.01035762,
      "auxiliary_loss_mlp": 0.01003627,
      "balance_loss_clip": 1.01489305,
      "balance_loss_mlp": 1.0020293,
      "epoch": 0.2672999458906992,
      "flos": 64189929767040.0,
      "grad_norm": 1.1402715761803843,
      "language_loss": 0.6483233,
      "learning_rate": 3.4379699457609033e-06,
      "loss": 0.66871715,
      "num_input_tokens_seen": 47737700,
      "step": 2223,
      "time_per_iteration": 3.249392032623291
    },
    {
      "auxiliary_loss_clip": 0.01173069,
      "auxiliary_loss_mlp": 0.01036321,
      "balance_loss_clip": 1.05300367,
      "balance_loss_mlp": 1.02603316,
      "epoch": 0.26742018878133833,
      "flos": 16909042573440.0,
      "grad_norm": 1.8043774245299127,
      "language_loss": 0.9039917,
      "learning_rate": 3.4374284314550755e-06,
      "loss": 0.92608559,
      "num_input_tokens_seen": 47756740,
      "step": 2224,
      "time_per_iteration": 2.7601118087768555
    },
    {
      "auxiliary_loss_clip": 0.01105234,
      "auxiliary_loss_mlp": 0.01031158,
      "balance_loss_clip": 1.04775214,
      "balance_loss_mlp": 1.02058423,
      "epoch": 0.2675404316719774,
      "flos": 20667560964480.0,
      "grad_norm": 1.8047952929348465,
      "language_loss": 0.80328327,
      "learning_rate": 3.436886699092498e-06,
      "loss": 0.82464719,
      "num_input_tokens_seen": 47775255,
      "step": 2225,
      "time_per_iteration": 2.9320008754730225
    },
    {
      "auxiliary_loss_clip": 0.01195082,
      "auxiliary_loss_mlp": 0.0103622,
      "balance_loss_clip": 1.06011367,
      "balance_loss_mlp": 1.02405453,
      "epoch": 0.2676606745626165,
      "flos": 17485013157120.0,
      "grad_norm": 8.16393844870237,
      "language_loss": 0.7174772,
      "learning_rate": 3.4363447487553502e-06,
      "loss": 0.7397902,
      "num_input_tokens_seen": 47788570,
      "step": 2226,
      "time_per_iteration": 2.6752312183380127
    },
    {
      "auxiliary_loss_clip": 0.01171998,
      "auxiliary_loss_mlp": 0.01048997,
      "balance_loss_clip": 1.05406797,
      "balance_loss_mlp": 1.0386672,
      "epoch": 0.26778091745325555,
      "flos": 27852675143040.0,
      "grad_norm": 1.8543936611334195,
      "language_loss": 0.7762273,
      "learning_rate": 3.4358025805258455e-06,
      "loss": 0.7984373,
      "num_input_tokens_seen": 47808275,
      "step": 2227,
      "time_per_iteration": 2.8104536533355713
    },
    {
      "auxiliary_loss_clip": 0.01154252,
      "auxiliary_loss_mlp": 0.01032632,
      "balance_loss_clip": 1.04950392,
      "balance_loss_mlp": 1.02218318,
      "epoch": 0.26790116034389466,
      "flos": 20955995176320.0,
      "grad_norm": 2.7503741298778532,
      "language_loss": 0.83626842,
      "learning_rate": 3.435260194486232e-06,
      "loss": 0.85813719,
      "num_input_tokens_seen": 47826245,
      "step": 2228,
      "time_per_iteration": 2.8117096424102783
    },
    {
      "auxiliary_loss_clip": 0.01128307,
      "auxiliary_loss_mlp": 0.01036361,
      "balance_loss_clip": 1.04308605,
      "balance_loss_mlp": 1.0248518,
      "epoch": 0.2680214032345338,
      "flos": 18040659621120.0,
      "grad_norm": 2.1894249159713532,
      "language_loss": 0.82367975,
      "learning_rate": 3.4347175907187875e-06,
      "loss": 0.84532648,
      "num_input_tokens_seen": 47843235,
      "step": 2229,
      "time_per_iteration": 2.7997610569000244
    },
    {
      "auxiliary_loss_clip": 0.01188848,
      "auxiliary_loss_mlp": 0.01036563,
      "balance_loss_clip": 1.05833781,
      "balance_loss_mlp": 1.02635241,
      "epoch": 0.26814164612517283,
      "flos": 22419427086720.0,
      "grad_norm": 1.7802222926045004,
      "language_loss": 0.88241774,
      "learning_rate": 3.4341747693058254e-06,
      "loss": 0.90467185,
      "num_input_tokens_seen": 47861710,
      "step": 2230,
      "time_per_iteration": 2.731583595275879
    },
    {
      "auxiliary_loss_clip": 0.01144006,
      "auxiliary_loss_mlp": 0.01040167,
      "balance_loss_clip": 1.04631567,
      "balance_loss_mlp": 1.02947402,
      "epoch": 0.26826188901581194,
      "flos": 35627371159680.0,
      "grad_norm": 1.7506699934529546,
      "language_loss": 0.77391183,
      "learning_rate": 3.4336317303296916e-06,
      "loss": 0.7957536,
      "num_input_tokens_seen": 47882685,
      "step": 2231,
      "time_per_iteration": 2.862308979034424
    },
    {
      "auxiliary_loss_clip": 0.01143849,
      "auxiliary_loss_mlp": 0.01029447,
      "balance_loss_clip": 1.05442429,
      "balance_loss_mlp": 1.01963603,
      "epoch": 0.26838213190645105,
      "flos": 17639788861440.0,
      "grad_norm": 2.3765590941717236,
      "language_loss": 0.7526648,
      "learning_rate": 3.4330884738727635e-06,
      "loss": 0.77439779,
      "num_input_tokens_seen": 47900860,
      "step": 2232,
      "time_per_iteration": 2.7346982955932617
    },
    {
      "auxiliary_loss_clip": 0.01174801,
      "auxiliary_loss_mlp": 0.01026016,
      "balance_loss_clip": 1.05468667,
      "balance_loss_mlp": 1.01542449,
      "epoch": 0.2685023747970901,
      "flos": 22674823764480.0,
      "grad_norm": 2.8605424234780394,
      "language_loss": 0.71192634,
      "learning_rate": 3.4325450000174535e-06,
      "loss": 0.73393452,
      "num_input_tokens_seen": 47917500,
      "step": 2233,
      "time_per_iteration": 2.697563648223877
    },
    {
      "auxiliary_loss_clip": 0.01161805,
      "auxiliary_loss_mlp": 0.01037525,
      "balance_loss_clip": 1.05234122,
      "balance_loss_mlp": 1.02739859,
      "epoch": 0.2686226176877292,
      "flos": 20120533764480.0,
      "grad_norm": 6.7408743570567315,
      "language_loss": 0.74708748,
      "learning_rate": 3.4320013088462067e-06,
      "loss": 0.76908076,
      "num_input_tokens_seen": 47934860,
      "step": 2234,
      "time_per_iteration": 2.6656312942504883
    },
    {
      "auxiliary_loss_clip": 0.01168233,
      "auxiliary_loss_mlp": 0.01032466,
      "balance_loss_clip": 1.05852914,
      "balance_loss_mlp": 1.02274489,
      "epoch": 0.2687428605783683,
      "flos": 21872040750720.0,
      "grad_norm": 1.4779080159875955,
      "language_loss": 0.81714201,
      "learning_rate": 3.431457400441499e-06,
      "loss": 0.839149,
      "num_input_tokens_seen": 47955255,
      "step": 2235,
      "time_per_iteration": 2.748307704925537
    },
    {
      "auxiliary_loss_clip": 0.01062621,
      "auxiliary_loss_mlp": 0.01002059,
      "balance_loss_clip": 1.01429498,
      "balance_loss_mlp": 1.00040245,
      "epoch": 0.2688631034690074,
      "flos": 69943320766080.0,
      "grad_norm": 0.9104033504112127,
      "language_loss": 0.60898262,
      "learning_rate": 3.4309132748858424e-06,
      "loss": 0.62962943,
      "num_input_tokens_seen": 48016245,
      "step": 2236,
      "time_per_iteration": 4.148430585861206
    },
    {
      "auxiliary_loss_clip": 0.01173284,
      "auxiliary_loss_mlp": 0.01031179,
      "balance_loss_clip": 1.05500388,
      "balance_loss_mlp": 1.02188039,
      "epoch": 0.2689833463596465,
      "flos": 22856639431680.0,
      "grad_norm": 1.7320036131921834,
      "language_loss": 0.83538562,
      "learning_rate": 3.430368932261779e-06,
      "loss": 0.85743022,
      "num_input_tokens_seen": 48036600,
      "step": 2237,
      "time_per_iteration": 3.6978087425231934
    },
    {
      "auxiliary_loss_clip": 0.01124686,
      "auxiliary_loss_mlp": 0.01033176,
      "balance_loss_clip": 1.0456568,
      "balance_loss_mlp": 1.02347279,
      "epoch": 0.2691035892502856,
      "flos": 17200242132480.0,
      "grad_norm": 1.9663589229913925,
      "language_loss": 0.75106871,
      "learning_rate": 3.429824372651886e-06,
      "loss": 0.77264732,
      "num_input_tokens_seen": 48054750,
      "step": 2238,
      "time_per_iteration": 2.7704267501831055
    },
    {
      "auxiliary_loss_clip": 0.01157107,
      "auxiliary_loss_mlp": 0.01029981,
      "balance_loss_clip": 1.05608678,
      "balance_loss_mlp": 1.01949072,
      "epoch": 0.26922383214092466,
      "flos": 17747484814080.0,
      "grad_norm": 2.9967636756375104,
      "language_loss": 0.8347801,
      "learning_rate": 3.4292795961387732e-06,
      "loss": 0.85665101,
      "num_input_tokens_seen": 48072650,
      "step": 2239,
      "time_per_iteration": 3.6578757762908936
    },
    {
      "auxiliary_loss_clip": 0.01185954,
      "auxiliary_loss_mlp": 0.01036406,
      "balance_loss_clip": 1.05721903,
      "balance_loss_mlp": 1.02619028,
      "epoch": 0.26934407503156377,
      "flos": 16173376122240.0,
      "grad_norm": 3.282407889403582,
      "language_loss": 0.87681949,
      "learning_rate": 3.4287346028050818e-06,
      "loss": 0.89904308,
      "num_input_tokens_seen": 48088720,
      "step": 2240,
      "time_per_iteration": 3.5981497764587402
    },
    {
      "auxiliary_loss_clip": 0.01141591,
      "auxiliary_loss_mlp": 0.01034491,
      "balance_loss_clip": 1.05250752,
      "balance_loss_mlp": 1.0249784,
      "epoch": 0.2694643179222028,
      "flos": 23732895715200.0,
      "grad_norm": 1.5857176024929818,
      "language_loss": 0.79632175,
      "learning_rate": 3.4281893927334866e-06,
      "loss": 0.81808257,
      "num_input_tokens_seen": 48108630,
      "step": 2241,
      "time_per_iteration": 2.7531402111053467
    },
    {
      "auxiliary_loss_clip": 0.01190479,
      "auxiliary_loss_mlp": 0.0103459,
      "balance_loss_clip": 1.05642903,
      "balance_loss_mlp": 1.02522588,
      "epoch": 0.26958456081284193,
      "flos": 24718140840960.0,
      "grad_norm": 2.1843500625593224,
      "language_loss": 0.75348449,
      "learning_rate": 3.4276439660066963e-06,
      "loss": 0.7757352,
      "num_input_tokens_seen": 48128330,
      "step": 2242,
      "time_per_iteration": 2.702131986618042
    },
    {
      "auxiliary_loss_clip": 0.01163815,
      "auxiliary_loss_mlp": 0.00763897,
      "balance_loss_clip": 1.05442142,
      "balance_loss_mlp": 1.00021172,
      "epoch": 0.26970480370348104,
      "flos": 18112588606080.0,
      "grad_norm": 2.056673201528145,
      "language_loss": 0.84118617,
      "learning_rate": 3.427098322707452e-06,
      "loss": 0.86046332,
      "num_input_tokens_seen": 48144295,
      "step": 2243,
      "time_per_iteration": 2.642643690109253
    },
    {
      "auxiliary_loss_clip": 0.01140821,
      "auxiliary_loss_mlp": 0.01038615,
      "balance_loss_clip": 1.051368,
      "balance_loss_mlp": 1.02899492,
      "epoch": 0.2698250465941201,
      "flos": 10816546250880.0,
      "grad_norm": 2.4735244342997156,
      "language_loss": 0.8926304,
      "learning_rate": 3.426552462918526e-06,
      "loss": 0.91442478,
      "num_input_tokens_seen": 48162230,
      "step": 2244,
      "time_per_iteration": 2.6872997283935547
    },
    {
      "auxiliary_loss_clip": 0.01156857,
      "auxiliary_loss_mlp": 0.01032314,
      "balance_loss_clip": 1.05388093,
      "balance_loss_mlp": 1.02311683,
      "epoch": 0.2699452894847592,
      "flos": 17308117653120.0,
      "grad_norm": 2.310129621284818,
      "language_loss": 0.72953272,
      "learning_rate": 3.426006386722726e-06,
      "loss": 0.75142443,
      "num_input_tokens_seen": 48180290,
      "step": 2245,
      "time_per_iteration": 2.7400436401367188
    },
    {
      "auxiliary_loss_clip": 0.01136689,
      "auxiliary_loss_mlp": 0.01035256,
      "balance_loss_clip": 1.05454242,
      "balance_loss_mlp": 1.02534437,
      "epoch": 0.2700655323753983,
      "flos": 18078150441600.0,
      "grad_norm": 2.4112606236942544,
      "language_loss": 0.92052239,
      "learning_rate": 3.4254600942028914e-06,
      "loss": 0.94224179,
      "num_input_tokens_seen": 48198165,
      "step": 2246,
      "time_per_iteration": 2.772778034210205
    },
    {
      "auxiliary_loss_clip": 0.01166561,
      "auxiliary_loss_mlp": 0.0103803,
      "balance_loss_clip": 1.05580401,
      "balance_loss_mlp": 1.0284934,
      "epoch": 0.2701857752660374,
      "flos": 18186636493440.0,
      "grad_norm": 2.3148747871654916,
      "language_loss": 0.82909358,
      "learning_rate": 3.424913585441893e-06,
      "loss": 0.85113949,
      "num_input_tokens_seen": 48216000,
      "step": 2247,
      "time_per_iteration": 2.6435744762420654
    },
    {
      "auxiliary_loss_clip": 0.0120281,
      "auxiliary_loss_mlp": 0.01034877,
      "balance_loss_clip": 1.05840158,
      "balance_loss_mlp": 1.02492905,
      "epoch": 0.2703060181566765,
      "flos": 16319496648960.0,
      "grad_norm": 2.203281031538236,
      "language_loss": 0.87731922,
      "learning_rate": 3.4243668605226374e-06,
      "loss": 0.89969599,
      "num_input_tokens_seen": 48233025,
      "step": 2248,
      "time_per_iteration": 2.600879669189453
    },
    {
      "auxiliary_loss_clip": 0.01144874,
      "auxiliary_loss_mlp": 0.01039306,
      "balance_loss_clip": 1.0492115,
      "balance_loss_mlp": 1.02815974,
      "epoch": 0.2704262610473156,
      "flos": 19572357329280.0,
      "grad_norm": 2.982822127828276,
      "language_loss": 0.82783031,
      "learning_rate": 3.423819919528061e-06,
      "loss": 0.84967214,
      "num_input_tokens_seen": 48251110,
      "step": 2249,
      "time_per_iteration": 2.7453863620758057
    },
    {
      "auxiliary_loss_clip": 0.01166384,
      "auxiliary_loss_mlp": 0.01027645,
      "balance_loss_clip": 1.05419934,
      "balance_loss_mlp": 1.01726758,
      "epoch": 0.27054650393795465,
      "flos": 20740746925440.0,
      "grad_norm": 1.7998415895268454,
      "language_loss": 0.784042,
      "learning_rate": 3.4232727625411355e-06,
      "loss": 0.80598229,
      "num_input_tokens_seen": 48270215,
      "step": 2250,
      "time_per_iteration": 2.693774461746216
    },
    {
      "auxiliary_loss_clip": 0.0117,
      "auxiliary_loss_mlp": 0.01030455,
      "balance_loss_clip": 1.05299962,
      "balance_loss_mlp": 1.02125835,
      "epoch": 0.27066674682859376,
      "flos": 18658322916480.0,
      "grad_norm": 1.646094577151664,
      "language_loss": 0.86675346,
      "learning_rate": 3.4227253896448626e-06,
      "loss": 0.88875806,
      "num_input_tokens_seen": 48288075,
      "step": 2251,
      "time_per_iteration": 2.690833568572998
    },
    {
      "auxiliary_loss_clip": 0.01120655,
      "auxiliary_loss_mlp": 0.01034565,
      "balance_loss_clip": 1.04350734,
      "balance_loss_mlp": 1.02441442,
      "epoch": 0.2707869897192329,
      "flos": 23002759958400.0,
      "grad_norm": 4.641375683183339,
      "language_loss": 0.82582498,
      "learning_rate": 3.42217780092228e-06,
      "loss": 0.84737718,
      "num_input_tokens_seen": 48306415,
      "step": 2252,
      "time_per_iteration": 2.8395047187805176
    },
    {
      "auxiliary_loss_clip": 0.01106261,
      "auxiliary_loss_mlp": 0.01001059,
      "balance_loss_clip": 1.04613638,
      "balance_loss_mlp": 0.99924749,
      "epoch": 0.27090723260987193,
      "flos": 58323240293760.0,
      "grad_norm": 0.7952852772257865,
      "language_loss": 0.60324365,
      "learning_rate": 3.421629996456456e-06,
      "loss": 0.62431687,
      "num_input_tokens_seen": 48365035,
      "step": 2253,
      "time_per_iteration": 3.1426992416381836
    },
    {
      "auxiliary_loss_clip": 0.01135895,
      "auxiliary_loss_mlp": 0.0076387,
      "balance_loss_clip": 1.052279,
      "balance_loss_mlp": 1.00033402,
      "epoch": 0.27102747550051104,
      "flos": 11984540797440.0,
      "grad_norm": 2.014328382716697,
      "language_loss": 0.82803392,
      "learning_rate": 3.421081976330491e-06,
      "loss": 0.84703165,
      "num_input_tokens_seen": 48383550,
      "step": 2254,
      "time_per_iteration": 2.74092960357666
    },
    {
      "auxiliary_loss_clip": 0.01123725,
      "auxiliary_loss_mlp": 0.01032891,
      "balance_loss_clip": 1.0473578,
      "balance_loss_mlp": 1.02262688,
      "epoch": 0.27114771839115015,
      "flos": 19900401264000.0,
      "grad_norm": 2.1273756116256006,
      "language_loss": 0.8766849,
      "learning_rate": 3.4205337406275207e-06,
      "loss": 0.89825106,
      "num_input_tokens_seen": 48403670,
      "step": 2255,
      "time_per_iteration": 2.8263278007507324
    },
    {
      "auxiliary_loss_clip": 0.01098597,
      "auxiliary_loss_mlp": 0.01033869,
      "balance_loss_clip": 1.04830122,
      "balance_loss_mlp": 1.02405226,
      "epoch": 0.2712679612817892,
      "flos": 18331966920960.0,
      "grad_norm": 2.5871804079361405,
      "language_loss": 0.75099635,
      "learning_rate": 3.4199852894307114e-06,
      "loss": 0.77232105,
      "num_input_tokens_seen": 48420420,
      "step": 2256,
      "time_per_iteration": 2.8915212154388428
    },
    {
      "auxiliary_loss_clip": 0.01135189,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.0513978,
      "balance_loss_mlp": 1.02095032,
      "epoch": 0.2713882041724283,
      "flos": 24460302038400.0,
      "grad_norm": 1.8875798188300519,
      "language_loss": 0.78900909,
      "learning_rate": 3.419436622823262e-06,
      "loss": 0.81067681,
      "num_input_tokens_seen": 48441140,
      "step": 2257,
      "time_per_iteration": 3.0892281532287598
    },
    {
      "auxiliary_loss_clip": 0.01153744,
      "auxiliary_loss_mlp": 0.01029785,
      "balance_loss_clip": 1.05337334,
      "balance_loss_mlp": 1.02031398,
      "epoch": 0.27150844706306737,
      "flos": 23039317025280.0,
      "grad_norm": 1.5747112218783306,
      "language_loss": 0.74024773,
      "learning_rate": 3.4188877408884063e-06,
      "loss": 0.76208305,
      "num_input_tokens_seen": 48461845,
      "step": 2258,
      "time_per_iteration": 2.733814239501953
    },
    {
      "auxiliary_loss_clip": 0.01177169,
      "auxiliary_loss_mlp": 0.01035839,
      "balance_loss_clip": 1.05653548,
      "balance_loss_mlp": 1.0258255,
      "epoch": 0.2716286899537065,
      "flos": 22563644192640.0,
      "grad_norm": 2.3257404466071128,
      "language_loss": 0.64690155,
      "learning_rate": 3.4183386437094088e-06,
      "loss": 0.66903162,
      "num_input_tokens_seen": 48478510,
      "step": 2259,
      "time_per_iteration": 2.6496236324310303
    },
    {
      "auxiliary_loss_clip": 0.0113768,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.05261397,
      "balance_loss_mlp": 1.01954782,
      "epoch": 0.2717489328443456,
      "flos": 13115044523520.0,
      "grad_norm": 2.041980110851635,
      "language_loss": 0.81841588,
      "learning_rate": 3.417789331369565e-06,
      "loss": 0.84008527,
      "num_input_tokens_seen": 48494300,
      "step": 2260,
      "time_per_iteration": 2.775667667388916
    },
    {
      "auxiliary_loss_clip": 0.01177074,
      "auxiliary_loss_mlp": 0.01034248,
      "balance_loss_clip": 1.05610991,
      "balance_loss_mlp": 1.02423477,
      "epoch": 0.27186917573498465,
      "flos": 29278688060160.0,
      "grad_norm": 2.5513367446166626,
      "language_loss": 0.91445917,
      "learning_rate": 3.4172398039522088e-06,
      "loss": 0.93657243,
      "num_input_tokens_seen": 48515585,
      "step": 2261,
      "time_per_iteration": 2.757850408554077
    },
    {
      "auxiliary_loss_clip": 0.01161594,
      "auxiliary_loss_mlp": 0.01033696,
      "balance_loss_clip": 1.0532819,
      "balance_loss_mlp": 1.02368248,
      "epoch": 0.27198941862562376,
      "flos": 26032220000640.0,
      "grad_norm": 1.844460617228299,
      "language_loss": 0.79915619,
      "learning_rate": 3.4166900615407e-06,
      "loss": 0.82110918,
      "num_input_tokens_seen": 48533500,
      "step": 2262,
      "time_per_iteration": 3.916639566421509
    },
    {
      "auxiliary_loss_clip": 0.01194295,
      "auxiliary_loss_mlp": 0.00763789,
      "balance_loss_clip": 1.05870581,
      "balance_loss_mlp": 1.00024748,
      "epoch": 0.27210966151626287,
      "flos": 32780983760640.0,
      "grad_norm": 2.2150011892898496,
      "language_loss": 0.75135088,
      "learning_rate": 3.416140104218436e-06,
      "loss": 0.77093172,
      "num_input_tokens_seen": 48552865,
      "step": 2263,
      "time_per_iteration": 3.7570669651031494
    },
    {
      "auxiliary_loss_clip": 0.01062407,
      "auxiliary_loss_mlp": 0.01002654,
      "balance_loss_clip": 1.01475704,
      "balance_loss_mlp": 1.00112808,
      "epoch": 0.2722299044069019,
      "flos": 65471043219840.0,
      "grad_norm": 0.839536409271716,
      "language_loss": 0.69681686,
      "learning_rate": 3.4155899320688437e-06,
      "loss": 0.71746743,
      "num_input_tokens_seen": 48618940,
      "step": 2264,
      "time_per_iteration": 3.276496648788452
    },
    {
      "auxiliary_loss_clip": 0.01121386,
      "auxiliary_loss_mlp": 0.01033515,
      "balance_loss_clip": 1.04651988,
      "balance_loss_mlp": 1.02384686,
      "epoch": 0.27235014729754103,
      "flos": 15334143782400.0,
      "grad_norm": 2.0993799933218744,
      "language_loss": 0.74299955,
      "learning_rate": 3.415039545175384e-06,
      "loss": 0.76454866,
      "num_input_tokens_seen": 48634665,
      "step": 2265,
      "time_per_iteration": 3.720205068588257
    },
    {
      "auxiliary_loss_clip": 0.01192151,
      "auxiliary_loss_mlp": 0.01032754,
      "balance_loss_clip": 1.05841064,
      "balance_loss_mlp": 1.02254415,
      "epoch": 0.27247039018818014,
      "flos": 21872363973120.0,
      "grad_norm": 5.338783261482775,
      "language_loss": 0.65072721,
      "learning_rate": 3.414488943621551e-06,
      "loss": 0.67297626,
      "num_input_tokens_seen": 48653330,
      "step": 2266,
      "time_per_iteration": 3.893923282623291
    },
    {
      "auxiliary_loss_clip": 0.0119105,
      "auxiliary_loss_mlp": 0.01028866,
      "balance_loss_clip": 1.05563021,
      "balance_loss_mlp": 1.01885867,
      "epoch": 0.2725906330788192,
      "flos": 18695490514560.0,
      "grad_norm": 1.8692336980693418,
      "language_loss": 0.73558092,
      "learning_rate": 3.41393812749087e-06,
      "loss": 0.75778008,
      "num_input_tokens_seen": 48671375,
      "step": 2267,
      "time_per_iteration": 2.602783203125
    },
    {
      "auxiliary_loss_clip": 0.01169965,
      "auxiliary_loss_mlp": 0.00764537,
      "balance_loss_clip": 1.05537963,
      "balance_loss_mlp": 1.00029397,
      "epoch": 0.2727108759694583,
      "flos": 17886099398400.0,
      "grad_norm": 2.265543287192323,
      "language_loss": 0.72028804,
      "learning_rate": 3.4133870968668984e-06,
      "loss": 0.73963308,
      "num_input_tokens_seen": 48686175,
      "step": 2268,
      "time_per_iteration": 2.6458866596221924
    },
    {
      "auxiliary_loss_clip": 0.0110184,
      "auxiliary_loss_mlp": 0.01033254,
      "balance_loss_clip": 1.04610896,
      "balance_loss_mlp": 1.02172649,
      "epoch": 0.2728311188600974,
      "flos": 24461666755200.0,
      "grad_norm": 1.8753536891900282,
      "language_loss": 0.7861563,
      "learning_rate": 3.412835851833229e-06,
      "loss": 0.80750722,
      "num_input_tokens_seen": 48708370,
      "step": 2269,
      "time_per_iteration": 3.059894323348999
    },
    {
      "auxiliary_loss_clip": 0.01154352,
      "auxiliary_loss_mlp": 0.01039742,
      "balance_loss_clip": 1.05267322,
      "balance_loss_mlp": 1.02953756,
      "epoch": 0.2729513617507365,
      "flos": 30993314757120.0,
      "grad_norm": 1.8750185156555166,
      "language_loss": 0.77722287,
      "learning_rate": 3.4122843924734834e-06,
      "loss": 0.79916382,
      "num_input_tokens_seen": 48730670,
      "step": 2270,
      "time_per_iteration": 3.0616073608398438
    },
    {
      "auxiliary_loss_clip": 0.01132053,
      "auxiliary_loss_mlp": 0.01034307,
      "balance_loss_clip": 1.04896021,
      "balance_loss_mlp": 1.02432907,
      "epoch": 0.2730716046413756,
      "flos": 19094637421440.0,
      "grad_norm": 2.0913188961620564,
      "language_loss": 0.88013625,
      "learning_rate": 3.411732718871319e-06,
      "loss": 0.90179986,
      "num_input_tokens_seen": 48746510,
      "step": 2271,
      "time_per_iteration": 2.7409424781799316
    },
    {
      "auxiliary_loss_clip": 0.0117921,
      "auxiliary_loss_mlp": 0.01032416,
      "balance_loss_clip": 1.05851591,
      "balance_loss_mlp": 1.02366865,
      "epoch": 0.27319184753201464,
      "flos": 26944566474240.0,
      "grad_norm": 1.4978811398988434,
      "language_loss": 0.78740281,
      "learning_rate": 3.4111808311104227e-06,
      "loss": 0.80951905,
      "num_input_tokens_seen": 48768825,
      "step": 2272,
      "time_per_iteration": 2.729586601257324
    },
    {
      "auxiliary_loss_clip": 0.01104791,
      "auxiliary_loss_mlp": 0.01040353,
      "balance_loss_clip": 1.05060399,
      "balance_loss_mlp": 1.03015471,
      "epoch": 0.27331209042265375,
      "flos": 31759828012800.0,
      "grad_norm": 2.026443502861916,
      "language_loss": 0.69598234,
      "learning_rate": 3.410628729274517e-06,
      "loss": 0.71743375,
      "num_input_tokens_seen": 48790345,
      "step": 2273,
      "time_per_iteration": 2.8639280796051025
    },
    {
      "auxiliary_loss_clip": 0.01099869,
      "auxiliary_loss_mlp": 0.01031783,
      "balance_loss_clip": 1.03898978,
      "balance_loss_mlp": 1.02259767,
      "epoch": 0.27343233331329286,
      "flos": 25739081107200.0,
      "grad_norm": 2.357521932502255,
      "language_loss": 0.825459,
      "learning_rate": 3.4100764134473546e-06,
      "loss": 0.84677553,
      "num_input_tokens_seen": 48809630,
      "step": 2274,
      "time_per_iteration": 2.866203546524048
    },
    {
      "auxiliary_loss_clip": 0.0115051,
      "auxiliary_loss_mlp": 0.01031019,
      "balance_loss_clip": 1.05205429,
      "balance_loss_mlp": 1.02141106,
      "epoch": 0.2735525762039319,
      "flos": 24389414547840.0,
      "grad_norm": 2.586882444991514,
      "language_loss": 0.85006434,
      "learning_rate": 3.4095238837127215e-06,
      "loss": 0.87187964,
      "num_input_tokens_seen": 48828770,
      "step": 2275,
      "time_per_iteration": 2.8127849102020264
    },
    {
      "auxiliary_loss_clip": 0.01168708,
      "auxiliary_loss_mlp": 0.01041257,
      "balance_loss_clip": 1.05321348,
      "balance_loss_mlp": 1.03149998,
      "epoch": 0.27367281909457103,
      "flos": 14465357527680.0,
      "grad_norm": 2.126226080803989,
      "language_loss": 0.79106086,
      "learning_rate": 3.4089711401544355e-06,
      "loss": 0.81316054,
      "num_input_tokens_seen": 48846365,
      "step": 2276,
      "time_per_iteration": 2.724574565887451
    },
    {
      "auxiliary_loss_clip": 0.01123475,
      "auxiliary_loss_mlp": 0.00763749,
      "balance_loss_clip": 1.04680121,
      "balance_loss_mlp": 1.00019479,
      "epoch": 0.27379306198521014,
      "flos": 23476996247040.0,
      "grad_norm": 2.4480899482198493,
      "language_loss": 0.67455482,
      "learning_rate": 3.4084181828563486e-06,
      "loss": 0.69342703,
      "num_input_tokens_seen": 48863085,
      "step": 2277,
      "time_per_iteration": 2.8047549724578857
    },
    {
      "auxiliary_loss_clip": 0.01202118,
      "auxiliary_loss_mlp": 0.01028818,
      "balance_loss_clip": 1.05863011,
      "balance_loss_mlp": 1.0194782,
      "epoch": 0.2739133048758492,
      "flos": 17458152762240.0,
      "grad_norm": 1.6629860327995796,
      "language_loss": 0.70314646,
      "learning_rate": 3.4078650119023428e-06,
      "loss": 0.72545582,
      "num_input_tokens_seen": 48881400,
      "step": 2278,
      "time_per_iteration": 2.6151230335235596
    },
    {
      "auxiliary_loss_clip": 0.01120052,
      "auxiliary_loss_mlp": 0.01042202,
      "balance_loss_clip": 1.0484457,
      "balance_loss_mlp": 1.03195024,
      "epoch": 0.2740335477664883,
      "flos": 19273113123840.0,
      "grad_norm": 2.361666781704344,
      "language_loss": 0.74188608,
      "learning_rate": 3.4073116273763337e-06,
      "loss": 0.76350862,
      "num_input_tokens_seen": 48895845,
      "step": 2279,
      "time_per_iteration": 2.735368013381958
    },
    {
      "auxiliary_loss_clip": 0.01143194,
      "auxiliary_loss_mlp": 0.01033679,
      "balance_loss_clip": 1.05101824,
      "balance_loss_mlp": 1.02315867,
      "epoch": 0.2741537906571274,
      "flos": 26104723603200.0,
      "grad_norm": 2.0928344290933465,
      "language_loss": 0.80975497,
      "learning_rate": 3.40675802936227e-06,
      "loss": 0.83152366,
      "num_input_tokens_seen": 48916630,
      "step": 2280,
      "time_per_iteration": 2.804858684539795
    },
    {
      "auxiliary_loss_clip": 0.01174595,
      "auxiliary_loss_mlp": 0.00764072,
      "balance_loss_clip": 1.05155957,
      "balance_loss_mlp": 1.00018895,
      "epoch": 0.27427403354776647,
      "flos": 34164190644480.0,
      "grad_norm": 3.1013344304708372,
      "language_loss": 0.71960193,
      "learning_rate": 3.4062042179441318e-06,
      "loss": 0.73898864,
      "num_input_tokens_seen": 48937100,
      "step": 2281,
      "time_per_iteration": 2.830155849456787
    },
    {
      "auxiliary_loss_clip": 0.01121403,
      "auxiliary_loss_mlp": 0.01030985,
      "balance_loss_clip": 1.04498923,
      "balance_loss_mlp": 1.02193117,
      "epoch": 0.2743942764384056,
      "flos": 18766988536320.0,
      "grad_norm": 1.9405885312568891,
      "language_loss": 0.80623949,
      "learning_rate": 3.4056501932059314e-06,
      "loss": 0.82776338,
      "num_input_tokens_seen": 48955175,
      "step": 2282,
      "time_per_iteration": 2.826223373413086
    },
    {
      "auxiliary_loss_clip": 0.01090695,
      "auxiliary_loss_mlp": 0.01003448,
      "balance_loss_clip": 1.04534864,
      "balance_loss_mlp": 1.00176668,
      "epoch": 0.2745145193290447,
      "flos": 64904048058240.0,
      "grad_norm": 0.7603377042040929,
      "language_loss": 0.58116025,
      "learning_rate": 3.405095955231715e-06,
      "loss": 0.60210168,
      "num_input_tokens_seen": 49006830,
      "step": 2283,
      "time_per_iteration": 3.199119806289673
    },
    {
      "auxiliary_loss_clip": 0.01158116,
      "auxiliary_loss_mlp": 0.01030204,
      "balance_loss_clip": 1.04924572,
      "balance_loss_mlp": 1.01985693,
      "epoch": 0.27463476221968375,
      "flos": 16136926796160.0,
      "grad_norm": 3.4885603155343885,
      "language_loss": 0.94325316,
      "learning_rate": 3.4045415041055585e-06,
      "loss": 0.96513641,
      "num_input_tokens_seen": 49022470,
      "step": 2284,
      "time_per_iteration": 2.6449408531188965
    },
    {
      "auxiliary_loss_clip": 0.01158231,
      "auxiliary_loss_mlp": 0.01036044,
      "balance_loss_clip": 1.04854572,
      "balance_loss_mlp": 1.02612543,
      "epoch": 0.27475500511032286,
      "flos": 10376712213120.0,
      "grad_norm": 2.100148562299794,
      "language_loss": 0.78145981,
      "learning_rate": 3.4039868399115728e-06,
      "loss": 0.80340254,
      "num_input_tokens_seen": 49037110,
      "step": 2285,
      "time_per_iteration": 2.6615288257598877
    },
    {
      "auxiliary_loss_clip": 0.01165257,
      "auxiliary_loss_mlp": 0.01039319,
      "balance_loss_clip": 1.05318928,
      "balance_loss_mlp": 1.02967453,
      "epoch": 0.27487524800096197,
      "flos": 17311062568320.0,
      "grad_norm": 1.8276650696941406,
      "language_loss": 0.80141079,
      "learning_rate": 3.4034319627339003e-06,
      "loss": 0.82345653,
      "num_input_tokens_seen": 49053975,
      "step": 2286,
      "time_per_iteration": 2.675473928451538
    },
    {
      "auxiliary_loss_clip": 0.01123047,
      "auxiliary_loss_mlp": 0.01037092,
      "balance_loss_clip": 1.04561222,
      "balance_loss_mlp": 1.02677488,
      "epoch": 0.274995490891601,
      "flos": 27120205002240.0,
      "grad_norm": 2.5904135311559786,
      "language_loss": 0.69618893,
      "learning_rate": 3.402876872656715e-06,
      "loss": 0.71779031,
      "num_input_tokens_seen": 49072295,
      "step": 2287,
      "time_per_iteration": 2.8310437202453613
    },
    {
      "auxiliary_loss_clip": 0.01098235,
      "auxiliary_loss_mlp": 0.0103358,
      "balance_loss_clip": 1.04151821,
      "balance_loss_mlp": 1.02322102,
      "epoch": 0.27511573378224013,
      "flos": 23436093634560.0,
      "grad_norm": 1.848856672139973,
      "language_loss": 0.89770103,
      "learning_rate": 3.402321569764223e-06,
      "loss": 0.91901916,
      "num_input_tokens_seen": 49091600,
      "step": 2288,
      "time_per_iteration": 5.048674821853638
    },
    {
      "auxiliary_loss_clip": 0.0111928,
      "auxiliary_loss_mlp": 0.01037229,
      "balance_loss_clip": 1.04534173,
      "balance_loss_mlp": 1.0264107,
      "epoch": 0.2752359766728792,
      "flos": 16722019434240.0,
      "grad_norm": 2.1731975382997013,
      "language_loss": 0.83458751,
      "learning_rate": 3.4017660541406635e-06,
      "loss": 0.85615265,
      "num_input_tokens_seen": 49107665,
      "step": 2289,
      "time_per_iteration": 2.82672381401062
    },
    {
      "auxiliary_loss_clip": 0.01145953,
      "auxiliary_loss_mlp": 0.01027943,
      "balance_loss_clip": 1.04500377,
      "balance_loss_mlp": 1.01784563,
      "epoch": 0.2753562195635183,
      "flos": 25297738698240.0,
      "grad_norm": 2.3440235435632393,
      "language_loss": 0.74565792,
      "learning_rate": 3.4012103258703092e-06,
      "loss": 0.76739687,
      "num_input_tokens_seen": 49126420,
      "step": 2290,
      "time_per_iteration": 2.8392069339752197
    },
    {
      "auxiliary_loss_clip": 0.01108337,
      "auxiliary_loss_mlp": 0.01029719,
      "balance_loss_clip": 1.04813969,
      "balance_loss_mlp": 1.02003336,
      "epoch": 0.2754764624541574,
      "flos": 27338972785920.0,
      "grad_norm": 2.0163440554144114,
      "language_loss": 0.83211392,
      "learning_rate": 3.4006543850374616e-06,
      "loss": 0.85349452,
      "num_input_tokens_seen": 49141470,
      "step": 2291,
      "time_per_iteration": 5.007181406021118
    },
    {
      "auxiliary_loss_clip": 0.01189936,
      "auxiliary_loss_mlp": 0.01032117,
      "balance_loss_clip": 1.05815172,
      "balance_loss_mlp": 1.02239537,
      "epoch": 0.27559670534479647,
      "flos": 17238379397760.0,
      "grad_norm": 2.642601015442636,
      "language_loss": 0.75109458,
      "learning_rate": 3.400098231726458e-06,
      "loss": 0.77331507,
      "num_input_tokens_seen": 49158570,
      "step": 2292,
      "time_per_iteration": 2.624467134475708
    },
    {
      "auxiliary_loss_clip": 0.01178691,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.05425811,
      "balance_loss_mlp": 1.02007604,
      "epoch": 0.2757169482354356,
      "flos": 21939085486080.0,
      "grad_norm": 1.7837598775308494,
      "language_loss": 0.86689967,
      "learning_rate": 3.3995418660216657e-06,
      "loss": 0.88899302,
      "num_input_tokens_seen": 49176025,
      "step": 2293,
      "time_per_iteration": 2.6781623363494873
    },
    {
      "auxiliary_loss_clip": 0.01131765,
      "auxiliary_loss_mlp": 0.01024965,
      "balance_loss_clip": 1.05426681,
      "balance_loss_mlp": 1.01471901,
      "epoch": 0.2758371911260747,
      "flos": 20850669521280.0,
      "grad_norm": 2.554763975278054,
      "language_loss": 0.80602956,
      "learning_rate": 3.3989852880074848e-06,
      "loss": 0.8275969,
      "num_input_tokens_seen": 49197455,
      "step": 2294,
      "time_per_iteration": 2.8064236640930176
    },
    {
      "auxiliary_loss_clip": 0.01072766,
      "auxiliary_loss_mlp": 0.01002402,
      "balance_loss_clip": 1.01804435,
      "balance_loss_mlp": 1.00095367,
      "epoch": 0.27595743401671374,
      "flos": 69269063592960.0,
      "grad_norm": 0.9065840818635238,
      "language_loss": 0.60577333,
      "learning_rate": 3.398428497768348e-06,
      "loss": 0.62652493,
      "num_input_tokens_seen": 49262625,
      "step": 2295,
      "time_per_iteration": 3.539659261703491
    },
    {
      "auxiliary_loss_clip": 0.01140156,
      "auxiliary_loss_mlp": 0.01036564,
      "balance_loss_clip": 1.04855943,
      "balance_loss_mlp": 1.02619827,
      "epoch": 0.27607767690735285,
      "flos": 21215019127680.0,
      "grad_norm": 1.664763926585048,
      "language_loss": 0.71889615,
      "learning_rate": 3.3978714953887205e-06,
      "loss": 0.74066329,
      "num_input_tokens_seen": 49282380,
      "step": 2296,
      "time_per_iteration": 2.7773478031158447
    },
    {
      "auxiliary_loss_clip": 0.01066006,
      "auxiliary_loss_mlp": 0.01030428,
      "balance_loss_clip": 1.0442915,
      "balance_loss_mlp": 1.02090907,
      "epoch": 0.27619791979799196,
      "flos": 24825334003200.0,
      "grad_norm": 1.7267262356302868,
      "language_loss": 0.85938144,
      "learning_rate": 3.397314280953098e-06,
      "loss": 0.88034582,
      "num_input_tokens_seen": 49303205,
      "step": 2297,
      "time_per_iteration": 3.270078659057617
    },
    {
      "auxiliary_loss_clip": 0.01149743,
      "auxiliary_loss_mlp": 0.01036653,
      "balance_loss_clip": 1.05187464,
      "balance_loss_mlp": 1.02689576,
      "epoch": 0.276318162688631,
      "flos": 24753548672640.0,
      "grad_norm": 1.7982725332956133,
      "language_loss": 0.79859298,
      "learning_rate": 3.3967568545460108e-06,
      "loss": 0.82045692,
      "num_input_tokens_seen": 49322745,
      "step": 2298,
      "time_per_iteration": 3.047654151916504
    },
    {
      "auxiliary_loss_clip": 0.01149177,
      "auxiliary_loss_mlp": 0.01032845,
      "balance_loss_clip": 1.04501927,
      "balance_loss_mlp": 1.02356482,
      "epoch": 0.27643840557927013,
      "flos": 18150007599360.0,
      "grad_norm": 1.860729578887045,
      "language_loss": 0.80391753,
      "learning_rate": 3.3961992162520185e-06,
      "loss": 0.82573771,
      "num_input_tokens_seen": 49341370,
      "step": 2299,
      "time_per_iteration": 2.6943047046661377
    },
    {
      "auxiliary_loss_clip": 0.01189381,
      "auxiliary_loss_mlp": 0.01035976,
      "balance_loss_clip": 1.05345821,
      "balance_loss_mlp": 1.0263145,
      "epoch": 0.27655864846990924,
      "flos": 24823933372800.0,
      "grad_norm": 2.0758711593881096,
      "language_loss": 0.71936995,
      "learning_rate": 3.3956413661557156e-06,
      "loss": 0.74162352,
      "num_input_tokens_seen": 49361545,
      "step": 2300,
      "time_per_iteration": 2.68784236907959
    },
    {
      "auxiliary_loss_clip": 0.01182499,
      "auxiliary_loss_mlp": 0.0102629,
      "balance_loss_clip": 1.05470979,
      "balance_loss_mlp": 1.01635408,
      "epoch": 0.2766788913605483,
      "flos": 20266582464000.0,
      "grad_norm": 3.223559114163457,
      "language_loss": 0.65958703,
      "learning_rate": 3.3950833043417273e-06,
      "loss": 0.68167484,
      "num_input_tokens_seen": 49379690,
      "step": 2301,
      "time_per_iteration": 2.7451610565185547
    },
    {
      "auxiliary_loss_clip": 0.01133605,
      "auxiliary_loss_mlp": 0.01038334,
      "balance_loss_clip": 1.05048752,
      "balance_loss_mlp": 1.02764702,
      "epoch": 0.2767991342511874,
      "flos": 21470272151040.0,
      "grad_norm": 3.2534513189219427,
      "language_loss": 0.72755778,
      "learning_rate": 3.3945250308947105e-06,
      "loss": 0.74927711,
      "num_input_tokens_seen": 49395995,
      "step": 2302,
      "time_per_iteration": 2.723289728164673
    },
    {
      "auxiliary_loss_clip": 0.01052577,
      "auxiliary_loss_mlp": 0.01002576,
      "balance_loss_clip": 1.01311696,
      "balance_loss_mlp": 1.001122,
      "epoch": 0.2769193771418265,
      "flos": 66002627571840.0,
      "grad_norm": 1.2792640625682665,
      "language_loss": 0.68385589,
      "learning_rate": 3.3939665458993556e-06,
      "loss": 0.70440733,
      "num_input_tokens_seen": 49450415,
      "step": 2303,
      "time_per_iteration": 3.2308835983276367
    },
    {
      "auxiliary_loss_clip": 0.01117273,
      "auxiliary_loss_mlp": 0.0102902,
      "balance_loss_clip": 1.05134666,
      "balance_loss_mlp": 1.01930451,
      "epoch": 0.27703962003246557,
      "flos": 20704441253760.0,
      "grad_norm": 2.125991708334389,
      "language_loss": 0.7679711,
      "learning_rate": 3.3934078494403843e-06,
      "loss": 0.78943408,
      "num_input_tokens_seen": 49469990,
      "step": 2304,
      "time_per_iteration": 2.778982162475586
    },
    {
      "auxiliary_loss_clip": 0.01192313,
      "auxiliary_loss_mlp": 0.01035501,
      "balance_loss_clip": 1.06081641,
      "balance_loss_mlp": 1.02468288,
      "epoch": 0.2771598629231047,
      "flos": 22929897219840.0,
      "grad_norm": 3.1033748161704353,
      "language_loss": 0.8173669,
      "learning_rate": 3.3928489416025495e-06,
      "loss": 0.83964503,
      "num_input_tokens_seen": 49490835,
      "step": 2305,
      "time_per_iteration": 2.6960599422454834
    },
    {
      "auxiliary_loss_clip": 0.0115466,
      "auxiliary_loss_mlp": 0.01032532,
      "balance_loss_clip": 1.05369496,
      "balance_loss_mlp": 1.02258396,
      "epoch": 0.27728010581374374,
      "flos": 18369457741440.0,
      "grad_norm": 2.738241832606527,
      "language_loss": 0.78622824,
      "learning_rate": 3.392289822470638e-06,
      "loss": 0.80810016,
      "num_input_tokens_seen": 49508815,
      "step": 2306,
      "time_per_iteration": 2.679733991622925
    },
    {
      "auxiliary_loss_clip": 0.01203794,
      "auxiliary_loss_mlp": 0.00764112,
      "balance_loss_clip": 1.05802369,
      "balance_loss_mlp": 1.000108,
      "epoch": 0.27740034870438285,
      "flos": 19427637432960.0,
      "grad_norm": 2.345545875754048,
      "language_loss": 0.75623477,
      "learning_rate": 3.3917304921294674e-06,
      "loss": 0.77591377,
      "num_input_tokens_seen": 49526980,
      "step": 2307,
      "time_per_iteration": 2.673004627227783
    },
    {
      "auxiliary_loss_clip": 0.01146943,
      "auxiliary_loss_mlp": 0.01041406,
      "balance_loss_clip": 1.04885089,
      "balance_loss_mlp": 1.02952695,
      "epoch": 0.27752059159502196,
      "flos": 21614776565760.0,
      "grad_norm": 1.7088311376015601,
      "language_loss": 0.80700243,
      "learning_rate": 3.3911709506638876e-06,
      "loss": 0.82888591,
      "num_input_tokens_seen": 49546290,
      "step": 2308,
      "time_per_iteration": 2.764427423477173
    },
    {
      "auxiliary_loss_clip": 0.01160065,
      "auxiliary_loss_mlp": 0.01029338,
      "balance_loss_clip": 1.05334187,
      "balance_loss_mlp": 1.01877606,
      "epoch": 0.277640834485661,
      "flos": 26608011016320.0,
      "grad_norm": 2.466719063356355,
      "language_loss": 0.8133198,
      "learning_rate": 3.390611198158781e-06,
      "loss": 0.8352139,
      "num_input_tokens_seen": 49564165,
      "step": 2309,
      "time_per_iteration": 2.729781150817871
    },
    {
      "auxiliary_loss_clip": 0.01186142,
      "auxiliary_loss_mlp": 0.00764891,
      "balance_loss_clip": 1.05689728,
      "balance_loss_mlp": 1.00013018,
      "epoch": 0.2777610773763001,
      "flos": 19492814661120.0,
      "grad_norm": 2.129004217576595,
      "language_loss": 0.9004966,
      "learning_rate": 3.3900512346990612e-06,
      "loss": 0.92000687,
      "num_input_tokens_seen": 49580155,
      "step": 2310,
      "time_per_iteration": 2.7065343856811523
    },
    {
      "auxiliary_loss_clip": 0.01174876,
      "auxiliary_loss_mlp": 0.01035043,
      "balance_loss_clip": 1.05735147,
      "balance_loss_mlp": 1.02427828,
      "epoch": 0.27788132026693924,
      "flos": 38290650001920.0,
      "grad_norm": 1.8163835490378635,
      "language_loss": 0.65348125,
      "learning_rate": 3.389491060369674e-06,
      "loss": 0.6755805,
      "num_input_tokens_seen": 49605830,
      "step": 2311,
      "time_per_iteration": 2.8209280967712402
    },
    {
      "auxiliary_loss_clip": 0.01100352,
      "auxiliary_loss_mlp": 0.01033513,
      "balance_loss_clip": 1.04697633,
      "balance_loss_mlp": 1.0241071,
      "epoch": 0.2780015631575783,
      "flos": 22382546797440.0,
      "grad_norm": 5.747054758045989,
      "language_loss": 0.89534724,
      "learning_rate": 3.388930675255598e-06,
      "loss": 0.91668594,
      "num_input_tokens_seen": 49625680,
      "step": 2312,
      "time_per_iteration": 2.832345485687256
    },
    {
      "auxiliary_loss_clip": 0.01144247,
      "auxiliary_loss_mlp": 0.01029982,
      "balance_loss_clip": 1.04990602,
      "balance_loss_mlp": 1.02018332,
      "epoch": 0.2781218060482174,
      "flos": 12203200840320.0,
      "grad_norm": 2.790287537490699,
      "language_loss": 0.79747272,
      "learning_rate": 3.388370079441843e-06,
      "loss": 0.81921494,
      "num_input_tokens_seen": 49641195,
      "step": 2313,
      "time_per_iteration": 4.417813539505005
    },
    {
      "auxiliary_loss_clip": 0.01144373,
      "auxiliary_loss_mlp": 0.00764292,
      "balance_loss_clip": 1.05467308,
      "balance_loss_mlp": 1.00017285,
      "epoch": 0.2782420489388565,
      "flos": 18107632529280.0,
      "grad_norm": 2.1764669617306387,
      "language_loss": 0.92651129,
      "learning_rate": 3.3878092730134505e-06,
      "loss": 0.94559789,
      "num_input_tokens_seen": 49659180,
      "step": 2314,
      "time_per_iteration": 3.674851417541504
    },
    {
      "auxiliary_loss_clip": 0.01127608,
      "auxiliary_loss_mlp": 0.01031685,
      "balance_loss_clip": 1.05255258,
      "balance_loss_mlp": 1.02084875,
      "epoch": 0.27836229182949557,
      "flos": 18514752255360.0,
      "grad_norm": 1.781609516880436,
      "language_loss": 0.80710059,
      "learning_rate": 3.3872482560554947e-06,
      "loss": 0.82869351,
      "num_input_tokens_seen": 49677955,
      "step": 2315,
      "time_per_iteration": 2.8475232124328613
    },
    {
      "auxiliary_loss_clip": 0.01078766,
      "auxiliary_loss_mlp": 0.01002569,
      "balance_loss_clip": 1.01616037,
      "balance_loss_mlp": 1.0010308,
      "epoch": 0.2784825347201347,
      "flos": 67079230940160.0,
      "grad_norm": 0.8123526026628693,
      "language_loss": 0.56956995,
      "learning_rate": 3.386687028653082e-06,
      "loss": 0.59038329,
      "num_input_tokens_seen": 49740800,
      "step": 2316,
      "time_per_iteration": 4.170092344284058
    },
    {
      "auxiliary_loss_clip": 0.01143376,
      "auxiliary_loss_mlp": 0.0076468,
      "balance_loss_clip": 1.05159807,
      "balance_loss_mlp": 1.00015545,
      "epoch": 0.2786027776107738,
      "flos": 22631119891200.0,
      "grad_norm": 1.901995766931992,
      "language_loss": 0.85329294,
      "learning_rate": 3.386125590891349e-06,
      "loss": 0.87237346,
      "num_input_tokens_seen": 49757675,
      "step": 2317,
      "time_per_iteration": 3.745311975479126
    },
    {
      "auxiliary_loss_clip": 0.01171369,
      "auxiliary_loss_mlp": 0.01037274,
      "balance_loss_clip": 1.05672753,
      "balance_loss_mlp": 1.02835727,
      "epoch": 0.27872302050141284,
      "flos": 15778826156160.0,
      "grad_norm": 2.4990136440421558,
      "language_loss": 0.83155239,
      "learning_rate": 3.3855639428554657e-06,
      "loss": 0.85363877,
      "num_input_tokens_seen": 49775205,
      "step": 2318,
      "time_per_iteration": 2.7109031677246094
    },
    {
      "auxiliary_loss_clip": 0.01146371,
      "auxiliary_loss_mlp": 0.00764368,
      "balance_loss_clip": 1.05314398,
      "balance_loss_mlp": 1.00015652,
      "epoch": 0.27884326339205195,
      "flos": 22126970551680.0,
      "grad_norm": 1.9567801749646097,
      "language_loss": 0.80143678,
      "learning_rate": 3.385002084630635e-06,
      "loss": 0.82054412,
      "num_input_tokens_seen": 49794175,
      "step": 2319,
      "time_per_iteration": 2.859862804412842
    },
    {
      "auxiliary_loss_clip": 0.01189844,
      "auxiliary_loss_mlp": 0.00764271,
      "balance_loss_clip": 1.06093788,
      "balance_loss_mlp": 1.00016868,
      "epoch": 0.278963506282691,
      "flos": 20558715776640.0,
      "grad_norm": 2.5550559949341833,
      "language_loss": 0.84810287,
      "learning_rate": 3.384440016302088e-06,
      "loss": 0.86764401,
      "num_input_tokens_seen": 49812850,
      "step": 2320,
      "time_per_iteration": 2.701064109802246
    },
    {
      "auxiliary_loss_clip": 0.01169466,
      "auxiliary_loss_mlp": 0.00764937,
      "balance_loss_clip": 1.05421853,
      "balance_loss_mlp": 1.000126,
      "epoch": 0.2790837491733301,
      "flos": 21942928241280.0,
      "grad_norm": 2.318511800191822,
      "language_loss": 0.62336934,
      "learning_rate": 3.3838777379550923e-06,
      "loss": 0.64271337,
      "num_input_tokens_seen": 49832295,
      "step": 2321,
      "time_per_iteration": 2.7426466941833496
    },
    {
      "auxiliary_loss_clip": 0.01178659,
      "auxiliary_loss_mlp": 0.0076454,
      "balance_loss_clip": 1.05531931,
      "balance_loss_mlp": 1.00014925,
      "epoch": 0.27920399206396923,
      "flos": 26286790665600.0,
      "grad_norm": 1.9544938761706252,
      "language_loss": 0.78391582,
      "learning_rate": 3.383315249674944e-06,
      "loss": 0.80334783,
      "num_input_tokens_seen": 49850860,
      "step": 2322,
      "time_per_iteration": 2.7780566215515137
    },
    {
      "auxiliary_loss_clip": 0.01124102,
      "auxiliary_loss_mlp": 0.01032397,
      "balance_loss_clip": 1.04744923,
      "balance_loss_mlp": 1.02239513,
      "epoch": 0.2793242349546083,
      "flos": 25400981364480.0,
      "grad_norm": 2.102402052880469,
      "language_loss": 0.85868883,
      "learning_rate": 3.3827525515469715e-06,
      "loss": 0.88025379,
      "num_input_tokens_seen": 49865765,
      "step": 2323,
      "time_per_iteration": 2.8356776237487793
    },
    {
      "auxiliary_loss_clip": 0.01156181,
      "auxiliary_loss_mlp": 0.01032569,
      "balance_loss_clip": 1.05328834,
      "balance_loss_mlp": 1.02192998,
      "epoch": 0.2794444778452474,
      "flos": 20850346298880.0,
      "grad_norm": 2.4325896204254476,
      "language_loss": 0.71622324,
      "learning_rate": 3.3821896436565367e-06,
      "loss": 0.73811078,
      "num_input_tokens_seen": 49885425,
      "step": 2324,
      "time_per_iteration": 2.6948115825653076
    },
    {
      "auxiliary_loss_clip": 0.01135023,
      "auxiliary_loss_mlp": 0.01037406,
      "balance_loss_clip": 1.05788016,
      "balance_loss_mlp": 1.02736306,
      "epoch": 0.2795647207358865,
      "flos": 21576244250880.0,
      "grad_norm": 1.8768172875111335,
      "language_loss": 0.70164615,
      "learning_rate": 3.381626526089032e-06,
      "loss": 0.72337043,
      "num_input_tokens_seen": 49904990,
      "step": 2325,
      "time_per_iteration": 2.7957029342651367
    },
    {
      "auxiliary_loss_clip": 0.011641,
      "auxiliary_loss_mlp": 0.00765328,
      "balance_loss_clip": 1.0541743,
      "balance_loss_mlp": 1.00013709,
      "epoch": 0.27968496362652556,
      "flos": 21471744608640.0,
      "grad_norm": 2.0330467014744102,
      "language_loss": 0.79014754,
      "learning_rate": 3.3810631989298815e-06,
      "loss": 0.80944186,
      "num_input_tokens_seen": 49924600,
      "step": 2326,
      "time_per_iteration": 2.7104907035827637
    },
    {
      "auxiliary_loss_clip": 0.01134957,
      "auxiliary_loss_mlp": 0.01037055,
      "balance_loss_clip": 1.05219483,
      "balance_loss_mlp": 1.02472305,
      "epoch": 0.2798052065171647,
      "flos": 23258695340160.0,
      "grad_norm": 5.395684900669667,
      "language_loss": 0.84940362,
      "learning_rate": 3.3804996622645423e-06,
      "loss": 0.87112379,
      "num_input_tokens_seen": 49942600,
      "step": 2327,
      "time_per_iteration": 2.8133366107940674
    },
    {
      "auxiliary_loss_clip": 0.01139189,
      "auxiliary_loss_mlp": 0.01031639,
      "balance_loss_clip": 1.0507704,
      "balance_loss_mlp": 1.0214287,
      "epoch": 0.2799254494078038,
      "flos": 21539328048000.0,
      "grad_norm": 2.108839627372388,
      "language_loss": 0.90038663,
      "learning_rate": 3.3799359161785015e-06,
      "loss": 0.92209494,
      "num_input_tokens_seen": 49962250,
      "step": 2328,
      "time_per_iteration": 2.865649700164795
    },
    {
      "auxiliary_loss_clip": 0.01149116,
      "auxiliary_loss_mlp": 0.01034848,
      "balance_loss_clip": 1.05425727,
      "balance_loss_mlp": 1.02433968,
      "epoch": 0.28004569229844284,
      "flos": 26393912000640.0,
      "grad_norm": 1.5676295699971585,
      "language_loss": 0.85967731,
      "learning_rate": 3.3793719607572798e-06,
      "loss": 0.88151693,
      "num_input_tokens_seen": 49983215,
      "step": 2329,
      "time_per_iteration": 2.7522480487823486
    },
    {
      "auxiliary_loss_clip": 0.01188942,
      "auxiliary_loss_mlp": 0.01035214,
      "balance_loss_clip": 1.05735433,
      "balance_loss_mlp": 1.02472997,
      "epoch": 0.28016593518908195,
      "flos": 33547676584320.0,
      "grad_norm": 2.096107270257157,
      "language_loss": 0.77101922,
      "learning_rate": 3.378807796086428e-06,
      "loss": 0.79326075,
      "num_input_tokens_seen": 50006075,
      "step": 2330,
      "time_per_iteration": 2.7771871089935303
    },
    {
      "auxiliary_loss_clip": 0.01151586,
      "auxiliary_loss_mlp": 0.01039612,
      "balance_loss_clip": 1.05189085,
      "balance_loss_mlp": 1.02931213,
      "epoch": 0.28028617807972106,
      "flos": 15340823712000.0,
      "grad_norm": 2.130756678297092,
      "language_loss": 0.7683531,
      "learning_rate": 3.37824342225153e-06,
      "loss": 0.79026502,
      "num_input_tokens_seen": 50022495,
      "step": 2331,
      "time_per_iteration": 2.750938653945923
    },
    {
      "auxiliary_loss_clip": 0.01146016,
      "auxiliary_loss_mlp": 0.01028097,
      "balance_loss_clip": 1.05140567,
      "balance_loss_mlp": 1.01871514,
      "epoch": 0.2804064209703601,
      "flos": 25520277409920.0,
      "grad_norm": 1.838416592861169,
      "language_loss": 0.7771188,
      "learning_rate": 3.3776788393382006e-06,
      "loss": 0.79885989,
      "num_input_tokens_seen": 50041975,
      "step": 2332,
      "time_per_iteration": 2.799628734588623
    },
    {
      "auxiliary_loss_clip": 0.01138259,
      "auxiliary_loss_mlp": 0.01038145,
      "balance_loss_clip": 1.04862773,
      "balance_loss_mlp": 1.02623582,
      "epoch": 0.2805266638609992,
      "flos": 29351766280320.0,
      "grad_norm": 2.449657917055747,
      "language_loss": 0.76569718,
      "learning_rate": 3.3771140474320872e-06,
      "loss": 0.78746122,
      "num_input_tokens_seen": 50061925,
      "step": 2333,
      "time_per_iteration": 2.8521535396575928
    },
    {
      "auxiliary_loss_clip": 0.01174597,
      "auxiliary_loss_mlp": 0.01040973,
      "balance_loss_clip": 1.05715609,
      "balance_loss_mlp": 1.0306673,
      "epoch": 0.28064690675163834,
      "flos": 21463735875840.0,
      "grad_norm": 1.924590268976589,
      "language_loss": 0.79371667,
      "learning_rate": 3.3765490466188664e-06,
      "loss": 0.81587243,
      "num_input_tokens_seen": 50079325,
      "step": 2334,
      "time_per_iteration": 2.778287887573242
    },
    {
      "auxiliary_loss_clip": 0.01179717,
      "auxiliary_loss_mlp": 0.01040789,
      "balance_loss_clip": 1.05764759,
      "balance_loss_mlp": 1.03008425,
      "epoch": 0.2807671496422774,
      "flos": 20995640812800.0,
      "grad_norm": 2.619489489574715,
      "language_loss": 0.73760968,
      "learning_rate": 3.3759838369842508e-06,
      "loss": 0.75981474,
      "num_input_tokens_seen": 50097400,
      "step": 2335,
      "time_per_iteration": 2.7499279975891113
    },
    {
      "auxiliary_loss_clip": 0.01110273,
      "auxiliary_loss_mlp": 0.01032661,
      "balance_loss_clip": 1.05097961,
      "balance_loss_mlp": 1.0226593,
      "epoch": 0.2808873925329165,
      "flos": 21506577822720.0,
      "grad_norm": 5.437385086686854,
      "language_loss": 0.73161566,
      "learning_rate": 3.375418418613981e-06,
      "loss": 0.75304496,
      "num_input_tokens_seen": 50116425,
      "step": 2336,
      "time_per_iteration": 2.951927900314331
    },
    {
      "auxiliary_loss_clip": 0.01147017,
      "auxiliary_loss_mlp": 0.00765362,
      "balance_loss_clip": 1.04934788,
      "balance_loss_mlp": 1.00013435,
      "epoch": 0.28100763542355556,
      "flos": 16070815814400.0,
      "grad_norm": 2.2365382637858855,
      "language_loss": 0.83730286,
      "learning_rate": 3.374852791593831e-06,
      "loss": 0.85642666,
      "num_input_tokens_seen": 50132625,
      "step": 2337,
      "time_per_iteration": 3.2870430946350098
    },
    {
      "auxiliary_loss_clip": 0.01173295,
      "auxiliary_loss_mlp": 0.00765183,
      "balance_loss_clip": 1.05623269,
      "balance_loss_mlp": 1.00016117,
      "epoch": 0.28112787831419467,
      "flos": 19062605468160.0,
      "grad_norm": 4.635040696810698,
      "language_loss": 0.54247743,
      "learning_rate": 3.374286956009605e-06,
      "loss": 0.56186223,
      "num_input_tokens_seen": 50151190,
      "step": 2338,
      "time_per_iteration": 2.8077406883239746
    },
    {
      "auxiliary_loss_clip": 0.01149144,
      "auxiliary_loss_mlp": 0.01032079,
      "balance_loss_clip": 1.04964936,
      "balance_loss_mlp": 1.02180958,
      "epoch": 0.2812481212048338,
      "flos": 12823629482880.0,
      "grad_norm": 3.3291585824455816,
      "language_loss": 0.75471509,
      "learning_rate": 3.3737209119471405e-06,
      "loss": 0.77652735,
      "num_input_tokens_seen": 50167700,
      "step": 2339,
      "time_per_iteration": 3.661303758621216
    },
    {
      "auxiliary_loss_clip": 0.01171988,
      "auxiliary_loss_mlp": 0.01041207,
      "balance_loss_clip": 1.05408335,
      "balance_loss_mlp": 1.029948,
      "epoch": 0.28136836409547283,
      "flos": 15633064765440.0,
      "grad_norm": 2.379495852864982,
      "language_loss": 0.63860589,
      "learning_rate": 3.373154659492306e-06,
      "loss": 0.66073787,
      "num_input_tokens_seen": 50185840,
      "step": 2340,
      "time_per_iteration": 3.8763482570648193
    },
    {
      "auxiliary_loss_clip": 0.01194736,
      "auxiliary_loss_mlp": 0.01033629,
      "balance_loss_clip": 1.05889082,
      "balance_loss_mlp": 1.0231626,
      "epoch": 0.28148860698611194,
      "flos": 19933726106880.0,
      "grad_norm": 1.791675567405481,
      "language_loss": 0.85298955,
      "learning_rate": 3.3725881987310016e-06,
      "loss": 0.87527323,
      "num_input_tokens_seen": 50203375,
      "step": 2341,
      "time_per_iteration": 2.6649868488311768
    },
    {
      "auxiliary_loss_clip": 0.01173568,
      "auxiliary_loss_mlp": 0.01028813,
      "balance_loss_clip": 1.05677402,
      "balance_loss_mlp": 1.01884067,
      "epoch": 0.28160884987675106,
      "flos": 17457219008640.0,
      "grad_norm": 2.389052755715976,
      "language_loss": 0.87739247,
      "learning_rate": 3.372021529749159e-06,
      "loss": 0.89941633,
      "num_input_tokens_seen": 50222435,
      "step": 2342,
      "time_per_iteration": 3.586425542831421
    },
    {
      "auxiliary_loss_clip": 0.01129921,
      "auxiliary_loss_mlp": 0.0103496,
      "balance_loss_clip": 1.04765379,
      "balance_loss_mlp": 1.02476776,
      "epoch": 0.2817290927673901,
      "flos": 16834743290880.0,
      "grad_norm": 3.240896528348502,
      "language_loss": 0.9240796,
      "learning_rate": 3.3714546526327405e-06,
      "loss": 0.94572836,
      "num_input_tokens_seen": 50240435,
      "step": 2343,
      "time_per_iteration": 3.7473433017730713
    },
    {
      "auxiliary_loss_clip": 0.01186351,
      "auxiliary_loss_mlp": 0.01031414,
      "balance_loss_clip": 1.05747104,
      "balance_loss_mlp": 1.02070284,
      "epoch": 0.2818493356580292,
      "flos": 15414081500160.0,
      "grad_norm": 3.5097861625522384,
      "language_loss": 0.87850404,
      "learning_rate": 3.3708875674677423e-06,
      "loss": 0.90068161,
      "num_input_tokens_seen": 50258410,
      "step": 2344,
      "time_per_iteration": 2.71134877204895
    },
    {
      "auxiliary_loss_clip": 0.01181331,
      "auxiliary_loss_mlp": 0.00765622,
      "balance_loss_clip": 1.06044436,
      "balance_loss_mlp": 1.00020456,
      "epoch": 0.28196957854866833,
      "flos": 20412451595520.0,
      "grad_norm": 3.321038532443056,
      "language_loss": 0.83150196,
      "learning_rate": 3.37032027434019e-06,
      "loss": 0.85097158,
      "num_input_tokens_seen": 50277930,
      "step": 2345,
      "time_per_iteration": 2.7391436100006104
    },
    {
      "auxiliary_loss_clip": 0.01153245,
      "auxiliary_loss_mlp": 0.01041869,
      "balance_loss_clip": 1.04961514,
      "balance_loss_mlp": 1.02914906,
      "epoch": 0.2820898214393074,
      "flos": 19973120348160.0,
      "grad_norm": 1.7139507727384782,
      "language_loss": 0.82923901,
      "learning_rate": 3.369752773336141e-06,
      "loss": 0.85119015,
      "num_input_tokens_seen": 50297410,
      "step": 2346,
      "time_per_iteration": 2.7361955642700195
    },
    {
      "auxiliary_loss_clip": 0.0114439,
      "auxiliary_loss_mlp": 0.01033724,
      "balance_loss_clip": 1.05344772,
      "balance_loss_mlp": 1.02175546,
      "epoch": 0.2822100643299465,
      "flos": 22528308188160.0,
      "grad_norm": 1.7747997978865597,
      "language_loss": 0.78546333,
      "learning_rate": 3.3691850645416864e-06,
      "loss": 0.80724448,
      "num_input_tokens_seen": 50317120,
      "step": 2347,
      "time_per_iteration": 2.7478253841400146
    },
    {
      "auxiliary_loss_clip": 0.01159099,
      "auxiliary_loss_mlp": 0.00765137,
      "balance_loss_clip": 1.05606747,
      "balance_loss_mlp": 1.00016522,
      "epoch": 0.2823303072205856,
      "flos": 11546682007680.0,
      "grad_norm": 1.9813521371497003,
      "language_loss": 0.82885557,
      "learning_rate": 3.368617148042945e-06,
      "loss": 0.84809792,
      "num_input_tokens_seen": 50334790,
      "step": 2348,
      "time_per_iteration": 2.7115478515625
    },
    {
      "auxiliary_loss_clip": 0.01163022,
      "auxiliary_loss_mlp": 0.00765469,
      "balance_loss_clip": 1.05757356,
      "balance_loss_mlp": 1.00014615,
      "epoch": 0.28245055011122466,
      "flos": 18259894281600.0,
      "grad_norm": 2.101254113539135,
      "language_loss": 0.84607708,
      "learning_rate": 3.368049023926071e-06,
      "loss": 0.86536199,
      "num_input_tokens_seen": 50353785,
      "step": 2349,
      "time_per_iteration": 2.7343766689300537
    },
    {
      "auxiliary_loss_clip": 0.0114238,
      "auxiliary_loss_mlp": 0.01037604,
      "balance_loss_clip": 1.05552781,
      "balance_loss_mlp": 1.02806091,
      "epoch": 0.2825707930018638,
      "flos": 24608110504320.0,
      "grad_norm": 1.5704278633076865,
      "language_loss": 0.83807403,
      "learning_rate": 3.3674806922772476e-06,
      "loss": 0.85987389,
      "num_input_tokens_seen": 50374670,
      "step": 2350,
      "time_per_iteration": 2.811335325241089
    },
    {
      "auxiliary_loss_clip": 0.01181622,
      "auxiliary_loss_mlp": 0.01039262,
      "balance_loss_clip": 1.05424881,
      "balance_loss_mlp": 1.02908134,
      "epoch": 0.28269103589250283,
      "flos": 25226994862080.0,
      "grad_norm": 1.8041093272279733,
      "language_loss": 0.75105888,
      "learning_rate": 3.3669121531826904e-06,
      "loss": 0.77326775,
      "num_input_tokens_seen": 50395650,
      "step": 2351,
      "time_per_iteration": 2.7376320362091064
    },
    {
      "auxiliary_loss_clip": 0.01140857,
      "auxiliary_loss_mlp": 0.00765344,
      "balance_loss_clip": 1.05080295,
      "balance_loss_mlp": 1.00016272,
      "epoch": 0.28281127878314194,
      "flos": 19281552819840.0,
      "grad_norm": 1.8121655280906521,
      "language_loss": 0.83219337,
      "learning_rate": 3.366343406728647e-06,
      "loss": 0.85125542,
      "num_input_tokens_seen": 50415100,
      "step": 2352,
      "time_per_iteration": 2.8072760105133057
    },
    {
      "auxiliary_loss_clip": 0.01123116,
      "auxiliary_loss_mlp": 0.01027218,
      "balance_loss_clip": 1.04480982,
      "balance_loss_mlp": 1.01701975,
      "epoch": 0.28293152167378105,
      "flos": 23878405710720.0,
      "grad_norm": 1.8113392871135554,
      "language_loss": 0.68464667,
      "learning_rate": 3.3657744530013946e-06,
      "loss": 0.70615,
      "num_input_tokens_seen": 50434335,
      "step": 2353,
      "time_per_iteration": 2.8329687118530273
    },
    {
      "auxiliary_loss_clip": 0.01164957,
      "auxiliary_loss_mlp": 0.01039331,
      "balance_loss_clip": 1.05948186,
      "balance_loss_mlp": 1.02871513,
      "epoch": 0.2830517645644201,
      "flos": 43866965928960.0,
      "grad_norm": 1.9377635470591228,
      "language_loss": 0.71363211,
      "learning_rate": 3.3652052920872437e-06,
      "loss": 0.73567498,
      "num_input_tokens_seen": 50457200,
      "step": 2354,
      "time_per_iteration": 2.931483507156372
    },
    {
      "auxiliary_loss_clip": 0.01154832,
      "auxiliary_loss_mlp": 0.01040447,
      "balance_loss_clip": 1.05314529,
      "balance_loss_mlp": 1.02980733,
      "epoch": 0.2831720074550592,
      "flos": 26651750803200.0,
      "grad_norm": 1.8749066606370421,
      "language_loss": 0.85310066,
      "learning_rate": 3.3646359240725355e-06,
      "loss": 0.87505347,
      "num_input_tokens_seen": 50476390,
      "step": 2355,
      "time_per_iteration": 2.741882562637329
    },
    {
      "auxiliary_loss_clip": 0.01192676,
      "auxiliary_loss_mlp": 0.01037242,
      "balance_loss_clip": 1.06075931,
      "balance_loss_mlp": 1.02717435,
      "epoch": 0.2832922503456983,
      "flos": 31029979564800.0,
      "grad_norm": 2.2949813445441594,
      "language_loss": 0.68142021,
      "learning_rate": 3.364066349043643e-06,
      "loss": 0.70371944,
      "num_input_tokens_seen": 50497595,
      "step": 2356,
      "time_per_iteration": 2.7827744483947754
    },
    {
      "auxiliary_loss_clip": 0.0117067,
      "auxiliary_loss_mlp": 0.01031691,
      "balance_loss_clip": 1.05631542,
      "balance_loss_mlp": 1.02243447,
      "epoch": 0.2834124932363374,
      "flos": 20405699838720.0,
      "grad_norm": 1.756677696828894,
      "language_loss": 0.82068264,
      "learning_rate": 3.363496567086969e-06,
      "loss": 0.84270626,
      "num_input_tokens_seen": 50514690,
      "step": 2357,
      "time_per_iteration": 2.6715352535247803
    },
    {
      "auxiliary_loss_clip": 0.01176076,
      "auxiliary_loss_mlp": 0.00764433,
      "balance_loss_clip": 1.05623949,
      "balance_loss_mlp": 1.00015283,
      "epoch": 0.2835327361269765,
      "flos": 39384848056320.0,
      "grad_norm": 2.5427349772945105,
      "language_loss": 0.75906056,
      "learning_rate": 3.3629265782889506e-06,
      "loss": 0.77846563,
      "num_input_tokens_seen": 50536515,
      "step": 2358,
      "time_per_iteration": 2.7665605545043945
    },
    {
      "auxiliary_loss_clip": 0.01178882,
      "auxiliary_loss_mlp": 0.01037238,
      "balance_loss_clip": 1.05600512,
      "balance_loss_mlp": 1.02727771,
      "epoch": 0.2836529790176156,
      "flos": 30261598801920.0,
      "grad_norm": 2.0022241979940265,
      "language_loss": 0.71966064,
      "learning_rate": 3.362356382736054e-06,
      "loss": 0.74182189,
      "num_input_tokens_seen": 50557120,
      "step": 2359,
      "time_per_iteration": 2.78812837600708
    },
    {
      "auxiliary_loss_clip": 0.01154376,
      "auxiliary_loss_mlp": 0.01027933,
      "balance_loss_clip": 1.04999828,
      "balance_loss_mlp": 1.01871192,
      "epoch": 0.28377322190825466,
      "flos": 12677796264960.0,
      "grad_norm": 2.0301725083794424,
      "language_loss": 0.9060055,
      "learning_rate": 3.361785980514777e-06,
      "loss": 0.92782855,
      "num_input_tokens_seen": 50573320,
      "step": 2360,
      "time_per_iteration": 2.676764726638794
    },
    {
      "auxiliary_loss_clip": 0.0116227,
      "auxiliary_loss_mlp": 0.01032325,
      "balance_loss_clip": 1.05107713,
      "balance_loss_mlp": 1.0224483,
      "epoch": 0.28389346479889377,
      "flos": 18296666830080.0,
      "grad_norm": 2.966779916111583,
      "language_loss": 0.76956415,
      "learning_rate": 3.361215371711649e-06,
      "loss": 0.79151011,
      "num_input_tokens_seen": 50592415,
      "step": 2361,
      "time_per_iteration": 2.722966432571411
    },
    {
      "auxiliary_loss_clip": 0.01146315,
      "auxiliary_loss_mlp": 0.01026042,
      "balance_loss_clip": 1.04616249,
      "balance_loss_mlp": 1.01601076,
      "epoch": 0.2840137076895329,
      "flos": 20406992728320.0,
      "grad_norm": 2.573167555019569,
      "language_loss": 0.83866251,
      "learning_rate": 3.3606445564132326e-06,
      "loss": 0.86038607,
      "num_input_tokens_seen": 50609710,
      "step": 2362,
      "time_per_iteration": 2.695765256881714
    },
    {
      "auxiliary_loss_clip": 0.01187848,
      "auxiliary_loss_mlp": 0.0103662,
      "balance_loss_clip": 1.05864775,
      "balance_loss_mlp": 1.02747703,
      "epoch": 0.28413395058017193,
      "flos": 20048030161920.0,
      "grad_norm": 7.81242377105535,
      "language_loss": 0.8207233,
      "learning_rate": 3.360073534706118e-06,
      "loss": 0.84296799,
      "num_input_tokens_seen": 50626865,
      "step": 2363,
      "time_per_iteration": 2.697080135345459
    },
    {
      "auxiliary_loss_clip": 0.01167623,
      "auxiliary_loss_mlp": 0.01038539,
      "balance_loss_clip": 1.05280602,
      "balance_loss_mlp": 1.02820349,
      "epoch": 0.28425419347081105,
      "flos": 37663613256960.0,
      "grad_norm": 1.9542311951960234,
      "language_loss": 0.75853294,
      "learning_rate": 3.35950230667693e-06,
      "loss": 0.78059453,
      "num_input_tokens_seen": 50648560,
      "step": 2364,
      "time_per_iteration": 2.809440851211548
    },
    {
      "auxiliary_loss_clip": 0.01139435,
      "auxiliary_loss_mlp": 0.01032092,
      "balance_loss_clip": 1.04588544,
      "balance_loss_mlp": 1.02250171,
      "epoch": 0.28437443636145016,
      "flos": 13845072539520.0,
      "grad_norm": 2.112532696139439,
      "language_loss": 0.85905057,
      "learning_rate": 3.358930872412323e-06,
      "loss": 0.8807658,
      "num_input_tokens_seen": 50665725,
      "step": 2365,
      "time_per_iteration": 2.769566297531128
    },
    {
      "auxiliary_loss_clip": 0.0116928,
      "auxiliary_loss_mlp": 0.01032192,
      "balance_loss_clip": 1.05175304,
      "balance_loss_mlp": 1.02216053,
      "epoch": 0.2844946792520892,
      "flos": 22747794243840.0,
      "grad_norm": 1.7685979223493071,
      "language_loss": 0.81003684,
      "learning_rate": 3.3583592319989825e-06,
      "loss": 0.83205158,
      "num_input_tokens_seen": 50685095,
      "step": 2366,
      "time_per_iteration": 3.7350308895111084
    },
    {
      "auxiliary_loss_clip": 0.01142262,
      "auxiliary_loss_mlp": 0.01041849,
      "balance_loss_clip": 1.05194759,
      "balance_loss_mlp": 1.03128111,
      "epoch": 0.2846149221427283,
      "flos": 32415987709440.0,
      "grad_norm": 2.1629808959246204,
      "language_loss": 0.69000387,
      "learning_rate": 3.357787385523627e-06,
      "loss": 0.71184492,
      "num_input_tokens_seen": 50706500,
      "step": 2367,
      "time_per_iteration": 2.8688607215881348
    },
    {
      "auxiliary_loss_clip": 0.01140493,
      "auxiliary_loss_mlp": 0.01035082,
      "balance_loss_clip": 1.04813862,
      "balance_loss_mlp": 1.02515197,
      "epoch": 0.2847351650333674,
      "flos": 28475976873600.0,
      "grad_norm": 1.8466293337963,
      "language_loss": 0.82746005,
      "learning_rate": 3.3572153330730048e-06,
      "loss": 0.84921581,
      "num_input_tokens_seen": 50727595,
      "step": 2368,
      "time_per_iteration": 3.7562708854675293
    },
    {
      "auxiliary_loss_clip": 0.01072982,
      "auxiliary_loss_mlp": 0.01004052,
      "balance_loss_clip": 1.02590156,
      "balance_loss_mlp": 1.00233543,
      "epoch": 0.2848554079240065,
      "flos": 55753399704960.0,
      "grad_norm": 0.8298930241794007,
      "language_loss": 0.64690822,
      "learning_rate": 3.3566430747338956e-06,
      "loss": 0.66767859,
      "num_input_tokens_seen": 50782800,
      "step": 2369,
      "time_per_iteration": 3.120100975036621
    },
    {
      "auxiliary_loss_clip": 0.01173304,
      "auxiliary_loss_mlp": 0.0102891,
      "balance_loss_clip": 1.05408835,
      "balance_loss_mlp": 1.0194751,
      "epoch": 0.2849756508146456,
      "flos": 11836875985920.0,
      "grad_norm": 2.1240195369227552,
      "language_loss": 0.86685294,
      "learning_rate": 3.35607061059311e-06,
      "loss": 0.88887513,
      "num_input_tokens_seen": 50797730,
      "step": 2370,
      "time_per_iteration": 2.6581499576568604
    },
    {
      "auxiliary_loss_clip": 0.01148064,
      "auxiliary_loss_mlp": 0.01031845,
      "balance_loss_clip": 1.05373359,
      "balance_loss_mlp": 1.0216763,
      "epoch": 0.28509589370528465,
      "flos": 25155209531520.0,
      "grad_norm": 1.9867925861856683,
      "language_loss": 0.75184119,
      "learning_rate": 3.3554979407374917e-06,
      "loss": 0.77364028,
      "num_input_tokens_seen": 50819840,
      "step": 2371,
      "time_per_iteration": 2.82029128074646
    },
    {
      "auxiliary_loss_clip": 0.0119976,
      "auxiliary_loss_mlp": 0.01034045,
      "balance_loss_clip": 1.05791545,
      "balance_loss_mlp": 1.02488375,
      "epoch": 0.28521613659592376,
      "flos": 19974808287360.0,
      "grad_norm": 1.5209933090501866,
      "language_loss": 0.73599195,
      "learning_rate": 3.3549250652539134e-06,
      "loss": 0.75832999,
      "num_input_tokens_seen": 50838935,
      "step": 2372,
      "time_per_iteration": 2.628692150115967
    },
    {
      "auxiliary_loss_clip": 0.0115281,
      "auxiliary_loss_mlp": 0.01038016,
      "balance_loss_clip": 1.05026436,
      "balance_loss_mlp": 1.02760947,
      "epoch": 0.2853363794865629,
      "flos": 23367971491200.0,
      "grad_norm": 1.8862295722283944,
      "language_loss": 0.81140554,
      "learning_rate": 3.3543519842292794e-06,
      "loss": 0.83331382,
      "num_input_tokens_seen": 50858590,
      "step": 2373,
      "time_per_iteration": 2.778146743774414
    },
    {
      "auxiliary_loss_clip": 0.01130477,
      "auxiliary_loss_mlp": 0.01034558,
      "balance_loss_clip": 1.04945278,
      "balance_loss_mlp": 1.02447331,
      "epoch": 0.28545662237720193,
      "flos": 19861940776320.0,
      "grad_norm": 1.9312605316846316,
      "language_loss": 0.8337869,
      "learning_rate": 3.353778697750527e-06,
      "loss": 0.85543728,
      "num_input_tokens_seen": 50876995,
      "step": 2374,
      "time_per_iteration": 2.7588043212890625
    },
    {
      "auxiliary_loss_clip": 0.01142998,
      "auxiliary_loss_mlp": 0.00764095,
      "balance_loss_clip": 1.04649329,
      "balance_loss_mlp": 1.00017214,
      "epoch": 0.28557686526784104,
      "flos": 23879016241920.0,
      "grad_norm": 1.6487792750693435,
      "language_loss": 0.89123297,
      "learning_rate": 3.353205205904622e-06,
      "loss": 0.91030389,
      "num_input_tokens_seen": 50896105,
      "step": 2375,
      "time_per_iteration": 2.7772200107574463
    },
    {
      "auxiliary_loss_clip": 0.0115759,
      "auxiliary_loss_mlp": 0.01032085,
      "balance_loss_clip": 1.05142009,
      "balance_loss_mlp": 1.02201819,
      "epoch": 0.28569710815848015,
      "flos": 44890384233600.0,
      "grad_norm": 1.8384536582946422,
      "language_loss": 0.71794355,
      "learning_rate": 3.3526315087785637e-06,
      "loss": 0.73984027,
      "num_input_tokens_seen": 50917220,
      "step": 2376,
      "time_per_iteration": 2.9539899826049805
    },
    {
      "auxiliary_loss_clip": 0.01168541,
      "auxiliary_loss_mlp": 0.01033762,
      "balance_loss_clip": 1.05441594,
      "balance_loss_mlp": 1.02446342,
      "epoch": 0.2858173510491192,
      "flos": 26829759628800.0,
      "grad_norm": 2.609242514664903,
      "language_loss": 0.81236351,
      "learning_rate": 3.3520576064593805e-06,
      "loss": 0.83438653,
      "num_input_tokens_seen": 50937175,
      "step": 2377,
      "time_per_iteration": 2.6888458728790283
    },
    {
      "auxiliary_loss_clip": 0.0117942,
      "auxiliary_loss_mlp": 0.01034744,
      "balance_loss_clip": 1.05946136,
      "balance_loss_mlp": 1.02512383,
      "epoch": 0.2859375939397583,
      "flos": 23148916398720.0,
      "grad_norm": 1.502599810174506,
      "language_loss": 0.81684411,
      "learning_rate": 3.3514834990341337e-06,
      "loss": 0.83898568,
      "num_input_tokens_seen": 50957500,
      "step": 2378,
      "time_per_iteration": 2.7454981803894043
    },
    {
      "auxiliary_loss_clip": 0.01190929,
      "auxiliary_loss_mlp": 0.01032416,
      "balance_loss_clip": 1.0575397,
      "balance_loss_mlp": 1.02339196,
      "epoch": 0.2860578368303974,
      "flos": 12129799397760.0,
      "grad_norm": 2.6909478359149563,
      "language_loss": 0.92934728,
      "learning_rate": 3.3509091865899144e-06,
      "loss": 0.9515807,
      "num_input_tokens_seen": 50972690,
      "step": 2379,
      "time_per_iteration": 2.679994583129883
    },
    {
      "auxiliary_loss_clip": 0.0113249,
      "auxiliary_loss_mlp": 0.01033775,
      "balance_loss_clip": 1.04857218,
      "balance_loss_mlp": 1.02443552,
      "epoch": 0.2861780797210365,
      "flos": 19938035738880.0,
      "grad_norm": 2.2504198863613145,
      "language_loss": 0.7072649,
      "learning_rate": 3.350334669213846e-06,
      "loss": 0.72892761,
      "num_input_tokens_seen": 50990095,
      "step": 2380,
      "time_per_iteration": 2.8304953575134277
    },
    {
      "auxiliary_loss_clip": 0.01189594,
      "auxiliary_loss_mlp": 0.0103275,
      "balance_loss_clip": 1.05745125,
      "balance_loss_mlp": 1.02311242,
      "epoch": 0.2862983226116756,
      "flos": 27563127609600.0,
      "grad_norm": 1.9725844081141637,
      "language_loss": 0.75536668,
      "learning_rate": 3.3497599469930816e-06,
      "loss": 0.77759016,
      "num_input_tokens_seen": 51008305,
      "step": 2381,
      "time_per_iteration": 2.6869678497314453
    },
    {
      "auxiliary_loss_clip": 0.01162391,
      "auxiliary_loss_mlp": 0.01042424,
      "balance_loss_clip": 1.05477333,
      "balance_loss_mlp": 1.03236866,
      "epoch": 0.28641856550231465,
      "flos": 22053964158720.0,
      "grad_norm": 2.4147856265731873,
      "language_loss": 0.83341533,
      "learning_rate": 3.349185020014807e-06,
      "loss": 0.8554635,
      "num_input_tokens_seen": 51025570,
      "step": 2382,
      "time_per_iteration": 2.689668655395508
    },
    {
      "auxiliary_loss_clip": 0.01126549,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.04876065,
      "balance_loss_mlp": 1.01988721,
      "epoch": 0.28653880839295376,
      "flos": 22378775869440.0,
      "grad_norm": 2.2392582276753803,
      "language_loss": 0.74558938,
      "learning_rate": 3.348609888366237e-06,
      "loss": 0.76714921,
      "num_input_tokens_seen": 51044585,
      "step": 2383,
      "time_per_iteration": 2.7780399322509766
    },
    {
      "auxiliary_loss_clip": 0.01184944,
      "auxiliary_loss_mlp": 0.01028847,
      "balance_loss_clip": 1.05616522,
      "balance_loss_mlp": 1.0199182,
      "epoch": 0.28665905128359287,
      "flos": 23367971491200.0,
      "grad_norm": 3.367829754600112,
      "language_loss": 0.63142604,
      "learning_rate": 3.348034552134619e-06,
      "loss": 0.65356398,
      "num_input_tokens_seen": 51063990,
      "step": 2384,
      "time_per_iteration": 2.6821606159210205
    },
    {
      "auxiliary_loss_clip": 0.01187076,
      "auxiliary_loss_mlp": 0.00763347,
      "balance_loss_clip": 1.05638599,
      "balance_loss_mlp": 1.00016618,
      "epoch": 0.2867792941742319,
      "flos": 20881695893760.0,
      "grad_norm": 1.8803065611079648,
      "language_loss": 0.84228176,
      "learning_rate": 3.3474590114072316e-06,
      "loss": 0.86178601,
      "num_input_tokens_seen": 51081990,
      "step": 2385,
      "time_per_iteration": 2.64638090133667
    },
    {
      "auxiliary_loss_clip": 0.01115807,
      "auxiliary_loss_mlp": 0.01027943,
      "balance_loss_clip": 1.04535341,
      "balance_loss_mlp": 1.01870489,
      "epoch": 0.28689953706487104,
      "flos": 20664005518080.0,
      "grad_norm": 2.3031421478319207,
      "language_loss": 0.83334154,
      "learning_rate": 3.3468832662713836e-06,
      "loss": 0.85477906,
      "num_input_tokens_seen": 51100235,
      "step": 2386,
      "time_per_iteration": 3.008270502090454
    },
    {
      "auxiliary_loss_clip": 0.01162877,
      "auxiliary_loss_mlp": 0.01032577,
      "balance_loss_clip": 1.05258322,
      "balance_loss_mlp": 1.02327251,
      "epoch": 0.28701977995551015,
      "flos": 12675533708160.0,
      "grad_norm": 2.898498549333057,
      "language_loss": 0.84039688,
      "learning_rate": 3.346307316814415e-06,
      "loss": 0.86235142,
      "num_input_tokens_seen": 51115405,
      "step": 2387,
      "time_per_iteration": 2.9773781299591064
    },
    {
      "auxiliary_loss_clip": 0.01123688,
      "auxiliary_loss_mlp": 0.01030569,
      "balance_loss_clip": 1.05280054,
      "balance_loss_mlp": 1.02007222,
      "epoch": 0.2871400228461492,
      "flos": 21252366293760.0,
      "grad_norm": 2.208603334320773,
      "language_loss": 0.75864524,
      "learning_rate": 3.3457311631236965e-06,
      "loss": 0.78018785,
      "num_input_tokens_seen": 51136390,
      "step": 2388,
      "time_per_iteration": 2.841770887374878
    },
    {
      "auxiliary_loss_clip": 0.01147169,
      "auxiliary_loss_mlp": 0.0103286,
      "balance_loss_clip": 1.05099416,
      "balance_loss_mlp": 1.02322745,
      "epoch": 0.2872602657367883,
      "flos": 25119262995840.0,
      "grad_norm": 2.41572534585922,
      "language_loss": 0.84786212,
      "learning_rate": 3.345154805286631e-06,
      "loss": 0.8696624,
      "num_input_tokens_seen": 51156650,
      "step": 2389,
      "time_per_iteration": 2.77532958984375
    },
    {
      "auxiliary_loss_clip": 0.01126902,
      "auxiliary_loss_mlp": 0.01030817,
      "balance_loss_clip": 1.05239129,
      "balance_loss_mlp": 1.02120829,
      "epoch": 0.2873805086274274,
      "flos": 16646606830080.0,
      "grad_norm": 2.604162470238548,
      "language_loss": 0.76261348,
      "learning_rate": 3.344578243390651e-06,
      "loss": 0.78419065,
      "num_input_tokens_seen": 51172210,
      "step": 2390,
      "time_per_iteration": 2.7134954929351807
    },
    {
      "auxiliary_loss_clip": 0.01157299,
      "auxiliary_loss_mlp": 0.0076319,
      "balance_loss_clip": 1.05278158,
      "balance_loss_mlp": 1.00011468,
      "epoch": 0.2875007515180665,
      "flos": 17420123237760.0,
      "grad_norm": 2.9230510685313393,
      "language_loss": 0.78554583,
      "learning_rate": 3.3440014775232206e-06,
      "loss": 0.80475068,
      "num_input_tokens_seen": 51190265,
      "step": 2391,
      "time_per_iteration": 3.7149064540863037
    },
    {
      "auxiliary_loss_clip": 0.01157473,
      "auxiliary_loss_mlp": 0.01032855,
      "balance_loss_clip": 1.05467606,
      "balance_loss_mlp": 1.02424216,
      "epoch": 0.2876209944087056,
      "flos": 23434190213760.0,
      "grad_norm": 2.3017645225848664,
      "language_loss": 0.71169311,
      "learning_rate": 3.343424507771834e-06,
      "loss": 0.73359644,
      "num_input_tokens_seen": 51208475,
      "step": 2392,
      "time_per_iteration": 2.7489380836486816
    },
    {
      "auxiliary_loss_clip": 0.01197467,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.05576205,
      "balance_loss_mlp": 1.01902962,
      "epoch": 0.2877412372993447,
      "flos": 13735509079680.0,
      "grad_norm": 2.663738508376235,
      "language_loss": 0.86368656,
      "learning_rate": 3.342847334224018e-06,
      "loss": 0.88593763,
      "num_input_tokens_seen": 51225875,
      "step": 2393,
      "time_per_iteration": 2.617743730545044
    },
    {
      "auxiliary_loss_clip": 0.01068493,
      "auxiliary_loss_mlp": 0.01002512,
      "balance_loss_clip": 1.01685905,
      "balance_loss_mlp": 1.00101554,
      "epoch": 0.28786148018998375,
      "flos": 58079695104000.0,
      "grad_norm": 0.9372552769478648,
      "language_loss": 0.62402326,
      "learning_rate": 3.342269956967329e-06,
      "loss": 0.64473331,
      "num_input_tokens_seen": 51287780,
      "step": 2394,
      "time_per_iteration": 4.273821592330933
    },
    {
      "auxiliary_loss_clip": 0.01193412,
      "auxiliary_loss_mlp": 0.00764501,
      "balance_loss_clip": 1.05622125,
      "balance_loss_mlp": 1.00017476,
      "epoch": 0.28798172308062286,
      "flos": 23435052140160.0,
      "grad_norm": 2.60127768048474,
      "language_loss": 0.72058988,
      "learning_rate": 3.341692376089355e-06,
      "loss": 0.74016905,
      "num_input_tokens_seen": 51303335,
      "step": 2395,
      "time_per_iteration": 3.6757848262786865
    },
    {
      "auxiliary_loss_clip": 0.01128706,
      "auxiliary_loss_mlp": 0.01032753,
      "balance_loss_clip": 1.0459969,
      "balance_loss_mlp": 1.02344847,
      "epoch": 0.288101965971262,
      "flos": 25110033200640.0,
      "grad_norm": 2.8290840455213964,
      "language_loss": 0.83412111,
      "learning_rate": 3.3411145916777146e-06,
      "loss": 0.85573572,
      "num_input_tokens_seen": 51317495,
      "step": 2396,
      "time_per_iteration": 2.7852320671081543
    },
    {
      "auxiliary_loss_clip": 0.011501,
      "auxiliary_loss_mlp": 0.01032173,
      "balance_loss_clip": 1.05176568,
      "balance_loss_mlp": 1.02218354,
      "epoch": 0.28822220886190103,
      "flos": 16252559654400.0,
      "grad_norm": 2.20614171412712,
      "language_loss": 0.90998793,
      "learning_rate": 3.3405366038200566e-06,
      "loss": 0.93181074,
      "num_input_tokens_seen": 51336430,
      "step": 2397,
      "time_per_iteration": 2.7425012588500977
    },
    {
      "auxiliary_loss_clip": 0.0114935,
      "auxiliary_loss_mlp": 0.01036045,
      "balance_loss_clip": 1.05545878,
      "balance_loss_mlp": 1.02642488,
      "epoch": 0.28834245175254014,
      "flos": 24535642815360.0,
      "grad_norm": 2.19982651652735,
      "language_loss": 0.84895587,
      "learning_rate": 3.3399584126040617e-06,
      "loss": 0.87080979,
      "num_input_tokens_seen": 51355930,
      "step": 2398,
      "time_per_iteration": 2.8133132457733154
    },
    {
      "auxiliary_loss_clip": 0.01124679,
      "auxiliary_loss_mlp": 0.00763486,
      "balance_loss_clip": 1.04791427,
      "balance_loss_mlp": 1.00015569,
      "epoch": 0.2884626946431792,
      "flos": 24571445696640.0,
      "grad_norm": 1.834842556634728,
      "language_loss": 0.90831828,
      "learning_rate": 3.339380018117441e-06,
      "loss": 0.92719984,
      "num_input_tokens_seen": 51376765,
      "step": 2399,
      "time_per_iteration": 2.878488779067993
    },
    {
      "auxiliary_loss_clip": 0.01119617,
      "auxiliary_loss_mlp": 0.01030198,
      "balance_loss_clip": 1.04717028,
      "balance_loss_mlp": 1.02053618,
      "epoch": 0.2885829375338183,
      "flos": 16544657053440.0,
      "grad_norm": 2.3654737740190823,
      "language_loss": 0.78322971,
      "learning_rate": 3.3388014204479366e-06,
      "loss": 0.80472785,
      "num_input_tokens_seen": 51394570,
      "step": 2400,
      "time_per_iteration": 2.7507948875427246
    },
    {
      "auxiliary_loss_clip": 0.01080627,
      "auxiliary_loss_mlp": 0.01031466,
      "balance_loss_clip": 1.0447855,
      "balance_loss_mlp": 1.02098107,
      "epoch": 0.2887031804244574,
      "flos": 24061226958720.0,
      "grad_norm": 2.203646961510135,
      "language_loss": 0.91467053,
      "learning_rate": 3.338222619683321e-06,
      "loss": 0.93579149,
      "num_input_tokens_seen": 51414535,
      "step": 2401,
      "time_per_iteration": 3.0968263149261475
    },
    {
      "auxiliary_loss_clip": 0.0113557,
      "auxiliary_loss_mlp": 0.01031449,
      "balance_loss_clip": 1.05013168,
      "balance_loss_mlp": 1.0219245,
      "epoch": 0.2888234233150965,
      "flos": 23330696152320.0,
      "grad_norm": 4.014253552515274,
      "language_loss": 0.73260236,
      "learning_rate": 3.337643615911398e-06,
      "loss": 0.75427258,
      "num_input_tokens_seen": 51434160,
      "step": 2402,
      "time_per_iteration": 3.201019287109375
    },
    {
      "auxiliary_loss_clip": 0.01187599,
      "auxiliary_loss_mlp": 0.01033837,
      "balance_loss_clip": 1.05305314,
      "balance_loss_mlp": 1.02394843,
      "epoch": 0.2889436662057356,
      "flos": 22272767856000.0,
      "grad_norm": 2.1159868490902256,
      "language_loss": 0.78651792,
      "learning_rate": 3.3370644092200026e-06,
      "loss": 0.80873227,
      "num_input_tokens_seen": 51451435,
      "step": 2403,
      "time_per_iteration": 2.6836469173431396
    },
    {
      "auxiliary_loss_clip": 0.01184416,
      "auxiliary_loss_mlp": 0.00763563,
      "balance_loss_clip": 1.0532254,
      "balance_loss_mlp": 1.00015426,
      "epoch": 0.2890639090963747,
      "flos": 21616931381760.0,
      "grad_norm": 1.8390417615525756,
      "language_loss": 0.78534585,
      "learning_rate": 3.3364849996969985e-06,
      "loss": 0.8048256,
      "num_input_tokens_seen": 51471455,
      "step": 2404,
      "time_per_iteration": 2.7250161170959473
    },
    {
      "auxiliary_loss_clip": 0.01135944,
      "auxiliary_loss_mlp": 0.00762836,
      "balance_loss_clip": 1.05039799,
      "balance_loss_mlp": 1.00014687,
      "epoch": 0.28918415198701375,
      "flos": 28585540333440.0,
      "grad_norm": 1.989824813760848,
      "language_loss": 0.85727847,
      "learning_rate": 3.335905387430283e-06,
      "loss": 0.8762663,
      "num_input_tokens_seen": 51492890,
      "step": 2405,
      "time_per_iteration": 2.8180606365203857
    },
    {
      "auxiliary_loss_clip": 0.01142516,
      "auxiliary_loss_mlp": 0.010374,
      "balance_loss_clip": 1.04747128,
      "balance_loss_mlp": 1.02782202,
      "epoch": 0.28930439487765286,
      "flos": 21944688007680.0,
      "grad_norm": 1.8712796118074624,
      "language_loss": 0.82855296,
      "learning_rate": 3.335325572507782e-06,
      "loss": 0.85035211,
      "num_input_tokens_seen": 51513390,
      "step": 2406,
      "time_per_iteration": 2.8203322887420654
    },
    {
      "auxiliary_loss_clip": 0.01153235,
      "auxiliary_loss_mlp": 0.00764063,
      "balance_loss_clip": 1.05250525,
      "balance_loss_mlp": 1.00018525,
      "epoch": 0.28942463776829197,
      "flos": 19281911955840.0,
      "grad_norm": 1.6755376128369985,
      "language_loss": 0.73793238,
      "learning_rate": 3.3347455550174537e-06,
      "loss": 0.75710535,
      "num_input_tokens_seen": 51532730,
      "step": 2407,
      "time_per_iteration": 2.738115072250366
    },
    {
      "auxiliary_loss_clip": 0.01154447,
      "auxiliary_loss_mlp": 0.00763655,
      "balance_loss_clip": 1.0474968,
      "balance_loss_mlp": 1.00021195,
      "epoch": 0.289544880658931,
      "flos": 14645700737280.0,
      "grad_norm": 1.9872905394815272,
      "language_loss": 0.68233848,
      "learning_rate": 3.3341653350472864e-06,
      "loss": 0.70151949,
      "num_input_tokens_seen": 51549560,
      "step": 2408,
      "time_per_iteration": 3.245424747467041
    },
    {
      "auxiliary_loss_clip": 0.0113482,
      "auxiliary_loss_mlp": 0.0104016,
      "balance_loss_clip": 1.04819059,
      "balance_loss_mlp": 1.02842987,
      "epoch": 0.28966512354957014,
      "flos": 28621881918720.0,
      "grad_norm": 2.502363207254137,
      "language_loss": 0.68694675,
      "learning_rate": 3.333584912685298e-06,
      "loss": 0.70869654,
      "num_input_tokens_seen": 51568180,
      "step": 2409,
      "time_per_iteration": 2.8783271312713623
    },
    {
      "auxiliary_loss_clip": 0.01063651,
      "auxiliary_loss_mlp": 0.01002683,
      "balance_loss_clip": 1.0192287,
      "balance_loss_mlp": 1.00107408,
      "epoch": 0.28978536644020925,
      "flos": 64711784511360.0,
      "grad_norm": 0.8727984716614485,
      "language_loss": 0.55547148,
      "learning_rate": 3.3330042880195385e-06,
      "loss": 0.5761348,
      "num_input_tokens_seen": 51622530,
      "step": 2410,
      "time_per_iteration": 3.2267343997955322
    },
    {
      "auxiliary_loss_clip": 0.01178687,
      "auxiliary_loss_mlp": 0.01039582,
      "balance_loss_clip": 1.05204594,
      "balance_loss_mlp": 1.03059363,
      "epoch": 0.2899056093308483,
      "flos": 18624638937600.0,
      "grad_norm": 2.0188389044883666,
      "language_loss": 0.78720772,
      "learning_rate": 3.3324234611380888e-06,
      "loss": 0.80939043,
      "num_input_tokens_seen": 51641260,
      "step": 2411,
      "time_per_iteration": 2.704883098602295
    },
    {
      "auxiliary_loss_clip": 0.0117543,
      "auxiliary_loss_mlp": 0.0103525,
      "balance_loss_clip": 1.0565418,
      "balance_loss_mlp": 1.02612996,
      "epoch": 0.2900258522214874,
      "flos": 22893735202560.0,
      "grad_norm": 1.6311942089560714,
      "language_loss": 0.81866723,
      "learning_rate": 3.3318424321290596e-06,
      "loss": 0.84077406,
      "num_input_tokens_seen": 51660975,
      "step": 2412,
      "time_per_iteration": 2.687671422958374
    },
    {
      "auxiliary_loss_clip": 0.01031395,
      "auxiliary_loss_mlp": 0.01002347,
      "balance_loss_clip": 1.01805997,
      "balance_loss_mlp": 1.00078583,
      "epoch": 0.2901460951121265,
      "flos": 71106036013440.0,
      "grad_norm": 0.8255732015927931,
      "language_loss": 0.5988723,
      "learning_rate": 3.3312612010805917e-06,
      "loss": 0.61920965,
      "num_input_tokens_seen": 51720550,
      "step": 2413,
      "time_per_iteration": 3.4102606773376465
    },
    {
      "auxiliary_loss_clip": 0.01136349,
      "auxiliary_loss_mlp": 0.01034587,
      "balance_loss_clip": 1.05020607,
      "balance_loss_mlp": 1.02494931,
      "epoch": 0.2902663380027656,
      "flos": 32160986081280.0,
      "grad_norm": 2.508971576441302,
      "language_loss": 0.70142215,
      "learning_rate": 3.330679768080858e-06,
      "loss": 0.72313154,
      "num_input_tokens_seen": 51744435,
      "step": 2414,
      "time_per_iteration": 2.882861375808716
    },
    {
      "auxiliary_loss_clip": 0.01111808,
      "auxiliary_loss_mlp": 0.01028652,
      "balance_loss_clip": 1.04847383,
      "balance_loss_mlp": 1.01848304,
      "epoch": 0.2903865808934047,
      "flos": 29351658539520.0,
      "grad_norm": 2.318257000740699,
      "language_loss": 0.83523345,
      "learning_rate": 3.3300981332180627e-06,
      "loss": 0.85663807,
      "num_input_tokens_seen": 51763640,
      "step": 2415,
      "time_per_iteration": 3.0018484592437744
    },
    {
      "auxiliary_loss_clip": 0.01203326,
      "auxiliary_loss_mlp": 0.01032451,
      "balance_loss_clip": 1.05903697,
      "balance_loss_mlp": 1.02298617,
      "epoch": 0.29050682378404374,
      "flos": 17089026647040.0,
      "grad_norm": 2.074285120001107,
      "language_loss": 0.80148375,
      "learning_rate": 3.3295162965804373e-06,
      "loss": 0.82384157,
      "num_input_tokens_seen": 51782135,
      "step": 2416,
      "time_per_iteration": 3.117766857147217
    },
    {
      "auxiliary_loss_clip": 0.01187373,
      "auxiliary_loss_mlp": 0.01031129,
      "balance_loss_clip": 1.05909586,
      "balance_loss_mlp": 1.02220058,
      "epoch": 0.29062706667468285,
      "flos": 17858233422720.0,
      "grad_norm": 2.4824334025364996,
      "language_loss": 0.78956807,
      "learning_rate": 3.328934258256247e-06,
      "loss": 0.81175303,
      "num_input_tokens_seen": 51800200,
      "step": 2417,
      "time_per_iteration": 4.782647609710693
    },
    {
      "auxiliary_loss_clip": 0.01118863,
      "auxiliary_loss_mlp": 0.00764324,
      "balance_loss_clip": 1.04246771,
      "balance_loss_mlp": 1.0002085,
      "epoch": 0.29074730956532197,
      "flos": 24279815174400.0,
      "grad_norm": 1.8570517564930173,
      "language_loss": 0.67236108,
      "learning_rate": 3.3283520183337856e-06,
      "loss": 0.69119298,
      "num_input_tokens_seen": 51819905,
      "step": 2418,
      "time_per_iteration": 2.7949962615966797
    },
    {
      "auxiliary_loss_clip": 0.0115559,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.05232608,
      "balance_loss_mlp": 1.01924419,
      "epoch": 0.290867552455961,
      "flos": 22340961826560.0,
      "grad_norm": 2.5076993070075866,
      "language_loss": 0.69210863,
      "learning_rate": 3.3277695769013797e-06,
      "loss": 0.713943,
      "num_input_tokens_seen": 51839350,
      "step": 2419,
      "time_per_iteration": 2.7775728702545166
    },
    {
      "auxiliary_loss_clip": 0.01137005,
      "auxiliary_loss_mlp": 0.01033786,
      "balance_loss_clip": 1.05042529,
      "balance_loss_mlp": 1.02436221,
      "epoch": 0.29098779534660013,
      "flos": 23186155824000.0,
      "grad_norm": 2.7585335333081145,
      "language_loss": 0.77573001,
      "learning_rate": 3.327186934047385e-06,
      "loss": 0.79743791,
      "num_input_tokens_seen": 51858045,
      "step": 2420,
      "time_per_iteration": 3.68530535697937
    },
    {
      "auxiliary_loss_clip": 0.01092814,
      "auxiliary_loss_mlp": 0.0102964,
      "balance_loss_clip": 1.04200923,
      "balance_loss_mlp": 1.02016926,
      "epoch": 0.29110803823723924,
      "flos": 15304194817920.0,
      "grad_norm": 1.8922424158791036,
      "language_loss": 0.65992332,
      "learning_rate": 3.3266040898601877e-06,
      "loss": 0.68114781,
      "num_input_tokens_seen": 51875880,
      "step": 2421,
      "time_per_iteration": 3.846189498901367
    },
    {
      "auxiliary_loss_clip": 0.01191609,
      "auxiliary_loss_mlp": 0.01035889,
      "balance_loss_clip": 1.06237876,
      "balance_loss_mlp": 1.02642965,
      "epoch": 0.2912282811278783,
      "flos": 22595352923520.0,
      "grad_norm": 1.9298941299794388,
      "language_loss": 0.78079247,
      "learning_rate": 3.3260210444282045e-06,
      "loss": 0.80306745,
      "num_input_tokens_seen": 51893835,
      "step": 2422,
      "time_per_iteration": 3.010197639465332
    },
    {
      "auxiliary_loss_clip": 0.0112342,
      "auxiliary_loss_mlp": 0.01030213,
      "balance_loss_clip": 1.04933119,
      "balance_loss_mlp": 1.02117109,
      "epoch": 0.2913485240185174,
      "flos": 24497900599680.0,
      "grad_norm": 2.4685035567037477,
      "language_loss": 0.73385394,
      "learning_rate": 3.325437797839883e-06,
      "loss": 0.75539029,
      "num_input_tokens_seen": 51912205,
      "step": 2423,
      "time_per_iteration": 2.8468544483184814
    },
    {
      "auxiliary_loss_clip": 0.01098234,
      "auxiliary_loss_mlp": 0.01028358,
      "balance_loss_clip": 1.04451752,
      "balance_loss_mlp": 1.01804018,
      "epoch": 0.2914687669091565,
      "flos": 17931024334080.0,
      "grad_norm": 3.100611511017553,
      "language_loss": 0.75483531,
      "learning_rate": 3.3248543501837015e-06,
      "loss": 0.77610123,
      "num_input_tokens_seen": 51929410,
      "step": 2424,
      "time_per_iteration": 2.890800714492798
    },
    {
      "auxiliary_loss_clip": 0.0114327,
      "auxiliary_loss_mlp": 0.01032631,
      "balance_loss_clip": 1.05106282,
      "balance_loss_mlp": 1.02272475,
      "epoch": 0.2915890097997956,
      "flos": 22529313768960.0,
      "grad_norm": 3.132792926175196,
      "language_loss": 0.77381873,
      "learning_rate": 3.3242707015481684e-06,
      "loss": 0.79557776,
      "num_input_tokens_seen": 51949345,
      "step": 2425,
      "time_per_iteration": 3.1095986366271973
    },
    {
      "auxiliary_loss_clip": 0.01146353,
      "auxiliary_loss_mlp": 0.01032399,
      "balance_loss_clip": 1.04990518,
      "balance_loss_mlp": 1.02314281,
      "epoch": 0.2917092526904347,
      "flos": 13845216193920.0,
      "grad_norm": 1.8419680125537834,
      "language_loss": 0.80451596,
      "learning_rate": 3.323686852021823e-06,
      "loss": 0.82630348,
      "num_input_tokens_seen": 51966855,
      "step": 2426,
      "time_per_iteration": 2.7590479850769043
    },
    {
      "auxiliary_loss_clip": 0.01154418,
      "auxiliary_loss_mlp": 0.01034779,
      "balance_loss_clip": 1.05307555,
      "balance_loss_mlp": 1.02506995,
      "epoch": 0.2918294955810738,
      "flos": 22674859678080.0,
      "grad_norm": 2.476245872619211,
      "language_loss": 0.79606247,
      "learning_rate": 3.323102801693235e-06,
      "loss": 0.81795442,
      "num_input_tokens_seen": 51985620,
      "step": 2427,
      "time_per_iteration": 2.783510446548462
    },
    {
      "auxiliary_loss_clip": 0.01160838,
      "auxiliary_loss_mlp": 0.01027264,
      "balance_loss_clip": 1.05469692,
      "balance_loss_mlp": 1.01739955,
      "epoch": 0.29194973847171285,
      "flos": 23438284364160.0,
      "grad_norm": 2.0852128140731194,
      "language_loss": 0.80781794,
      "learning_rate": 3.322518550651003e-06,
      "loss": 0.82969892,
      "num_input_tokens_seen": 52004930,
      "step": 2428,
      "time_per_iteration": 2.8347764015197754
    },
    {
      "auxiliary_loss_clip": 0.01129918,
      "auxiliary_loss_mlp": 0.01031205,
      "balance_loss_clip": 1.04914594,
      "balance_loss_mlp": 1.02209187,
      "epoch": 0.29206998136235196,
      "flos": 21909064694400.0,
      "grad_norm": 2.004778920184166,
      "language_loss": 0.80938387,
      "learning_rate": 3.3219340989837586e-06,
      "loss": 0.83099502,
      "num_input_tokens_seen": 52024920,
      "step": 2429,
      "time_per_iteration": 2.767841100692749
    },
    {
      "auxiliary_loss_clip": 0.01185658,
      "auxiliary_loss_mlp": 0.01027582,
      "balance_loss_clip": 1.05968261,
      "balance_loss_mlp": 1.01876068,
      "epoch": 0.292190224252991,
      "flos": 23215925220480.0,
      "grad_norm": 2.030318682920934,
      "language_loss": 0.80768228,
      "learning_rate": 3.3213494467801625e-06,
      "loss": 0.82981467,
      "num_input_tokens_seen": 52044095,
      "step": 2430,
      "time_per_iteration": 2.714914560317993
    },
    {
      "auxiliary_loss_clip": 0.01169938,
      "auxiliary_loss_mlp": 0.01031219,
      "balance_loss_clip": 1.05821025,
      "balance_loss_mlp": 1.02079415,
      "epoch": 0.2923104671436301,
      "flos": 20740818752640.0,
      "grad_norm": 2.130336132383923,
      "language_loss": 0.71490979,
      "learning_rate": 3.3207645941289063e-06,
      "loss": 0.73692137,
      "num_input_tokens_seen": 52062440,
      "step": 2431,
      "time_per_iteration": 2.719669818878174
    },
    {
      "auxiliary_loss_clip": 0.01104509,
      "auxiliary_loss_mlp": 0.01030384,
      "balance_loss_clip": 1.0450213,
      "balance_loss_mlp": 1.02105641,
      "epoch": 0.29243071003426924,
      "flos": 35809114999680.0,
      "grad_norm": 2.421794543441703,
      "language_loss": 0.80308807,
      "learning_rate": 3.320179541118711e-06,
      "loss": 0.82443702,
      "num_input_tokens_seen": 52084940,
      "step": 2432,
      "time_per_iteration": 2.9789607524871826
    },
    {
      "auxiliary_loss_clip": 0.01067204,
      "auxiliary_loss_mlp": 0.01002211,
      "balance_loss_clip": 1.01967788,
      "balance_loss_mlp": 1.00068521,
      "epoch": 0.2925509529249083,
      "flos": 58081598524800.0,
      "grad_norm": 1.0008955866725573,
      "language_loss": 0.60271275,
      "learning_rate": 3.3195942878383293e-06,
      "loss": 0.62340689,
      "num_input_tokens_seen": 52141040,
      "step": 2433,
      "time_per_iteration": 3.3042244911193848
    },
    {
      "auxiliary_loss_clip": 0.01124848,
      "auxiliary_loss_mlp": 0.01029666,
      "balance_loss_clip": 1.05095124,
      "balance_loss_mlp": 1.01921129,
      "epoch": 0.2926711958155474,
      "flos": 21397122103680.0,
      "grad_norm": 2.347261650644067,
      "language_loss": 0.78223044,
      "learning_rate": 3.319008834376543e-06,
      "loss": 0.80377561,
      "num_input_tokens_seen": 52160730,
      "step": 2434,
      "time_per_iteration": 2.858396053314209
    },
    {
      "auxiliary_loss_clip": 0.01104951,
      "auxiliary_loss_mlp": 0.01034371,
      "balance_loss_clip": 1.03860009,
      "balance_loss_mlp": 1.02382052,
      "epoch": 0.2927914387061865,
      "flos": 23185796688000.0,
      "grad_norm": 3.091241503252478,
      "language_loss": 0.8874011,
      "learning_rate": 3.3184231808221654e-06,
      "loss": 0.90879428,
      "num_input_tokens_seen": 52175055,
      "step": 2435,
      "time_per_iteration": 2.8692433834075928
    },
    {
      "auxiliary_loss_clip": 0.01123369,
      "auxiliary_loss_mlp": 0.01033264,
      "balance_loss_clip": 1.05232441,
      "balance_loss_mlp": 1.0231607,
      "epoch": 0.29291168159682557,
      "flos": 22455553190400.0,
      "grad_norm": 2.9212751369574548,
      "language_loss": 0.62930685,
      "learning_rate": 3.3178373272640394e-06,
      "loss": 0.65087318,
      "num_input_tokens_seen": 52194150,
      "step": 2436,
      "time_per_iteration": 2.8216090202331543
    },
    {
      "auxiliary_loss_clip": 0.01174546,
      "auxiliary_loss_mlp": 0.01031897,
      "balance_loss_clip": 1.05552244,
      "balance_loss_mlp": 1.0222764,
      "epoch": 0.2930319244874647,
      "flos": 21170632896000.0,
      "grad_norm": 8.266566814264538,
      "language_loss": 0.85503173,
      "learning_rate": 3.3172512737910387e-06,
      "loss": 0.87709612,
      "num_input_tokens_seen": 52211660,
      "step": 2437,
      "time_per_iteration": 2.700761556625366
    },
    {
      "auxiliary_loss_clip": 0.01174464,
      "auxiliary_loss_mlp": 0.01036331,
      "balance_loss_clip": 1.05953133,
      "balance_loss_mlp": 1.02645421,
      "epoch": 0.2931521673781038,
      "flos": 31357843931520.0,
      "grad_norm": 2.8454154393991433,
      "language_loss": 0.88200873,
      "learning_rate": 3.3166650204920674e-06,
      "loss": 0.90411669,
      "num_input_tokens_seen": 52232830,
      "step": 2438,
      "time_per_iteration": 2.777876377105713
    },
    {
      "auxiliary_loss_clip": 0.01125572,
      "auxiliary_loss_mlp": 0.01027762,
      "balance_loss_clip": 1.04899287,
      "balance_loss_mlp": 1.01821375,
      "epoch": 0.29327241026874284,
      "flos": 24200990778240.0,
      "grad_norm": 1.6839182598432347,
      "language_loss": 0.81634933,
      "learning_rate": 3.316078567456059e-06,
      "loss": 0.83788264,
      "num_input_tokens_seen": 52250670,
      "step": 2439,
      "time_per_iteration": 2.812483787536621
    },
    {
      "auxiliary_loss_clip": 0.01113705,
      "auxiliary_loss_mlp": 0.01035044,
      "balance_loss_clip": 1.04808974,
      "balance_loss_mlp": 1.02488136,
      "epoch": 0.29339265315938196,
      "flos": 24242611662720.0,
      "grad_norm": 1.4534923720829984,
      "language_loss": 0.75861454,
      "learning_rate": 3.3154919147719786e-06,
      "loss": 0.78010201,
      "num_input_tokens_seen": 52271685,
      "step": 2440,
      "time_per_iteration": 2.7958555221557617
    },
    {
      "auxiliary_loss_clip": 0.01186953,
      "auxiliary_loss_mlp": 0.00764508,
      "balance_loss_clip": 1.05821753,
      "balance_loss_mlp": 1.00045431,
      "epoch": 0.29351289605002107,
      "flos": 16946641134720.0,
      "grad_norm": 2.9175211698449517,
      "language_loss": 0.86692739,
      "learning_rate": 3.31490506252882e-06,
      "loss": 0.88644207,
      "num_input_tokens_seen": 52291065,
      "step": 2441,
      "time_per_iteration": 2.690713882446289
    },
    {
      "auxiliary_loss_clip": 0.01144519,
      "auxiliary_loss_mlp": 0.01030351,
      "balance_loss_clip": 1.04603279,
      "balance_loss_mlp": 1.02088618,
      "epoch": 0.2936331389406601,
      "flos": 19829082810240.0,
      "grad_norm": 1.8027543968817399,
      "language_loss": 0.84299362,
      "learning_rate": 3.31431801081561e-06,
      "loss": 0.86474228,
      "num_input_tokens_seen": 52310000,
      "step": 2442,
      "time_per_iteration": 3.797450542449951
    },
    {
      "auxiliary_loss_clip": 0.01044519,
      "auxiliary_loss_mlp": 0.01004204,
      "balance_loss_clip": 1.0387255,
      "balance_loss_mlp": 1.00206971,
      "epoch": 0.29375338183129923,
      "flos": 71416844398080.0,
      "grad_norm": 0.8921663067033458,
      "language_loss": 0.67895579,
      "learning_rate": 3.313730759721402e-06,
      "loss": 0.69944298,
      "num_input_tokens_seen": 52372930,
      "step": 2443,
      "time_per_iteration": 4.6702539920806885
    },
    {
      "auxiliary_loss_clip": 0.01153512,
      "auxiliary_loss_mlp": 0.01036695,
      "balance_loss_clip": 1.05285954,
      "balance_loss_mlp": 1.02684259,
      "epoch": 0.29387362472193834,
      "flos": 22054502862720.0,
      "grad_norm": 2.047768714652683,
      "language_loss": 0.8642875,
      "learning_rate": 3.313143309335282e-06,
      "loss": 0.88618952,
      "num_input_tokens_seen": 52391420,
      "step": 2444,
      "time_per_iteration": 3.149419069290161
    },
    {
      "auxiliary_loss_clip": 0.0120453,
      "auxiliary_loss_mlp": 0.01031958,
      "balance_loss_clip": 1.06134701,
      "balance_loss_mlp": 1.02283216,
      "epoch": 0.2939938676125774,
      "flos": 22966418373120.0,
      "grad_norm": 1.8269047823139433,
      "language_loss": 0.84584415,
      "learning_rate": 3.3125556597463665e-06,
      "loss": 0.868209,
      "num_input_tokens_seen": 52410725,
      "step": 2445,
      "time_per_iteration": 2.6104013919830322
    },
    {
      "auxiliary_loss_clip": 0.01145175,
      "auxiliary_loss_mlp": 0.00763255,
      "balance_loss_clip": 1.0501858,
      "balance_loss_mlp": 1.00047231,
      "epoch": 0.2941141105032165,
      "flos": 31358705857920.0,
      "grad_norm": 1.7973907498228898,
      "language_loss": 0.66455781,
      "learning_rate": 3.311967811043801e-06,
      "loss": 0.68364215,
      "num_input_tokens_seen": 52432645,
      "step": 2446,
      "time_per_iteration": 3.7833046913146973
    },
    {
      "auxiliary_loss_clip": 0.01095497,
      "auxiliary_loss_mlp": 0.01030494,
      "balance_loss_clip": 1.04689169,
      "balance_loss_mlp": 1.02020597,
      "epoch": 0.29423435339385556,
      "flos": 23222138273280.0,
      "grad_norm": 3.520374383727725,
      "language_loss": 0.82021344,
      "learning_rate": 3.3113797633167617e-06,
      "loss": 0.84147334,
      "num_input_tokens_seen": 52450940,
      "step": 2447,
      "time_per_iteration": 3.7384657859802246
    },
    {
      "auxiliary_loss_clip": 0.01182723,
      "auxiliary_loss_mlp": 0.01035948,
      "balance_loss_clip": 1.05602598,
      "balance_loss_mlp": 1.02527261,
      "epoch": 0.2943545962844947,
      "flos": 26864054138880.0,
      "grad_norm": 2.725453910400299,
      "language_loss": 0.68778074,
      "learning_rate": 3.310791516654455e-06,
      "loss": 0.70996749,
      "num_input_tokens_seen": 52468000,
      "step": 2448,
      "time_per_iteration": 2.6137073040008545
    },
    {
      "auxiliary_loss_clip": 0.01156942,
      "auxiliary_loss_mlp": 0.01033591,
      "balance_loss_clip": 1.05207348,
      "balance_loss_mlp": 1.02254057,
      "epoch": 0.2944748391751338,
      "flos": 20231677422720.0,
      "grad_norm": 1.7302164244562805,
      "language_loss": 0.7902354,
      "learning_rate": 3.3102030711461177e-06,
      "loss": 0.81214082,
      "num_input_tokens_seen": 52487575,
      "step": 2449,
      "time_per_iteration": 2.622363567352295
    },
    {
      "auxiliary_loss_clip": 0.01164688,
      "auxiliary_loss_mlp": 0.0102962,
      "balance_loss_clip": 1.05134189,
      "balance_loss_mlp": 1.01886177,
      "epoch": 0.29459508206577284,
      "flos": 15960965045760.0,
      "grad_norm": 2.268554592627871,
      "language_loss": 0.68072236,
      "learning_rate": 3.3096144268810156e-06,
      "loss": 0.70266545,
      "num_input_tokens_seen": 52506335,
      "step": 2450,
      "time_per_iteration": 2.5845131874084473
    },
    {
      "auxiliary_loss_clip": 0.01085102,
      "auxiliary_loss_mlp": 0.00765199,
      "balance_loss_clip": 1.04051709,
      "balance_loss_mlp": 1.00063729,
      "epoch": 0.29471532495641195,
      "flos": 20412882558720.0,
      "grad_norm": 1.8853729997695394,
      "language_loss": 0.73099434,
      "learning_rate": 3.3090255839484462e-06,
      "loss": 0.74949729,
      "num_input_tokens_seen": 52524330,
      "step": 2451,
      "time_per_iteration": 2.7399537563323975
    },
    {
      "auxiliary_loss_clip": 0.01174118,
      "auxiliary_loss_mlp": 0.01033126,
      "balance_loss_clip": 1.05356383,
      "balance_loss_mlp": 1.02254009,
      "epoch": 0.29483556784705106,
      "flos": 20376576887040.0,
      "grad_norm": 1.9657947447413424,
      "language_loss": 0.85400069,
      "learning_rate": 3.3084365424377366e-06,
      "loss": 0.87607312,
      "num_input_tokens_seen": 52543095,
      "step": 2452,
      "time_per_iteration": 2.8119823932647705
    },
    {
      "auxiliary_loss_clip": 0.01102729,
      "auxiliary_loss_mlp": 0.01005811,
      "balance_loss_clip": 1.02697241,
      "balance_loss_mlp": 1.00423741,
      "epoch": 0.2949558107376901,
      "flos": 68555660595840.0,
      "grad_norm": 0.7286871514685379,
      "language_loss": 0.55951244,
      "learning_rate": 3.307847302438245e-06,
      "loss": 0.58059782,
      "num_input_tokens_seen": 52597075,
      "step": 2453,
      "time_per_iteration": 3.0620346069335938
    },
    {
      "auxiliary_loss_clip": 0.01104148,
      "auxiliary_loss_mlp": 0.01032324,
      "balance_loss_clip": 1.0491662,
      "balance_loss_mlp": 1.02214909,
      "epoch": 0.2950760536283292,
      "flos": 16107085572480.0,
      "grad_norm": 2.5560884743659575,
      "language_loss": 0.77419138,
      "learning_rate": 3.3072578640393562e-06,
      "loss": 0.79555613,
      "num_input_tokens_seen": 52614410,
      "step": 2454,
      "time_per_iteration": 2.738950729370117
    },
    {
      "auxiliary_loss_clip": 0.01191127,
      "auxiliary_loss_mlp": 0.01034024,
      "balance_loss_clip": 1.05922532,
      "balance_loss_mlp": 1.0239625,
      "epoch": 0.29519629651896834,
      "flos": 20483626394880.0,
      "grad_norm": 2.4236784920460197,
      "language_loss": 0.79433382,
      "learning_rate": 3.3066682273304886e-06,
      "loss": 0.81658524,
      "num_input_tokens_seen": 52632055,
      "step": 2455,
      "time_per_iteration": 2.737192392349243
    },
    {
      "auxiliary_loss_clip": 0.01156968,
      "auxiliary_loss_mlp": 0.01028369,
      "balance_loss_clip": 1.05412018,
      "balance_loss_mlp": 1.01788485,
      "epoch": 0.2953165394096074,
      "flos": 18916484941440.0,
      "grad_norm": 2.2615668655972345,
      "language_loss": 0.78992784,
      "learning_rate": 3.3060783924010904e-06,
      "loss": 0.81178117,
      "num_input_tokens_seen": 52649980,
      "step": 2456,
      "time_per_iteration": 2.7486753463745117
    },
    {
      "auxiliary_loss_clip": 0.01173575,
      "auxiliary_loss_mlp": 0.01037818,
      "balance_loss_clip": 1.05680776,
      "balance_loss_mlp": 1.02702928,
      "epoch": 0.2954367823002465,
      "flos": 20624467622400.0,
      "grad_norm": 2.3640036477664728,
      "language_loss": 0.8451643,
      "learning_rate": 3.3054883593406387e-06,
      "loss": 0.86727822,
      "num_input_tokens_seen": 52664730,
      "step": 2457,
      "time_per_iteration": 2.6800243854522705
    },
    {
      "auxiliary_loss_clip": 0.01138148,
      "auxiliary_loss_mlp": 0.00764251,
      "balance_loss_clip": 1.05060792,
      "balance_loss_mlp": 1.00056589,
      "epoch": 0.2955570251908856,
      "flos": 31175525473920.0,
      "grad_norm": 2.2554362890584785,
      "language_loss": 0.64953393,
      "learning_rate": 3.3048981282386404e-06,
      "loss": 0.66855788,
      "num_input_tokens_seen": 52686040,
      "step": 2458,
      "time_per_iteration": 2.8256306648254395
    },
    {
      "auxiliary_loss_clip": 0.01171089,
      "auxiliary_loss_mlp": 0.01032676,
      "balance_loss_clip": 1.05714822,
      "balance_loss_mlp": 1.02219152,
      "epoch": 0.29567726808152467,
      "flos": 21650328051840.0,
      "grad_norm": 2.018594450796912,
      "language_loss": 0.82243973,
      "learning_rate": 3.304307699184634e-06,
      "loss": 0.84447742,
      "num_input_tokens_seen": 52704630,
      "step": 2459,
      "time_per_iteration": 2.7587738037109375
    },
    {
      "auxiliary_loss_clip": 0.01130763,
      "auxiliary_loss_mlp": 0.00763539,
      "balance_loss_clip": 1.05774689,
      "balance_loss_mlp": 1.0004046,
      "epoch": 0.2957975109721638,
      "flos": 24243868638720.0,
      "grad_norm": 1.6415359557354103,
      "language_loss": 0.78779644,
      "learning_rate": 3.3037170722681866e-06,
      "loss": 0.80673945,
      "num_input_tokens_seen": 52725465,
      "step": 2460,
      "time_per_iteration": 2.7902565002441406
    },
    {
      "auxiliary_loss_clip": 0.01105707,
      "auxiliary_loss_mlp": 0.01032828,
      "balance_loss_clip": 1.04668558,
      "balance_loss_mlp": 1.02327347,
      "epoch": 0.29591775386280283,
      "flos": 13479717352320.0,
      "grad_norm": 2.553572869353798,
      "language_loss": 0.67885208,
      "learning_rate": 3.3031262475788956e-06,
      "loss": 0.70023745,
      "num_input_tokens_seen": 52742405,
      "step": 2461,
      "time_per_iteration": 2.7942988872528076
    },
    {
      "auxiliary_loss_clip": 0.01186144,
      "auxiliary_loss_mlp": 0.01034349,
      "balance_loss_clip": 1.05725193,
      "balance_loss_mlp": 1.02394819,
      "epoch": 0.29603799675344195,
      "flos": 17749783284480.0,
      "grad_norm": 2.539744228366252,
      "language_loss": 0.72912037,
      "learning_rate": 3.3025352252063897e-06,
      "loss": 0.75132531,
      "num_input_tokens_seen": 52761100,
      "step": 2462,
      "time_per_iteration": 2.6419358253479004
    },
    {
      "auxiliary_loss_clip": 0.01167494,
      "auxiliary_loss_mlp": 0.00764287,
      "balance_loss_clip": 1.06126165,
      "balance_loss_mlp": 1.00044203,
      "epoch": 0.29615823964408106,
      "flos": 22783920347520.0,
      "grad_norm": 1.7468889661064262,
      "language_loss": 0.75074214,
      "learning_rate": 3.3019440052403252e-06,
      "loss": 0.77005994,
      "num_input_tokens_seen": 52780965,
      "step": 2463,
      "time_per_iteration": 2.7525055408477783
    },
    {
      "auxiliary_loss_clip": 0.01175167,
      "auxiliary_loss_mlp": 0.01036438,
      "balance_loss_clip": 1.05855393,
      "balance_loss_mlp": 1.02587605,
      "epoch": 0.2962784825347201,
      "flos": 23514199758720.0,
      "grad_norm": 1.7369223734356578,
      "language_loss": 0.71029723,
      "learning_rate": 3.30135258777039e-06,
      "loss": 0.73241329,
      "num_input_tokens_seen": 52800335,
      "step": 2464,
      "time_per_iteration": 2.7593696117401123
    },
    {
      "auxiliary_loss_clip": 0.01176278,
      "auxiliary_loss_mlp": 0.01031072,
      "balance_loss_clip": 1.05451334,
      "balance_loss_mlp": 1.02050376,
      "epoch": 0.2963987254253592,
      "flos": 16362769559040.0,
      "grad_norm": 2.1229904015593313,
      "language_loss": 0.70665658,
      "learning_rate": 3.3007609728863024e-06,
      "loss": 0.72873002,
      "num_input_tokens_seen": 52818425,
      "step": 2465,
      "time_per_iteration": 2.6748874187469482
    },
    {
      "auxiliary_loss_clip": 0.01168384,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.05389154,
      "balance_loss_mlp": 1.0223732,
      "epoch": 0.29651896831599833,
      "flos": 33472263980160.0,
      "grad_norm": 1.9965345121183715,
      "language_loss": 0.72781014,
      "learning_rate": 3.300169160677809e-06,
      "loss": 0.74981397,
      "num_input_tokens_seen": 52842340,
      "step": 2466,
      "time_per_iteration": 2.789761781692505
    },
    {
      "auxiliary_loss_clip": 0.01163152,
      "auxiliary_loss_mlp": 0.01042222,
      "balance_loss_clip": 1.05742693,
      "balance_loss_mlp": 1.03064644,
      "epoch": 0.2966392112066374,
      "flos": 23805363404160.0,
      "grad_norm": 6.506822936851549,
      "language_loss": 0.77849245,
      "learning_rate": 3.2995771512346878e-06,
      "loss": 0.80054623,
      "num_input_tokens_seen": 52860690,
      "step": 2467,
      "time_per_iteration": 2.717237710952759
    },
    {
      "auxiliary_loss_clip": 0.01150277,
      "auxiliary_loss_mlp": 0.00764799,
      "balance_loss_clip": 1.05144119,
      "balance_loss_mlp": 1.00044131,
      "epoch": 0.2967594540972765,
      "flos": 19938466702080.0,
      "grad_norm": 2.187359452096973,
      "language_loss": 0.73622787,
      "learning_rate": 3.298984944646746e-06,
      "loss": 0.75537866,
      "num_input_tokens_seen": 52879370,
      "step": 2468,
      "time_per_iteration": 4.030263423919678
    },
    {
      "auxiliary_loss_clip": 0.01149464,
      "auxiliary_loss_mlp": 0.0076353,
      "balance_loss_clip": 1.05388558,
      "balance_loss_mlp": 1.00039959,
      "epoch": 0.2968796969879156,
      "flos": 23732823888000.0,
      "grad_norm": 1.7535136832693332,
      "language_loss": 0.81787431,
      "learning_rate": 3.298392541003822e-06,
      "loss": 0.83700424,
      "num_input_tokens_seen": 52898775,
      "step": 2469,
      "time_per_iteration": 3.80946683883667
    },
    {
      "auxiliary_loss_clip": 0.01203348,
      "auxiliary_loss_mlp": 0.00763819,
      "balance_loss_clip": 1.06013274,
      "balance_loss_mlp": 1.00036573,
      "epoch": 0.29699993987855466,
      "flos": 22893699288960.0,
      "grad_norm": 1.6765966750461667,
      "language_loss": 0.89649606,
      "learning_rate": 3.2977999403957806e-06,
      "loss": 0.91616774,
      "num_input_tokens_seen": 52917535,
      "step": 2470,
      "time_per_iteration": 2.66635799407959
    },
    {
      "auxiliary_loss_clip": 0.01115016,
      "auxiliary_loss_mlp": 0.01040303,
      "balance_loss_clip": 1.05074716,
      "balance_loss_mlp": 1.02994943,
      "epoch": 0.2971201827691938,
      "flos": 33832555349760.0,
      "grad_norm": 13.21849761170167,
      "language_loss": 0.67060101,
      "learning_rate": 3.2972071429125207e-06,
      "loss": 0.69215423,
      "num_input_tokens_seen": 52938755,
      "step": 2471,
      "time_per_iteration": 3.8003904819488525
    },
    {
      "auxiliary_loss_clip": 0.01103402,
      "auxiliary_loss_mlp": 0.01035472,
      "balance_loss_clip": 1.05036438,
      "balance_loss_mlp": 1.02425408,
      "epoch": 0.2972404256598329,
      "flos": 22054359208320.0,
      "grad_norm": 2.6111765923051533,
      "language_loss": 0.88385713,
      "learning_rate": 3.2966141486439682e-06,
      "loss": 0.90524584,
      "num_input_tokens_seen": 52957945,
      "step": 2472,
      "time_per_iteration": 3.0888442993164062
    },
    {
      "auxiliary_loss_clip": 0.0118034,
      "auxiliary_loss_mlp": 0.01033678,
      "balance_loss_clip": 1.05639267,
      "balance_loss_mlp": 1.02300847,
      "epoch": 0.29736066855047194,
      "flos": 31978595796480.0,
      "grad_norm": 2.244944413184716,
      "language_loss": 0.6397236,
      "learning_rate": 3.29602095768008e-06,
      "loss": 0.66186374,
      "num_input_tokens_seen": 52978460,
      "step": 2473,
      "time_per_iteration": 4.396078586578369
    },
    {
      "auxiliary_loss_clip": 0.01090003,
      "auxiliary_loss_mlp": 0.01031385,
      "balance_loss_clip": 1.04290462,
      "balance_loss_mlp": 1.02075171,
      "epoch": 0.29748091144111105,
      "flos": 33510401245440.0,
      "grad_norm": 1.7853720922386496,
      "language_loss": 0.63513768,
      "learning_rate": 3.2954275701108437e-06,
      "loss": 0.65635151,
      "num_input_tokens_seen": 52999640,
      "step": 2474,
      "time_per_iteration": 2.937268018722534
    },
    {
      "auxiliary_loss_clip": 0.01150995,
      "auxiliary_loss_mlp": 0.01038381,
      "balance_loss_clip": 1.05386627,
      "balance_loss_mlp": 1.02770615,
      "epoch": 0.29760115433175016,
      "flos": 41283373409280.0,
      "grad_norm": 2.817893820605328,
      "language_loss": 0.68300295,
      "learning_rate": 3.294833986026275e-06,
      "loss": 0.70489681,
      "num_input_tokens_seen": 53022880,
      "step": 2475,
      "time_per_iteration": 2.9078502655029297
    },
    {
      "auxiliary_loss_clip": 0.01118796,
      "auxiliary_loss_mlp": 0.01030119,
      "balance_loss_clip": 1.05069578,
      "balance_loss_mlp": 1.01974142,
      "epoch": 0.2977213972223892,
      "flos": 24493339572480.0,
      "grad_norm": 2.1056731852255237,
      "language_loss": 0.8517617,
      "learning_rate": 3.29424020551642e-06,
      "loss": 0.87325084,
      "num_input_tokens_seen": 53041515,
      "step": 2476,
      "time_per_iteration": 2.8048317432403564
    },
    {
      "auxiliary_loss_clip": 0.01114219,
      "auxiliary_loss_mlp": 0.01032978,
      "balance_loss_clip": 1.04939914,
      "balance_loss_mlp": 1.02235055,
      "epoch": 0.2978416401130283,
      "flos": 21285116519040.0,
      "grad_norm": 1.9866494713428307,
      "language_loss": 0.7184037,
      "learning_rate": 3.2936462286713546e-06,
      "loss": 0.73987567,
      "num_input_tokens_seen": 53059865,
      "step": 2477,
      "time_per_iteration": 2.9124021530151367
    },
    {
      "auxiliary_loss_clip": 0.01175627,
      "auxiliary_loss_mlp": 0.01036484,
      "balance_loss_clip": 1.05762458,
      "balance_loss_mlp": 1.02542126,
      "epoch": 0.2979618830036674,
      "flos": 25772154554880.0,
      "grad_norm": 3.58820273621079,
      "language_loss": 0.77621627,
      "learning_rate": 3.2930520555811846e-06,
      "loss": 0.79833734,
      "num_input_tokens_seen": 53079490,
      "step": 2478,
      "time_per_iteration": 2.7427477836608887
    },
    {
      "auxiliary_loss_clip": 0.01115134,
      "auxiliary_loss_mlp": 0.01033104,
      "balance_loss_clip": 1.04322541,
      "balance_loss_mlp": 1.0230546,
      "epoch": 0.2980821258943065,
      "flos": 23476996247040.0,
      "grad_norm": 1.8276910581230406,
      "language_loss": 0.79583836,
      "learning_rate": 3.292457686336046e-06,
      "loss": 0.8173207,
      "num_input_tokens_seen": 53098810,
      "step": 2479,
      "time_per_iteration": 2.8623576164245605
    },
    {
      "auxiliary_loss_clip": 0.01052085,
      "auxiliary_loss_mlp": 0.01006027,
      "balance_loss_clip": 1.0243113,
      "balance_loss_mlp": 1.00475776,
      "epoch": 0.2982023687849456,
      "flos": 69752314195200.0,
      "grad_norm": 0.8619873199996234,
      "language_loss": 0.61249053,
      "learning_rate": 3.291863121026105e-06,
      "loss": 0.63307166,
      "num_input_tokens_seen": 53162590,
      "step": 2480,
      "time_per_iteration": 3.3785560131073
    },
    {
      "auxiliary_loss_clip": 0.01159975,
      "auxiliary_loss_mlp": 0.01035746,
      "balance_loss_clip": 1.05319691,
      "balance_loss_mlp": 1.02521944,
      "epoch": 0.29832261167558466,
      "flos": 29825930741760.0,
      "grad_norm": 1.9854560252397058,
      "language_loss": 0.77121919,
      "learning_rate": 3.2912683597415547e-06,
      "loss": 0.79317641,
      "num_input_tokens_seen": 53186675,
      "step": 2481,
      "time_per_iteration": 2.8242125511169434
    },
    {
      "auxiliary_loss_clip": 0.01174508,
      "auxiliary_loss_mlp": 0.01033266,
      "balance_loss_clip": 1.0558641,
      "balance_loss_mlp": 1.0228765,
      "epoch": 0.29844285456622377,
      "flos": 33910158683520.0,
      "grad_norm": 5.249827646997629,
      "language_loss": 0.78347349,
      "learning_rate": 3.2906734025726213e-06,
      "loss": 0.80555123,
      "num_input_tokens_seen": 53205940,
      "step": 2482,
      "time_per_iteration": 2.810451030731201
    },
    {
      "auxiliary_loss_clip": 0.01124572,
      "auxiliary_loss_mlp": 0.01031995,
      "balance_loss_clip": 1.04525375,
      "balance_loss_mlp": 1.02145076,
      "epoch": 0.2985630974568629,
      "flos": 23876933253120.0,
      "grad_norm": 2.6739374539675875,
      "language_loss": 0.88259643,
      "learning_rate": 3.290078249609559e-06,
      "loss": 0.90416205,
      "num_input_tokens_seen": 53225360,
      "step": 2483,
      "time_per_iteration": 2.743713617324829
    },
    {
      "auxiliary_loss_clip": 0.01123431,
      "auxiliary_loss_mlp": 0.01031642,
      "balance_loss_clip": 1.04947639,
      "balance_loss_mlp": 1.02149117,
      "epoch": 0.29868334034750194,
      "flos": 21799106184960.0,
      "grad_norm": 2.2209816106896967,
      "language_loss": 0.88047135,
      "learning_rate": 3.2894829009426514e-06,
      "loss": 0.90202212,
      "num_input_tokens_seen": 53243195,
      "step": 2484,
      "time_per_iteration": 2.806445598602295
    },
    {
      "auxiliary_loss_clip": 0.01170703,
      "auxiliary_loss_mlp": 0.0076365,
      "balance_loss_clip": 1.05718374,
      "balance_loss_mlp": 1.00026846,
      "epoch": 0.29880358323814105,
      "flos": 25666649331840.0,
      "grad_norm": 1.8045140992782787,
      "language_loss": 0.77766156,
      "learning_rate": 3.288887356662213e-06,
      "loss": 0.79700506,
      "num_input_tokens_seen": 53264530,
      "step": 2485,
      "time_per_iteration": 2.7401535511016846
    },
    {
      "auxiliary_loss_clip": 0.01049014,
      "auxiliary_loss_mlp": 0.01003599,
      "balance_loss_clip": 1.02467561,
      "balance_loss_mlp": 1.00191784,
      "epoch": 0.29892382612878016,
      "flos": 71005846003200.0,
      "grad_norm": 2.7066814478049204,
      "language_loss": 0.59653652,
      "learning_rate": 3.288291616858588e-06,
      "loss": 0.61706257,
      "num_input_tokens_seen": 53319920,
      "step": 2486,
      "time_per_iteration": 3.2252464294433594
    },
    {
      "auxiliary_loss_clip": 0.01174975,
      "auxiliary_loss_mlp": 0.01035661,
      "balance_loss_clip": 1.05734336,
      "balance_loss_mlp": 1.02511728,
      "epoch": 0.2990440690194192,
      "flos": 25481134563840.0,
      "grad_norm": 1.6716731727821141,
      "language_loss": 0.76714909,
      "learning_rate": 3.287695681622149e-06,
      "loss": 0.78925544,
      "num_input_tokens_seen": 53339270,
      "step": 2487,
      "time_per_iteration": 2.711996078491211
    },
    {
      "auxiliary_loss_clip": 0.01186023,
      "auxiliary_loss_mlp": 0.01031884,
      "balance_loss_clip": 1.05677474,
      "balance_loss_mlp": 1.02206707,
      "epoch": 0.2991643119100583,
      "flos": 23732357011200.0,
      "grad_norm": 1.7567254210424066,
      "language_loss": 0.8092376,
      "learning_rate": 3.2870995510432982e-06,
      "loss": 0.83141673,
      "num_input_tokens_seen": 53357750,
      "step": 2488,
      "time_per_iteration": 2.680511236190796
    },
    {
      "auxiliary_loss_clip": 0.01091677,
      "auxiliary_loss_mlp": 0.01035448,
      "balance_loss_clip": 1.04028082,
      "balance_loss_mlp": 1.02559495,
      "epoch": 0.29928455480069743,
      "flos": 27417545786880.0,
      "grad_norm": 1.8190289754654756,
      "language_loss": 0.7680136,
      "learning_rate": 3.2865032252124697e-06,
      "loss": 0.78928483,
      "num_input_tokens_seen": 53378265,
      "step": 2489,
      "time_per_iteration": 3.096278667449951
    },
    {
      "auxiliary_loss_clip": 0.01156456,
      "auxiliary_loss_mlp": 0.01032697,
      "balance_loss_clip": 1.05727458,
      "balance_loss_mlp": 1.02360725,
      "epoch": 0.2994047976913365,
      "flos": 33692935184640.0,
      "grad_norm": 1.4596462532029322,
      "language_loss": 0.77405238,
      "learning_rate": 3.2859067042201243e-06,
      "loss": 0.79594392,
      "num_input_tokens_seen": 53400305,
      "step": 2490,
      "time_per_iteration": 3.3406996726989746
    },
    {
      "auxiliary_loss_clip": 0.01185843,
      "auxiliary_loss_mlp": 0.0076396,
      "balance_loss_clip": 1.05893052,
      "balance_loss_mlp": 1.00039458,
      "epoch": 0.2995250405819756,
      "flos": 16763963541120.0,
      "grad_norm": 2.077048975724872,
      "language_loss": 0.78432369,
      "learning_rate": 3.2853099881567544e-06,
      "loss": 0.80382168,
      "num_input_tokens_seen": 53418705,
      "step": 2491,
      "time_per_iteration": 2.6882266998291016
    },
    {
      "auxiliary_loss_clip": 0.01146285,
      "auxiliary_loss_mlp": 0.00763574,
      "balance_loss_clip": 1.05255365,
      "balance_loss_mlp": 1.00032163,
      "epoch": 0.29964528347261465,
      "flos": 22963976248320.0,
      "grad_norm": 1.6028867736388877,
      "language_loss": 0.792409,
      "learning_rate": 3.284713077112881e-06,
      "loss": 0.81150758,
      "num_input_tokens_seen": 53438135,
      "step": 2492,
      "time_per_iteration": 2.7478201389312744
    },
    {
      "auxiliary_loss_clip": 0.01155646,
      "auxiliary_loss_mlp": 0.0103843,
      "balance_loss_clip": 1.05411315,
      "balance_loss_mlp": 1.0277195,
      "epoch": 0.29976552636325376,
      "flos": 16938021870720.0,
      "grad_norm": 2.4980156101738524,
      "language_loss": 0.86761665,
      "learning_rate": 3.284115971179056e-06,
      "loss": 0.88955736,
      "num_input_tokens_seen": 53452165,
      "step": 2493,
      "time_per_iteration": 3.762239933013916
    },
    {
      "auxiliary_loss_clip": 0.01170655,
      "auxiliary_loss_mlp": 0.00763852,
      "balance_loss_clip": 1.05508113,
      "balance_loss_mlp": 1.00029898,
      "epoch": 0.2998857692538929,
      "flos": 17056455989760.0,
      "grad_norm": 1.8312728814232622,
      "language_loss": 0.78801137,
      "learning_rate": 3.283518670445859e-06,
      "loss": 0.80735636,
      "num_input_tokens_seen": 53470075,
      "step": 2494,
      "time_per_iteration": 3.6250877380371094
    },
    {
      "auxiliary_loss_clip": 0.01063682,
      "auxiliary_loss_mlp": 0.01001543,
      "balance_loss_clip": 1.02415824,
      "balance_loss_mlp": 1.00000501,
      "epoch": 0.30000601214453193,
      "flos": 68831528025600.0,
      "grad_norm": 0.694551177095858,
      "language_loss": 0.54329729,
      "learning_rate": 3.2829211750038995e-06,
      "loss": 0.56394947,
      "num_input_tokens_seen": 53538705,
      "step": 2495,
      "time_per_iteration": 3.427942991256714
    },
    {
      "auxiliary_loss_clip": 0.01172446,
      "auxiliary_loss_mlp": 0.01034153,
      "balance_loss_clip": 1.05706108,
      "balance_loss_mlp": 1.02382946,
      "epoch": 0.30012625503517104,
      "flos": 17603267708160.0,
      "grad_norm": 1.7607918280958343,
      "language_loss": 0.89211714,
      "learning_rate": 3.2823234849438183e-06,
      "loss": 0.91418314,
      "num_input_tokens_seen": 53556740,
      "step": 2496,
      "time_per_iteration": 2.6724929809570312
    },
    {
      "auxiliary_loss_clip": 0.0112812,
      "auxiliary_loss_mlp": 0.01032298,
      "balance_loss_clip": 1.0474565,
      "balance_loss_mlp": 1.022964,
      "epoch": 0.30024649792581015,
      "flos": 21252581775360.0,
      "grad_norm": 2.078137818561309,
      "language_loss": 0.75652444,
      "learning_rate": 3.2817256003562836e-06,
      "loss": 0.77812862,
      "num_input_tokens_seen": 53577115,
      "step": 2497,
      "time_per_iteration": 3.6752707958221436
    },
    {
      "auxiliary_loss_clip": 0.01112482,
      "auxiliary_loss_mlp": 0.01029459,
      "balance_loss_clip": 1.05225873,
      "balance_loss_mlp": 1.02033305,
      "epoch": 0.3003667408164492,
      "flos": 23003262748800.0,
      "grad_norm": 2.9514735287222735,
      "language_loss": 0.65868658,
      "learning_rate": 3.281127521331995e-06,
      "loss": 0.68010598,
      "num_input_tokens_seen": 53598295,
      "step": 2498,
      "time_per_iteration": 3.968989849090576
    },
    {
      "auxiliary_loss_clip": 0.01042794,
      "auxiliary_loss_mlp": 0.01005586,
      "balance_loss_clip": 1.04918492,
      "balance_loss_mlp": 1.00395322,
      "epoch": 0.3004869837070883,
      "flos": 64232340750720.0,
      "grad_norm": 0.8848072156848285,
      "language_loss": 0.60709941,
      "learning_rate": 3.2805292479616798e-06,
      "loss": 0.62758315,
      "num_input_tokens_seen": 53657160,
      "step": 2499,
      "time_per_iteration": 3.613173484802246
    },
    {
      "auxiliary_loss_clip": 0.01163094,
      "auxiliary_loss_mlp": 0.00764135,
      "balance_loss_clip": 1.05406749,
      "balance_loss_mlp": 1.00042534,
      "epoch": 0.30060722659772743,
      "flos": 26248653400320.0,
      "grad_norm": 2.4375260230576936,
      "language_loss": 0.91517639,
      "learning_rate": 3.2799307803360955e-06,
      "loss": 0.93444866,
      "num_input_tokens_seen": 53673090,
      "step": 2500,
      "time_per_iteration": 2.7666265964508057
    },
    {
      "auxiliary_loss_clip": 0.01112343,
      "auxiliary_loss_mlp": 0.01027919,
      "balance_loss_clip": 1.0487349,
      "balance_loss_mlp": 1.01857328,
      "epoch": 0.3007274694883665,
      "flos": 24970879912320.0,
      "grad_norm": 2.598363846124977,
      "language_loss": 0.81381792,
      "learning_rate": 3.27933211854603e-06,
      "loss": 0.83522058,
      "num_input_tokens_seen": 53692145,
      "step": 2501,
      "time_per_iteration": 2.8373260498046875
    },
    {
      "auxiliary_loss_clip": 0.01130832,
      "auxiliary_loss_mlp": 0.01031036,
      "balance_loss_clip": 1.0526787,
      "balance_loss_mlp": 1.02118945,
      "epoch": 0.3008477123790056,
      "flos": 17055845458560.0,
      "grad_norm": 1.7122220633639595,
      "language_loss": 0.86862826,
      "learning_rate": 3.278733262682299e-06,
      "loss": 0.89024699,
      "num_input_tokens_seen": 53710000,
      "step": 2502,
      "time_per_iteration": 2.7716963291168213
    },
    {
      "auxiliary_loss_clip": 0.0118015,
      "auxiliary_loss_mlp": 0.0103439,
      "balance_loss_clip": 1.05378056,
      "balance_loss_mlp": 1.0244714,
      "epoch": 0.3009679552696447,
      "flos": 21506398254720.0,
      "grad_norm": 2.412467775762623,
      "language_loss": 0.8238408,
      "learning_rate": 3.2781342128357484e-06,
      "loss": 0.84598619,
      "num_input_tokens_seen": 53729355,
      "step": 2503,
      "time_per_iteration": 2.6848275661468506
    },
    {
      "auxiliary_loss_clip": 0.01171918,
      "auxiliary_loss_mlp": 0.01034943,
      "balance_loss_clip": 1.05859709,
      "balance_loss_mlp": 1.02568674,
      "epoch": 0.30108819816028376,
      "flos": 21134004001920.0,
      "grad_norm": 3.3661141810810076,
      "language_loss": 0.80357999,
      "learning_rate": 3.2775349690972547e-06,
      "loss": 0.82564855,
      "num_input_tokens_seen": 53743505,
      "step": 2504,
      "time_per_iteration": 2.6707539558410645
    },
    {
      "auxiliary_loss_clip": 0.01071808,
      "auxiliary_loss_mlp": 0.0100513,
      "balance_loss_clip": 1.04353762,
      "balance_loss_mlp": 1.00350857,
      "epoch": 0.30120844105092287,
      "flos": 71126434938240.0,
      "grad_norm": 0.7479389745423367,
      "language_loss": 0.51799977,
      "learning_rate": 3.276935531557722e-06,
      "loss": 0.53876919,
      "num_input_tokens_seen": 53808725,
      "step": 2505,
      "time_per_iteration": 3.439786434173584
    },
    {
      "auxiliary_loss_clip": 0.01096829,
      "auxiliary_loss_mlp": 0.01032961,
      "balance_loss_clip": 1.0451045,
      "balance_loss_mlp": 1.02233338,
      "epoch": 0.301328683941562,
      "flos": 20264571302400.0,
      "grad_norm": 2.590913437444009,
      "language_loss": 0.79286873,
      "learning_rate": 3.2763359003080837e-06,
      "loss": 0.81416667,
      "num_input_tokens_seen": 53825680,
      "step": 2506,
      "time_per_iteration": 2.9543638229370117
    },
    {
      "auxiliary_loss_clip": 0.01060288,
      "auxiliary_loss_mlp": 0.01006363,
      "balance_loss_clip": 1.02005982,
      "balance_loss_mlp": 1.00508714,
      "epoch": 0.30144892683220104,
      "flos": 70648212240000.0,
      "grad_norm": 0.8054084796232313,
      "language_loss": 0.62419552,
      "learning_rate": 3.2757360754393047e-06,
      "loss": 0.64486206,
      "num_input_tokens_seen": 53889750,
      "step": 2507,
      "time_per_iteration": 3.397045373916626
    },
    {
      "auxiliary_loss_clip": 0.01110496,
      "auxiliary_loss_mlp": 0.0103497,
      "balance_loss_clip": 1.04868948,
      "balance_loss_mlp": 1.0246824,
      "epoch": 0.30156916972284015,
      "flos": 22820549241600.0,
      "grad_norm": 3.81937744765038,
      "language_loss": 0.63862646,
      "learning_rate": 3.2751360570423767e-06,
      "loss": 0.66008109,
      "num_input_tokens_seen": 53908135,
      "step": 2508,
      "time_per_iteration": 2.8229119777679443
    },
    {
      "auxiliary_loss_clip": 0.01172749,
      "auxiliary_loss_mlp": 0.00764298,
      "balance_loss_clip": 1.05434012,
      "balance_loss_mlp": 1.00047755,
      "epoch": 0.3016894126134792,
      "flos": 29899188529920.0,
      "grad_norm": 2.0231128318048346,
      "language_loss": 0.75473648,
      "learning_rate": 3.2745358452083236e-06,
      "loss": 0.77410698,
      "num_input_tokens_seen": 53931035,
      "step": 2509,
      "time_per_iteration": 2.8422484397888184
    },
    {
      "auxiliary_loss_clip": 0.01124707,
      "auxiliary_loss_mlp": 0.01030374,
      "balance_loss_clip": 1.0469228,
      "balance_loss_mlp": 1.02162421,
      "epoch": 0.3018096555041183,
      "flos": 21546331200000.0,
      "grad_norm": 1.3646860614774792,
      "language_loss": 0.82405901,
      "learning_rate": 3.2739354400281955e-06,
      "loss": 0.8456099,
      "num_input_tokens_seen": 53952255,
      "step": 2510,
      "time_per_iteration": 2.8503470420837402
    },
    {
      "auxiliary_loss_clip": 0.01079037,
      "auxiliary_loss_mlp": 0.01001546,
      "balance_loss_clip": 1.02731454,
      "balance_loss_mlp": 1.00018144,
      "epoch": 0.3019298983947574,
      "flos": 59136294597120.0,
      "grad_norm": 0.8646767249636138,
      "language_loss": 0.63726234,
      "learning_rate": 3.2733348415930744e-06,
      "loss": 0.65806818,
      "num_input_tokens_seen": 54014125,
      "step": 2511,
      "time_per_iteration": 3.267677068710327
    },
    {
      "auxiliary_loss_clip": 0.01182517,
      "auxiliary_loss_mlp": 0.0102692,
      "balance_loss_clip": 1.05568326,
      "balance_loss_mlp": 1.0176692,
      "epoch": 0.3020501412853965,
      "flos": 34423070941440.0,
      "grad_norm": 2.0676644371935033,
      "language_loss": 0.80810404,
      "learning_rate": 3.27273404999407e-06,
      "loss": 0.83019841,
      "num_input_tokens_seen": 54036345,
      "step": 2512,
      "time_per_iteration": 2.7897729873657227
    },
    {
      "auxiliary_loss_clip": 0.01064373,
      "auxiliary_loss_mlp": 0.01002062,
      "balance_loss_clip": 1.01885271,
      "balance_loss_mlp": 1.00064373,
      "epoch": 0.3021703841760356,
      "flos": 71008288128000.0,
      "grad_norm": 0.8012613496934918,
      "language_loss": 0.60447586,
      "learning_rate": 3.272133065322322e-06,
      "loss": 0.62514019,
      "num_input_tokens_seen": 54094615,
      "step": 2513,
      "time_per_iteration": 3.2358593940734863
    },
    {
      "auxiliary_loss_clip": 0.01112481,
      "auxiliary_loss_mlp": 0.01031154,
      "balance_loss_clip": 1.04210925,
      "balance_loss_mlp": 1.02152753,
      "epoch": 0.3022906270666747,
      "flos": 21510528318720.0,
      "grad_norm": 1.6533079199549552,
      "language_loss": 0.7913748,
      "learning_rate": 3.271531887669e-06,
      "loss": 0.81281114,
      "num_input_tokens_seen": 54114675,
      "step": 2514,
      "time_per_iteration": 2.9417426586151123
    },
    {
      "auxiliary_loss_clip": 0.01150018,
      "auxiliary_loss_mlp": 0.01034188,
      "balance_loss_clip": 1.05217361,
      "balance_loss_mlp": 1.02337527,
      "epoch": 0.30241086995731375,
      "flos": 31132001168640.0,
      "grad_norm": 2.330250744139501,
      "language_loss": 0.63588095,
      "learning_rate": 3.2709305171253015e-06,
      "loss": 0.65772307,
      "num_input_tokens_seen": 54134795,
      "step": 2515,
      "time_per_iteration": 2.831024646759033
    },
    {
      "auxiliary_loss_clip": 0.01141824,
      "auxiliary_loss_mlp": 0.01032308,
      "balance_loss_clip": 1.05111206,
      "balance_loss_mlp": 1.02312243,
      "epoch": 0.30253111284795287,
      "flos": 23511542152320.0,
      "grad_norm": 1.8044466229322396,
      "language_loss": 0.77802467,
      "learning_rate": 3.2703289537824536e-06,
      "loss": 0.799766,
      "num_input_tokens_seen": 54154595,
      "step": 2516,
      "time_per_iteration": 2.8037869930267334
    },
    {
      "auxiliary_loss_clip": 0.01171392,
      "auxiliary_loss_mlp": 0.01031544,
      "balance_loss_clip": 1.06052232,
      "balance_loss_mlp": 1.02166188,
      "epoch": 0.302651355738592,
      "flos": 18725367651840.0,
      "grad_norm": 2.7047771112110457,
      "language_loss": 0.79255724,
      "learning_rate": 3.269727197731714e-06,
      "loss": 0.81458658,
      "num_input_tokens_seen": 54167360,
      "step": 2517,
      "time_per_iteration": 2.7079598903656006
    },
    {
      "auxiliary_loss_clip": 0.01156708,
      "auxiliary_loss_mlp": 0.01025764,
      "balance_loss_clip": 1.05141711,
      "balance_loss_mlp": 1.01636434,
      "epoch": 0.30277159862923103,
      "flos": 22418888382720.0,
      "grad_norm": 1.892419325105783,
      "language_loss": 0.78272551,
      "learning_rate": 3.269125249064367e-06,
      "loss": 0.80455023,
      "num_input_tokens_seen": 54187055,
      "step": 2518,
      "time_per_iteration": 2.6872661113739014
    },
    {
      "auxiliary_loss_clip": 0.01140654,
      "auxiliary_loss_mlp": 0.01034442,
      "balance_loss_clip": 1.04782188,
      "balance_loss_mlp": 1.02411819,
      "epoch": 0.30289184151987014,
      "flos": 22273126992000.0,
      "grad_norm": 3.244941629378159,
      "language_loss": 0.83251905,
      "learning_rate": 3.2685231078717297e-06,
      "loss": 0.85426998,
      "num_input_tokens_seen": 54207245,
      "step": 2519,
      "time_per_iteration": 3.6906447410583496
    },
    {
      "auxiliary_loss_clip": 0.01139112,
      "auxiliary_loss_mlp": 0.01036009,
      "balance_loss_clip": 1.04948616,
      "balance_loss_mlp": 1.0258348,
      "epoch": 0.30301208441050925,
      "flos": 25225594231680.0,
      "grad_norm": 7.142758420758249,
      "language_loss": 0.75287628,
      "learning_rate": 3.267920774245145e-06,
      "loss": 0.77462757,
      "num_input_tokens_seen": 54226650,
      "step": 2520,
      "time_per_iteration": 3.7708890438079834
    },
    {
      "auxiliary_loss_clip": 0.01104578,
      "auxiliary_loss_mlp": 0.01036392,
      "balance_loss_clip": 1.04215848,
      "balance_loss_mlp": 1.02489471,
      "epoch": 0.3031323273011483,
      "flos": 23039245198080.0,
      "grad_norm": 1.9386282597369902,
      "language_loss": 0.84721947,
      "learning_rate": 3.2673182482759876e-06,
      "loss": 0.86862916,
      "num_input_tokens_seen": 54245765,
      "step": 2521,
      "time_per_iteration": 2.8771464824676514
    },
    {
      "auxiliary_loss_clip": 0.01141103,
      "auxiliary_loss_mlp": 0.01034327,
      "balance_loss_clip": 1.04918253,
      "balance_loss_mlp": 1.0247848,
      "epoch": 0.3032525701917874,
      "flos": 18876695650560.0,
      "grad_norm": 2.273852973756494,
      "language_loss": 0.65741193,
      "learning_rate": 3.266715530055659e-06,
      "loss": 0.67916626,
      "num_input_tokens_seen": 54263915,
      "step": 2522,
      "time_per_iteration": 3.9609873294830322
    },
    {
      "auxiliary_loss_clip": 0.0112897,
      "auxiliary_loss_mlp": 0.01035857,
      "balance_loss_clip": 1.04982543,
      "balance_loss_mlp": 1.02600491,
      "epoch": 0.30337281308242653,
      "flos": 17782641250560.0,
      "grad_norm": 1.5500992856383249,
      "language_loss": 0.80300754,
      "learning_rate": 3.2661126196755927e-06,
      "loss": 0.82465577,
      "num_input_tokens_seen": 54283025,
      "step": 2523,
      "time_per_iteration": 2.841782331466675
    },
    {
      "auxiliary_loss_clip": 0.01086676,
      "auxiliary_loss_mlp": 0.01001656,
      "balance_loss_clip": 1.02187705,
      "balance_loss_mlp": 1.00026691,
      "epoch": 0.3034930559730656,
      "flos": 57824298426240.0,
      "grad_norm": 0.7753255017114641,
      "language_loss": 0.55905789,
      "learning_rate": 3.265509517227248e-06,
      "loss": 0.57994121,
      "num_input_tokens_seen": 54339840,
      "step": 2524,
      "time_per_iteration": 4.030854940414429
    },
    {
      "auxiliary_loss_clip": 0.01123003,
      "auxiliary_loss_mlp": 0.0104534,
      "balance_loss_clip": 1.04622352,
      "balance_loss_mlp": 1.03568459,
      "epoch": 0.3036132988637047,
      "flos": 14755587419520.0,
      "grad_norm": 6.345520252677623,
      "language_loss": 0.80867434,
      "learning_rate": 3.264906222802115e-06,
      "loss": 0.83035779,
      "num_input_tokens_seen": 54357690,
      "step": 2525,
      "time_per_iteration": 2.7662951946258545
    },
    {
      "auxiliary_loss_clip": 0.01191954,
      "auxiliary_loss_mlp": 0.01031727,
      "balance_loss_clip": 1.05676436,
      "balance_loss_mlp": 1.02164745,
      "epoch": 0.30373354175434375,
      "flos": 21033203460480.0,
      "grad_norm": 2.15752380520622,
      "language_loss": 0.77967775,
      "learning_rate": 3.264302736491715e-06,
      "loss": 0.80191457,
      "num_input_tokens_seen": 54377810,
      "step": 2526,
      "time_per_iteration": 2.697908639907837
    },
    {
      "auxiliary_loss_clip": 0.01153937,
      "auxiliary_loss_mlp": 0.01032452,
      "balance_loss_clip": 1.05475521,
      "balance_loss_mlp": 1.02295732,
      "epoch": 0.30385378464498286,
      "flos": 21143233797120.0,
      "grad_norm": 1.9394391831932247,
      "language_loss": 0.8722018,
      "learning_rate": 3.263699058387594e-06,
      "loss": 0.89406568,
      "num_input_tokens_seen": 54395245,
      "step": 2527,
      "time_per_iteration": 2.7451629638671875
    },
    {
      "auxiliary_loss_clip": 0.01168419,
      "auxiliary_loss_mlp": 0.01037766,
      "balance_loss_clip": 1.0561682,
      "balance_loss_mlp": 1.02720368,
      "epoch": 0.30397402753562197,
      "flos": 20629244131200.0,
      "grad_norm": 24.462837148815485,
      "language_loss": 0.90711033,
      "learning_rate": 3.2630951885813315e-06,
      "loss": 0.92917216,
      "num_input_tokens_seen": 54412640,
      "step": 2528,
      "time_per_iteration": 2.6988518238067627
    },
    {
      "auxiliary_loss_clip": 0.01188394,
      "auxiliary_loss_mlp": 0.01031499,
      "balance_loss_clip": 1.0557549,
      "balance_loss_mlp": 1.02220702,
      "epoch": 0.304094270426261,
      "flos": 15085678429440.0,
      "grad_norm": 1.9899110807195637,
      "language_loss": 0.78219283,
      "learning_rate": 3.262491127164533e-06,
      "loss": 0.80439168,
      "num_input_tokens_seen": 54431455,
      "step": 2529,
      "time_per_iteration": 2.654789686203003
    },
    {
      "auxiliary_loss_clip": 0.01181228,
      "auxiliary_loss_mlp": 0.01032969,
      "balance_loss_clip": 1.05563867,
      "balance_loss_mlp": 1.02212691,
      "epoch": 0.30421451331690014,
      "flos": 13845216193920.0,
      "grad_norm": 2.582111185625464,
      "language_loss": 0.80597675,
      "learning_rate": 3.2618868742288337e-06,
      "loss": 0.8281188,
      "num_input_tokens_seen": 54448380,
      "step": 2530,
      "time_per_iteration": 2.6487932205200195
    },
    {
      "auxiliary_loss_clip": 0.01116872,
      "auxiliary_loss_mlp": 0.01036136,
      "balance_loss_clip": 1.0505811,
      "balance_loss_mlp": 1.02593136,
      "epoch": 0.30433475620753925,
      "flos": 17384212615680.0,
      "grad_norm": 1.9214732684853402,
      "language_loss": 0.72070813,
      "learning_rate": 3.261282429865899e-06,
      "loss": 0.74223816,
      "num_input_tokens_seen": 54466385,
      "step": 2531,
      "time_per_iteration": 2.788752555847168
    },
    {
      "auxiliary_loss_clip": 0.01137518,
      "auxiliary_loss_mlp": 0.00763497,
      "balance_loss_clip": 1.05118585,
      "balance_loss_mlp": 1.00064504,
      "epoch": 0.3044549990981783,
      "flos": 18916951818240.0,
      "grad_norm": 1.623711772097314,
      "language_loss": 0.72286004,
      "learning_rate": 3.2606777941674225e-06,
      "loss": 0.74187016,
      "num_input_tokens_seen": 54485040,
      "step": 2532,
      "time_per_iteration": 2.723311185836792
    },
    {
      "auxiliary_loss_clip": 0.01166873,
      "auxiliary_loss_mlp": 0.01032587,
      "balance_loss_clip": 1.05460477,
      "balance_loss_mlp": 1.02272248,
      "epoch": 0.3045752419888174,
      "flos": 21068431724160.0,
      "grad_norm": 2.3033588917268206,
      "language_loss": 0.84753561,
      "learning_rate": 3.2600729672251276e-06,
      "loss": 0.8695302,
      "num_input_tokens_seen": 54502755,
      "step": 2533,
      "time_per_iteration": 2.700087308883667
    },
    {
      "auxiliary_loss_clip": 0.01171078,
      "auxiliary_loss_mlp": 0.01036874,
      "balance_loss_clip": 1.05912817,
      "balance_loss_mlp": 1.02699709,
      "epoch": 0.3046954848794565,
      "flos": 29096405516160.0,
      "grad_norm": 1.9747532879004797,
      "language_loss": 0.65080732,
      "learning_rate": 3.259467949130765e-06,
      "loss": 0.67288685,
      "num_input_tokens_seen": 54524165,
      "step": 2534,
      "time_per_iteration": 2.7852513790130615
    },
    {
      "auxiliary_loss_clip": 0.01140383,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.04905033,
      "balance_loss_mlp": 1.01835394,
      "epoch": 0.3048157277700956,
      "flos": 20295346279680.0,
      "grad_norm": 2.215575139930423,
      "language_loss": 0.82732427,
      "learning_rate": 3.2588627399761164e-06,
      "loss": 0.8490212,
      "num_input_tokens_seen": 54540160,
      "step": 2535,
      "time_per_iteration": 2.7882602214813232
    },
    {
      "auxiliary_loss_clip": 0.01181071,
      "auxiliary_loss_mlp": 0.0103038,
      "balance_loss_clip": 1.05761719,
      "balance_loss_mlp": 1.02045035,
      "epoch": 0.3049359706607347,
      "flos": 22739929165440.0,
      "grad_norm": 1.6191093806268975,
      "language_loss": 0.70795047,
      "learning_rate": 3.2582573398529903e-06,
      "loss": 0.73006493,
      "num_input_tokens_seen": 54557515,
      "step": 2536,
      "time_per_iteration": 2.725463390350342
    },
    {
      "auxiliary_loss_clip": 0.01135698,
      "auxiliary_loss_mlp": 0.01034402,
      "balance_loss_clip": 1.04893088,
      "balance_loss_mlp": 1.02463925,
      "epoch": 0.3050562135513738,
      "flos": 18434634969600.0,
      "grad_norm": 2.140444774897934,
      "language_loss": 0.73986685,
      "learning_rate": 3.2576517488532265e-06,
      "loss": 0.76156783,
      "num_input_tokens_seen": 54573865,
      "step": 2537,
      "time_per_iteration": 2.7230989933013916
    },
    {
      "auxiliary_loss_clip": 0.01151833,
      "auxiliary_loss_mlp": 0.00763937,
      "balance_loss_clip": 1.05317855,
      "balance_loss_mlp": 1.00063264,
      "epoch": 0.30517645644201286,
      "flos": 20370327920640.0,
      "grad_norm": 1.7558882369225601,
      "language_loss": 0.87240767,
      "learning_rate": 3.257045967068692e-06,
      "loss": 0.89156532,
      "num_input_tokens_seen": 54593120,
      "step": 2538,
      "time_per_iteration": 2.761528491973877
    },
    {
      "auxiliary_loss_clip": 0.01159809,
      "auxiliary_loss_mlp": 0.01036601,
      "balance_loss_clip": 1.05831671,
      "balance_loss_mlp": 1.02621198,
      "epoch": 0.30529669933265197,
      "flos": 21945118970880.0,
      "grad_norm": 1.582143775354837,
      "language_loss": 0.82169843,
      "learning_rate": 3.2564399945912848e-06,
      "loss": 0.8436625,
      "num_input_tokens_seen": 54612910,
      "step": 2539,
      "time_per_iteration": 2.74066162109375
    },
    {
      "auxiliary_loss_clip": 0.01185697,
      "auxiliary_loss_mlp": 0.00763052,
      "balance_loss_clip": 1.06016088,
      "balance_loss_mlp": 1.00058508,
      "epoch": 0.305416942223291,
      "flos": 21835411856640.0,
      "grad_norm": 2.24117020118918,
      "language_loss": 0.82216978,
      "learning_rate": 3.2558338315129287e-06,
      "loss": 0.84165734,
      "num_input_tokens_seen": 54631055,
      "step": 2540,
      "time_per_iteration": 2.693901300430298
    },
    {
      "auxiliary_loss_clip": 0.01140013,
      "auxiliary_loss_mlp": 0.0103207,
      "balance_loss_clip": 1.05264914,
      "balance_loss_mlp": 1.02253914,
      "epoch": 0.30553718511393013,
      "flos": 33911810709120.0,
      "grad_norm": 2.154699910112333,
      "language_loss": 0.76004386,
      "learning_rate": 3.2552274779255785e-06,
      "loss": 0.78176469,
      "num_input_tokens_seen": 54651985,
      "step": 2541,
      "time_per_iteration": 2.850451946258545
    },
    {
      "auxiliary_loss_clip": 0.01162612,
      "auxiliary_loss_mlp": 0.01035026,
      "balance_loss_clip": 1.05240059,
      "balance_loss_mlp": 1.02484596,
      "epoch": 0.30565742800456924,
      "flos": 22268530051200.0,
      "grad_norm": 2.058356703647301,
      "language_loss": 0.76919878,
      "learning_rate": 3.2546209339212184e-06,
      "loss": 0.79117513,
      "num_input_tokens_seen": 54671005,
      "step": 2542,
      "time_per_iteration": 2.734713077545166
    },
    {
      "auxiliary_loss_clip": 0.01149026,
      "auxiliary_loss_mlp": 0.01029987,
      "balance_loss_clip": 1.05327511,
      "balance_loss_mlp": 1.02068877,
      "epoch": 0.3057776708952083,
      "flos": 22565044823040.0,
      "grad_norm": 1.910676606193686,
      "language_loss": 0.77294612,
      "learning_rate": 3.25401419959186e-06,
      "loss": 0.79473627,
      "num_input_tokens_seen": 54691615,
      "step": 2543,
      "time_per_iteration": 2.7410237789154053
    },
    {
      "auxiliary_loss_clip": 0.01167546,
      "auxiliary_loss_mlp": 0.01034906,
      "balance_loss_clip": 1.06289756,
      "balance_loss_mlp": 1.02537489,
      "epoch": 0.3058979137858474,
      "flos": 21799213925760.0,
      "grad_norm": 2.101573094415959,
      "language_loss": 0.75843418,
      "learning_rate": 3.253407275029545e-06,
      "loss": 0.78045869,
      "num_input_tokens_seen": 54710520,
      "step": 2544,
      "time_per_iteration": 2.748110771179199
    },
    {
      "auxiliary_loss_clip": 0.01117571,
      "auxiliary_loss_mlp": 0.01039445,
      "balance_loss_clip": 1.04836321,
      "balance_loss_mlp": 1.02837634,
      "epoch": 0.3060181566764865,
      "flos": 26979435601920.0,
      "grad_norm": 1.9212644184721361,
      "language_loss": 0.80160785,
      "learning_rate": 3.2528001603263425e-06,
      "loss": 0.82317793,
      "num_input_tokens_seen": 54732590,
      "step": 2545,
      "time_per_iteration": 3.8128325939178467
    },
    {
      "auxiliary_loss_clip": 0.01176467,
      "auxiliary_loss_mlp": 0.01032717,
      "balance_loss_clip": 1.05972326,
      "balance_loss_mlp": 1.02164817,
      "epoch": 0.3061383995671256,
      "flos": 19865101173120.0,
      "grad_norm": 1.7947287785789974,
      "language_loss": 0.81488085,
      "learning_rate": 3.2521928555743514e-06,
      "loss": 0.83697265,
      "num_input_tokens_seen": 54749935,
      "step": 2546,
      "time_per_iteration": 3.6457669734954834
    },
    {
      "auxiliary_loss_clip": 0.01148286,
      "auxiliary_loss_mlp": 0.01036403,
      "balance_loss_clip": 1.05431008,
      "balance_loss_mlp": 1.02442229,
      "epoch": 0.3062586424577647,
      "flos": 22127509255680.0,
      "grad_norm": 1.8435234458481675,
      "language_loss": 0.67525208,
      "learning_rate": 3.2515853608657e-06,
      "loss": 0.69709891,
      "num_input_tokens_seen": 54767935,
      "step": 2547,
      "time_per_iteration": 2.706451416015625
    },
    {
      "auxiliary_loss_clip": 0.01128476,
      "auxiliary_loss_mlp": 0.01035081,
      "balance_loss_clip": 1.05103886,
      "balance_loss_mlp": 1.02493036,
      "epoch": 0.3063788853484038,
      "flos": 20845497962880.0,
      "grad_norm": 2.2004313350001485,
      "language_loss": 0.74943441,
      "learning_rate": 3.250977676292545e-06,
      "loss": 0.77107,
      "num_input_tokens_seen": 54786175,
      "step": 2548,
      "time_per_iteration": 3.7171669006347656
    },
    {
      "auxiliary_loss_clip": 0.01150409,
      "auxiliary_loss_mlp": 0.01035324,
      "balance_loss_clip": 1.05502963,
      "balance_loss_mlp": 1.02519727,
      "epoch": 0.30649912823904285,
      "flos": 16209717707520.0,
      "grad_norm": 2.258955340702177,
      "language_loss": 0.79020321,
      "learning_rate": 3.2503698019470712e-06,
      "loss": 0.81206059,
      "num_input_tokens_seen": 54801945,
      "step": 2549,
      "time_per_iteration": 2.686704397201538
    },
    {
      "auxiliary_loss_clip": 0.01178947,
      "auxiliary_loss_mlp": 0.01033973,
      "balance_loss_clip": 1.05857396,
      "balance_loss_mlp": 1.02297616,
      "epoch": 0.30661937112968196,
      "flos": 18617815353600.0,
      "grad_norm": 2.166500847693694,
      "language_loss": 0.77970243,
      "learning_rate": 3.249761737921492e-06,
      "loss": 0.8018316,
      "num_input_tokens_seen": 54818475,
      "step": 2550,
      "time_per_iteration": 3.6232426166534424
    },
    {
      "auxiliary_loss_clip": 0.01133717,
      "auxiliary_loss_mlp": 0.01030881,
      "balance_loss_clip": 1.05324233,
      "balance_loss_mlp": 1.01997972,
      "epoch": 0.30673961402032107,
      "flos": 31390809638400.0,
      "grad_norm": 2.093747061956546,
      "language_loss": 0.74221718,
      "learning_rate": 3.249153484308051e-06,
      "loss": 0.76386321,
      "num_input_tokens_seen": 54837090,
      "step": 2551,
      "time_per_iteration": 2.8769662380218506
    },
    {
      "auxiliary_loss_clip": 0.01169589,
      "auxiliary_loss_mlp": 0.01033051,
      "balance_loss_clip": 1.05783868,
      "balance_loss_mlp": 1.02190459,
      "epoch": 0.3068598569109601,
      "flos": 20229809915520.0,
      "grad_norm": 2.425761219764262,
      "language_loss": 0.77868307,
      "learning_rate": 3.2485450411990194e-06,
      "loss": 0.80070943,
      "num_input_tokens_seen": 54856445,
      "step": 2552,
      "time_per_iteration": 2.7543280124664307
    },
    {
      "auxiliary_loss_clip": 0.01153448,
      "auxiliary_loss_mlp": 0.01036699,
      "balance_loss_clip": 1.05299783,
      "balance_loss_mlp": 1.02566075,
      "epoch": 0.30698009980159924,
      "flos": 29601991399680.0,
      "grad_norm": 1.723006530690982,
      "language_loss": 0.82131928,
      "learning_rate": 3.2479364086866983e-06,
      "loss": 0.84322083,
      "num_input_tokens_seen": 54876700,
      "step": 2553,
      "time_per_iteration": 2.8638553619384766
    },
    {
      "auxiliary_loss_clip": 0.01169529,
      "auxiliary_loss_mlp": 0.01038613,
      "balance_loss_clip": 1.06128228,
      "balance_loss_mlp": 1.0290879,
      "epoch": 0.30710034269223835,
      "flos": 23842423261440.0,
      "grad_norm": 1.8940624120184992,
      "language_loss": 0.81532913,
      "learning_rate": 3.247327586863416e-06,
      "loss": 0.83741057,
      "num_input_tokens_seen": 54897580,
      "step": 2554,
      "time_per_iteration": 2.7503538131713867
    },
    {
      "auxiliary_loss_clip": 0.01138821,
      "auxiliary_loss_mlp": 0.01029826,
      "balance_loss_clip": 1.04797471,
      "balance_loss_mlp": 1.01916826,
      "epoch": 0.3072205855828774,
      "flos": 25884986152320.0,
      "grad_norm": 2.238394431462119,
      "language_loss": 0.7705819,
      "learning_rate": 3.2467185758215304e-06,
      "loss": 0.7922684,
      "num_input_tokens_seen": 54917320,
      "step": 2555,
      "time_per_iteration": 2.8785855770111084
    },
    {
      "auxiliary_loss_clip": 0.01131476,
      "auxiliary_loss_mlp": 0.0103458,
      "balance_loss_clip": 1.04920006,
      "balance_loss_mlp": 1.02457261,
      "epoch": 0.3073408284735165,
      "flos": 22236390357120.0,
      "grad_norm": 2.7527802808232944,
      "language_loss": 0.85640663,
      "learning_rate": 3.246109375653428e-06,
      "loss": 0.87806714,
      "num_input_tokens_seen": 54934085,
      "step": 2556,
      "time_per_iteration": 2.75565505027771
    },
    {
      "auxiliary_loss_clip": 0.01076941,
      "auxiliary_loss_mlp": 0.01034562,
      "balance_loss_clip": 1.04211497,
      "balance_loss_mlp": 1.02429795,
      "epoch": 0.30746107136415557,
      "flos": 19500284689920.0,
      "grad_norm": 2.7179804730437107,
      "language_loss": 0.78377974,
      "learning_rate": 3.2454999864515243e-06,
      "loss": 0.8048948,
      "num_input_tokens_seen": 54953460,
      "step": 2557,
      "time_per_iteration": 3.1013641357421875
    },
    {
      "auxiliary_loss_clip": 0.01174718,
      "auxiliary_loss_mlp": 0.01036426,
      "balance_loss_clip": 1.05426383,
      "balance_loss_mlp": 1.02502394,
      "epoch": 0.3075813142547947,
      "flos": 21724806902400.0,
      "grad_norm": 1.6787090890924263,
      "language_loss": 0.69700825,
      "learning_rate": 3.244890408308263e-06,
      "loss": 0.71911961,
      "num_input_tokens_seen": 54974165,
      "step": 2558,
      "time_per_iteration": 3.0141491889953613
    },
    {
      "auxiliary_loss_clip": 0.01156449,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.05411053,
      "balance_loss_mlp": 1.02155793,
      "epoch": 0.3077015571454338,
      "flos": 24097963593600.0,
      "grad_norm": 2.2196264972318174,
      "language_loss": 0.61643863,
      "learning_rate": 3.2442806413161165e-06,
      "loss": 0.63831377,
      "num_input_tokens_seen": 54993810,
      "step": 2559,
      "time_per_iteration": 2.7736926078796387
    },
    {
      "auxiliary_loss_clip": 0.0116918,
      "auxiliary_loss_mlp": 0.01036511,
      "balance_loss_clip": 1.05366278,
      "balance_loss_mlp": 1.02534747,
      "epoch": 0.30782180003607285,
      "flos": 18405476104320.0,
      "grad_norm": 2.1430254105879483,
      "language_loss": 0.76353508,
      "learning_rate": 3.243670685567586e-06,
      "loss": 0.78559202,
      "num_input_tokens_seen": 55011210,
      "step": 2560,
      "time_per_iteration": 2.7315242290496826
    },
    {
      "auxiliary_loss_clip": 0.01176214,
      "auxiliary_loss_mlp": 0.0103107,
      "balance_loss_clip": 1.05621314,
      "balance_loss_mlp": 1.02059174,
      "epoch": 0.30794204292671196,
      "flos": 23878549365120.0,
      "grad_norm": 2.8885922299453743,
      "language_loss": 0.7986027,
      "learning_rate": 3.2430605411552012e-06,
      "loss": 0.82067549,
      "num_input_tokens_seen": 55031325,
      "step": 2561,
      "time_per_iteration": 2.6995725631713867
    },
    {
      "auxiliary_loss_clip": 0.01100571,
      "auxiliary_loss_mlp": 0.01002063,
      "balance_loss_clip": 1.02700257,
      "balance_loss_mlp": 1.00011945,
      "epoch": 0.30806228581735107,
      "flos": 67927800816000.0,
      "grad_norm": 1.0043477831154586,
      "language_loss": 0.7051512,
      "learning_rate": 3.2424502081715205e-06,
      "loss": 0.72617745,
      "num_input_tokens_seen": 55094440,
      "step": 2562,
      "time_per_iteration": 3.2443604469299316
    },
    {
      "auxiliary_loss_clip": 0.01167075,
      "auxiliary_loss_mlp": 0.00764376,
      "balance_loss_clip": 1.05812061,
      "balance_loss_mlp": 1.00055075,
      "epoch": 0.3081825287079901,
      "flos": 23843213360640.0,
      "grad_norm": 1.7645608963678474,
      "language_loss": 0.7838546,
      "learning_rate": 3.241839686709132e-06,
      "loss": 0.80316907,
      "num_input_tokens_seen": 55115375,
      "step": 2563,
      "time_per_iteration": 2.8008017539978027
    },
    {
      "auxiliary_loss_clip": 0.01169021,
      "auxiliary_loss_mlp": 0.01030189,
      "balance_loss_clip": 1.05808651,
      "balance_loss_mlp": 1.01890016,
      "epoch": 0.30830277159862923,
      "flos": 16209969102720.0,
      "grad_norm": 2.661289929615969,
      "language_loss": 0.82329428,
      "learning_rate": 3.2412289768606495e-06,
      "loss": 0.84528637,
      "num_input_tokens_seen": 55131945,
      "step": 2564,
      "time_per_iteration": 2.7513716220855713
    },
    {
      "auxiliary_loss_clip": 0.01167178,
      "auxiliary_loss_mlp": 0.01030848,
      "balance_loss_clip": 1.05753875,
      "balance_loss_mlp": 1.020298,
      "epoch": 0.30842301448926834,
      "flos": 29349503723520.0,
      "grad_norm": 1.9082044925071757,
      "language_loss": 0.83352244,
      "learning_rate": 3.240618078718718e-06,
      "loss": 0.85550272,
      "num_input_tokens_seen": 55153405,
      "step": 2565,
      "time_per_iteration": 2.7542474269866943
    },
    {
      "auxiliary_loss_clip": 0.01123271,
      "auxiliary_loss_mlp": 0.01039171,
      "balance_loss_clip": 1.05224752,
      "balance_loss_mlp": 1.02837658,
      "epoch": 0.3085432573799074,
      "flos": 21945190798080.0,
      "grad_norm": 2.2870997019427057,
      "language_loss": 0.74458265,
      "learning_rate": 3.240006992376011e-06,
      "loss": 0.7662071,
      "num_input_tokens_seen": 55173030,
      "step": 2566,
      "time_per_iteration": 2.8223795890808105
    },
    {
      "auxiliary_loss_clip": 0.01181956,
      "auxiliary_loss_mlp": 0.01035867,
      "balance_loss_clip": 1.05878973,
      "balance_loss_mlp": 1.02493548,
      "epoch": 0.3086635002705465,
      "flos": 22054718344320.0,
      "grad_norm": 2.642061649185703,
      "language_loss": 0.76325804,
      "learning_rate": 3.2393957179252284e-06,
      "loss": 0.78543627,
      "num_input_tokens_seen": 55189565,
      "step": 2567,
      "time_per_iteration": 2.7035927772521973
    },
    {
      "auxiliary_loss_clip": 0.01183275,
      "auxiliary_loss_mlp": 0.01039272,
      "balance_loss_clip": 1.05957031,
      "balance_loss_mlp": 1.02870965,
      "epoch": 0.3087837431611856,
      "flos": 32665925520000.0,
      "grad_norm": 2.642489946854717,
      "language_loss": 0.80736959,
      "learning_rate": 3.2387842554591016e-06,
      "loss": 0.82959503,
      "num_input_tokens_seen": 55210380,
      "step": 2568,
      "time_per_iteration": 2.7203593254089355
    },
    {
      "auxiliary_loss_clip": 0.01177259,
      "auxiliary_loss_mlp": 0.0104016,
      "balance_loss_clip": 1.06290531,
      "balance_loss_mlp": 1.02857924,
      "epoch": 0.3089039860518247,
      "flos": 17599245384960.0,
      "grad_norm": 2.348165666682147,
      "language_loss": 0.88134837,
      "learning_rate": 3.238172605070388e-06,
      "loss": 0.90352261,
      "num_input_tokens_seen": 55225795,
      "step": 2569,
      "time_per_iteration": 2.6979501247406006
    },
    {
      "auxiliary_loss_clip": 0.01169612,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.05221331,
      "balance_loss_mlp": 1.01847053,
      "epoch": 0.3090242289424638,
      "flos": 14383839611520.0,
      "grad_norm": 2.3383347451297642,
      "language_loss": 0.78569579,
      "learning_rate": 3.2375607668518745e-06,
      "loss": 0.80768657,
      "num_input_tokens_seen": 55238830,
      "step": 2570,
      "time_per_iteration": 2.625988483428955
    },
    {
      "auxiliary_loss_clip": 0.01157846,
      "auxiliary_loss_mlp": 0.01029826,
      "balance_loss_clip": 1.0568862,
      "balance_loss_mlp": 1.01933575,
      "epoch": 0.30914447183310284,
      "flos": 16068625084800.0,
      "grad_norm": 2.1184340354088658,
      "language_loss": 0.90012926,
      "learning_rate": 3.236948740896377e-06,
      "loss": 0.92200595,
      "num_input_tokens_seen": 55253630,
      "step": 2571,
      "time_per_iteration": 3.9964804649353027
    },
    {
      "auxiliary_loss_clip": 0.01134263,
      "auxiliary_loss_mlp": 0.00765297,
      "balance_loss_clip": 1.05170226,
      "balance_loss_mlp": 1.00058413,
      "epoch": 0.30926471472374195,
      "flos": 32230221546240.0,
      "grad_norm": 1.6967799005337703,
      "language_loss": 0.84389853,
      "learning_rate": 3.2363365272967384e-06,
      "loss": 0.86289418,
      "num_input_tokens_seen": 55276200,
      "step": 2572,
      "time_per_iteration": 3.841850996017456
    },
    {
      "auxiliary_loss_clip": 0.01156694,
      "auxiliary_loss_mlp": 0.01036147,
      "balance_loss_clip": 1.05753636,
      "balance_loss_mlp": 1.0256983,
      "epoch": 0.30938495761438106,
      "flos": 20370722970240.0,
      "grad_norm": 2.0200187970550263,
      "language_loss": 0.81612241,
      "learning_rate": 3.235724126145832e-06,
      "loss": 0.83805084,
      "num_input_tokens_seen": 55292235,
      "step": 2573,
      "time_per_iteration": 3.6264469623565674
    },
    {
      "auxiliary_loss_clip": 0.01192295,
      "auxiliary_loss_mlp": 0.00765024,
      "balance_loss_clip": 1.0611825,
      "balance_loss_mlp": 1.00053382,
      "epoch": 0.3095052005050201,
      "flos": 24061155131520.0,
      "grad_norm": 1.6972416181752223,
      "language_loss": 0.77760202,
      "learning_rate": 3.235111537536558e-06,
      "loss": 0.79717523,
      "num_input_tokens_seen": 55313050,
      "step": 2574,
      "time_per_iteration": 2.640920639038086
    },
    {
      "auxiliary_loss_clip": 0.01205571,
      "auxiliary_loss_mlp": 0.00764362,
      "balance_loss_clip": 1.06237376,
      "balance_loss_mlp": 1.00056016,
      "epoch": 0.30962544339565923,
      "flos": 23401547729280.0,
      "grad_norm": 1.843904165987805,
      "language_loss": 0.8253454,
      "learning_rate": 3.2344987615618456e-06,
      "loss": 0.84504479,
      "num_input_tokens_seen": 55332885,
      "step": 2575,
      "time_per_iteration": 3.537795066833496
    },
    {
      "auxiliary_loss_clip": 0.01193601,
      "auxiliary_loss_mlp": 0.01037888,
      "balance_loss_clip": 1.05990124,
      "balance_loss_mlp": 1.02755213,
      "epoch": 0.30974568628629834,
      "flos": 33799984692480.0,
      "grad_norm": 1.699723427188966,
      "language_loss": 0.78382546,
      "learning_rate": 3.2338857983146533e-06,
      "loss": 0.8061403,
      "num_input_tokens_seen": 55354385,
      "step": 2576,
      "time_per_iteration": 2.804016590118408
    },
    {
      "auxiliary_loss_clip": 0.01089814,
      "auxiliary_loss_mlp": 0.01038611,
      "balance_loss_clip": 1.04792953,
      "balance_loss_mlp": 1.02697587,
      "epoch": 0.3098659291769374,
      "flos": 20229594433920.0,
      "grad_norm": 1.8924788038527582,
      "language_loss": 0.76405025,
      "learning_rate": 3.233272647887966e-06,
      "loss": 0.78533453,
      "num_input_tokens_seen": 55373275,
      "step": 2577,
      "time_per_iteration": 2.905064105987549
    },
    {
      "auxiliary_loss_clip": 0.01179652,
      "auxiliary_loss_mlp": 0.00765015,
      "balance_loss_clip": 1.06165409,
      "balance_loss_mlp": 1.00057054,
      "epoch": 0.3099861720675765,
      "flos": 24748556682240.0,
      "grad_norm": 1.8747786776862065,
      "language_loss": 0.89847213,
      "learning_rate": 3.2326593103747985e-06,
      "loss": 0.9179188,
      "num_input_tokens_seen": 55392290,
      "step": 2578,
      "time_per_iteration": 3.0027027130126953
    },
    {
      "auxiliary_loss_clip": 0.01119732,
      "auxiliary_loss_mlp": 0.01037783,
      "balance_loss_clip": 1.05191851,
      "balance_loss_mlp": 1.02733994,
      "epoch": 0.3101064149582156,
      "flos": 11765485704960.0,
      "grad_norm": 2.1063645896452967,
      "language_loss": 0.85232741,
      "learning_rate": 3.2320457858681936e-06,
      "loss": 0.87390256,
      "num_input_tokens_seen": 55410680,
      "step": 2579,
      "time_per_iteration": 2.7442610263824463
    },
    {
      "auxiliary_loss_clip": 0.011109,
      "auxiliary_loss_mlp": 0.00764135,
      "balance_loss_clip": 1.05034304,
      "balance_loss_mlp": 1.00054502,
      "epoch": 0.31022665784885467,
      "flos": 23033247626880.0,
      "grad_norm": 2.7668626439042905,
      "language_loss": 0.84836113,
      "learning_rate": 3.2314320744612228e-06,
      "loss": 0.86711144,
      "num_input_tokens_seen": 55425980,
      "step": 2580,
      "time_per_iteration": 2.990875005722046
    },
    {
      "auxiliary_loss_clip": 0.01134173,
      "auxiliary_loss_mlp": 0.01033724,
      "balance_loss_clip": 1.05153298,
      "balance_loss_mlp": 1.02464032,
      "epoch": 0.3103469007394938,
      "flos": 16289188548480.0,
      "grad_norm": 1.8029036662673819,
      "language_loss": 0.76879132,
      "learning_rate": 3.2308181762469854e-06,
      "loss": 0.79047036,
      "num_input_tokens_seen": 55443925,
      "step": 2581,
      "time_per_iteration": 3.290778875350952
    },
    {
      "auxiliary_loss_clip": 0.01108511,
      "auxiliary_loss_mlp": 0.01038552,
      "balance_loss_clip": 1.05035818,
      "balance_loss_mlp": 1.02604127,
      "epoch": 0.3104671436301329,
      "flos": 30515271626880.0,
      "grad_norm": 3.915698322106178,
      "language_loss": 0.78437811,
      "learning_rate": 3.230204091318609e-06,
      "loss": 0.80584878,
      "num_input_tokens_seen": 55464465,
      "step": 2582,
      "time_per_iteration": 3.137889862060547
    },
    {
      "auxiliary_loss_clip": 0.01066557,
      "auxiliary_loss_mlp": 0.01038843,
      "balance_loss_clip": 1.0468365,
      "balance_loss_mlp": 1.02812064,
      "epoch": 0.31058738652077195,
      "flos": 20047240062720.0,
      "grad_norm": 1.9878565919409317,
      "language_loss": 0.84608889,
      "learning_rate": 3.2295898197692503e-06,
      "loss": 0.86714286,
      "num_input_tokens_seen": 55483425,
      "step": 2583,
      "time_per_iteration": 3.727858781814575
    },
    {
      "auxiliary_loss_clip": 0.01160421,
      "auxiliary_loss_mlp": 0.0103894,
      "balance_loss_clip": 1.05496299,
      "balance_loss_mlp": 1.02853286,
      "epoch": 0.31070762941141106,
      "flos": 28074639237120.0,
      "grad_norm": 1.914840283536886,
      "language_loss": 0.7891838,
      "learning_rate": 3.228975361692094e-06,
      "loss": 0.81117737,
      "num_input_tokens_seen": 55504445,
      "step": 2584,
      "time_per_iteration": 3.316012144088745
    },
    {
      "auxiliary_loss_clip": 0.0116983,
      "auxiliary_loss_mlp": 0.01034522,
      "balance_loss_clip": 1.05855703,
      "balance_loss_mlp": 1.02351272,
      "epoch": 0.31082787230205017,
      "flos": 20521907314560.0,
      "grad_norm": 2.575785973377955,
      "language_loss": 0.80192828,
      "learning_rate": 3.228360717180352e-06,
      "loss": 0.82397187,
      "num_input_tokens_seen": 55521970,
      "step": 2585,
      "time_per_iteration": 2.862917184829712
    },
    {
      "auxiliary_loss_clip": 0.01065135,
      "auxiliary_loss_mlp": 0.01013082,
      "balance_loss_clip": 1.02634645,
      "balance_loss_mlp": 1.01128149,
      "epoch": 0.3109481151926892,
      "flos": 62445928723200.0,
      "grad_norm": 0.8233999959996435,
      "language_loss": 0.59441555,
      "learning_rate": 3.227745886327266e-06,
      "loss": 0.61519778,
      "num_input_tokens_seen": 55580665,
      "step": 2586,
      "time_per_iteration": 3.209357738494873
    },
    {
      "auxiliary_loss_clip": 0.01053349,
      "auxiliary_loss_mlp": 0.00756435,
      "balance_loss_clip": 1.02700639,
      "balance_loss_mlp": 1.00163674,
      "epoch": 0.31106835808332833,
      "flos": 44746744723200.0,
      "grad_norm": 0.8131267534890397,
      "language_loss": 0.55770385,
      "learning_rate": 3.227130869226105e-06,
      "loss": 0.57580173,
      "num_input_tokens_seen": 55637825,
      "step": 2587,
      "time_per_iteration": 3.2991836071014404
    },
    {
      "auxiliary_loss_clip": 0.01183789,
      "auxiliary_loss_mlp": 0.0103708,
      "balance_loss_clip": 1.06103075,
      "balance_loss_mlp": 1.02638125,
      "epoch": 0.3111886009739674,
      "flos": 23403056100480.0,
      "grad_norm": 2.90157277034463,
      "language_loss": 0.83104539,
      "learning_rate": 3.226515665970167e-06,
      "loss": 0.85325408,
      "num_input_tokens_seen": 55655365,
      "step": 2588,
      "time_per_iteration": 2.717397928237915
    },
    {
      "auxiliary_loss_clip": 0.01183973,
      "auxiliary_loss_mlp": 0.01037609,
      "balance_loss_clip": 1.05737019,
      "balance_loss_mlp": 1.02642167,
      "epoch": 0.3113088438646065,
      "flos": 17530728192000.0,
      "grad_norm": 3.610250276394621,
      "language_loss": 0.86381543,
      "learning_rate": 3.225900276652777e-06,
      "loss": 0.88603127,
      "num_input_tokens_seen": 55672140,
      "step": 2589,
      "time_per_iteration": 2.6584718227386475
    },
    {
      "auxiliary_loss_clip": 0.01204506,
      "auxiliary_loss_mlp": 0.00764385,
      "balance_loss_clip": 1.061831,
      "balance_loss_mlp": 1.00060403,
      "epoch": 0.3114290867552456,
      "flos": 28365802882560.0,
      "grad_norm": 1.7378860136210306,
      "language_loss": 0.75799555,
      "learning_rate": 3.2252847013672906e-06,
      "loss": 0.77768445,
      "num_input_tokens_seen": 55694800,
      "step": 2590,
      "time_per_iteration": 2.7717108726501465
    },
    {
      "auxiliary_loss_clip": 0.01146877,
      "auxiliary_loss_mlp": 0.01035643,
      "balance_loss_clip": 1.05305183,
      "balance_loss_mlp": 1.02523553,
      "epoch": 0.31154932964588467,
      "flos": 27379157126400.0,
      "grad_norm": 2.0427361638355523,
      "language_loss": 0.76039213,
      "learning_rate": 3.224668940207089e-06,
      "loss": 0.78221732,
      "num_input_tokens_seen": 55713785,
      "step": 2591,
      "time_per_iteration": 2.89076566696167
    },
    {
      "auxiliary_loss_clip": 0.01195645,
      "auxiliary_loss_mlp": 0.01032077,
      "balance_loss_clip": 1.06065702,
      "balance_loss_mlp": 1.02097249,
      "epoch": 0.3116695725365238,
      "flos": 26541864120960.0,
      "grad_norm": 1.9393030857244107,
      "language_loss": 0.87104917,
      "learning_rate": 3.2240529932655828e-06,
      "loss": 0.89332634,
      "num_input_tokens_seen": 55733050,
      "step": 2592,
      "time_per_iteration": 2.7057905197143555
    },
    {
      "auxiliary_loss_clip": 0.01145511,
      "auxiliary_loss_mlp": 0.01040654,
      "balance_loss_clip": 1.05425966,
      "balance_loss_mlp": 1.02985919,
      "epoch": 0.3117898154271629,
      "flos": 21177600134400.0,
      "grad_norm": 28.479695282284105,
      "language_loss": 0.89012468,
      "learning_rate": 3.223436860636211e-06,
      "loss": 0.91198635,
      "num_input_tokens_seen": 55748685,
      "step": 2593,
      "time_per_iteration": 2.7206900119781494
    },
    {
      "auxiliary_loss_clip": 0.01134236,
      "auxiliary_loss_mlp": 0.01033096,
      "balance_loss_clip": 1.05668092,
      "balance_loss_mlp": 1.02134788,
      "epoch": 0.31191005831780194,
      "flos": 27272430840960.0,
      "grad_norm": 2.2798066593728143,
      "language_loss": 0.74111843,
      "learning_rate": 3.2228205424124403e-06,
      "loss": 0.76279175,
      "num_input_tokens_seen": 55771840,
      "step": 2594,
      "time_per_iteration": 2.894965887069702
    },
    {
      "auxiliary_loss_clip": 0.01143544,
      "auxiliary_loss_mlp": 0.01035876,
      "balance_loss_clip": 1.05583405,
      "balance_loss_mlp": 1.02561212,
      "epoch": 0.31203030120844105,
      "flos": 12963501043200.0,
      "grad_norm": 2.5272656709923966,
      "language_loss": 0.74462527,
      "learning_rate": 3.222204038687765e-06,
      "loss": 0.76641947,
      "num_input_tokens_seen": 55784975,
      "step": 2595,
      "time_per_iteration": 2.7426552772521973
    },
    {
      "auxiliary_loss_clip": 0.01166168,
      "auxiliary_loss_mlp": 0.00764121,
      "balance_loss_clip": 1.05478024,
      "balance_loss_mlp": 1.00056052,
      "epoch": 0.31215054409908016,
      "flos": 27562014288000.0,
      "grad_norm": 1.8113772295123023,
      "language_loss": 0.88016927,
      "learning_rate": 3.221587349555709e-06,
      "loss": 0.89947212,
      "num_input_tokens_seen": 55805235,
      "step": 2596,
      "time_per_iteration": 3.719717025756836
    },
    {
      "auxiliary_loss_clip": 0.01156963,
      "auxiliary_loss_mlp": 0.01030785,
      "balance_loss_clip": 1.05998945,
      "balance_loss_mlp": 1.02161229,
      "epoch": 0.3122707869897192,
      "flos": 21506326427520.0,
      "grad_norm": 1.670807652139229,
      "language_loss": 0.69622648,
      "learning_rate": 3.2209704751098236e-06,
      "loss": 0.718104,
      "num_input_tokens_seen": 55824265,
      "step": 2597,
      "time_per_iteration": 3.736867666244507
    },
    {
      "auxiliary_loss_clip": 0.01205054,
      "auxiliary_loss_mlp": 0.00764751,
      "balance_loss_clip": 1.06215763,
      "balance_loss_mlp": 1.00056529,
      "epoch": 0.31239102988035833,
      "flos": 15187017674880.0,
      "grad_norm": 2.393963375907095,
      "language_loss": 0.83051562,
      "learning_rate": 3.2203534154436875e-06,
      "loss": 0.85021365,
      "num_input_tokens_seen": 55838620,
      "step": 2598,
      "time_per_iteration": 2.6801490783691406
    },
    {
      "auxiliary_loss_clip": 0.01188966,
      "auxiliary_loss_mlp": 0.01033361,
      "balance_loss_clip": 1.0576241,
      "balance_loss_mlp": 1.02320981,
      "epoch": 0.31251127277099744,
      "flos": 22053712763520.0,
      "grad_norm": 4.339044230166458,
      "language_loss": 0.75819463,
      "learning_rate": 3.2197361706509084e-06,
      "loss": 0.78041786,
      "num_input_tokens_seen": 55859375,
      "step": 2599,
      "time_per_iteration": 3.7262985706329346
    },
    {
      "auxiliary_loss_clip": 0.01152629,
      "auxiliary_loss_mlp": 0.01040375,
      "balance_loss_clip": 1.05308139,
      "balance_loss_mlp": 1.02902663,
      "epoch": 0.3126315156616365,
      "flos": 15193984913280.0,
      "grad_norm": 3.129693922591034,
      "language_loss": 0.83812797,
      "learning_rate": 3.2191187408251228e-06,
      "loss": 0.86005801,
      "num_input_tokens_seen": 55876535,
      "step": 2600,
      "time_per_iteration": 2.8152284622192383
    },
    {
      "auxiliary_loss_clip": 0.01102013,
      "auxiliary_loss_mlp": 0.00765306,
      "balance_loss_clip": 1.04537344,
      "balance_loss_mlp": 1.00067472,
      "epoch": 0.3127517585522756,
      "flos": 18145338831360.0,
      "grad_norm": 2.2158223670628336,
      "language_loss": 0.78963506,
      "learning_rate": 3.218501126059993e-06,
      "loss": 0.80830824,
      "num_input_tokens_seen": 55891930,
      "step": 2601,
      "time_per_iteration": 3.9832191467285156
    },
    {
      "auxiliary_loss_clip": 0.01157289,
      "auxiliary_loss_mlp": 0.01045109,
      "balance_loss_clip": 1.05254292,
      "balance_loss_mlp": 1.03439164,
      "epoch": 0.31287200144291466,
      "flos": 21908633731200.0,
      "grad_norm": 1.9760837622947278,
      "language_loss": 0.81416476,
      "learning_rate": 3.2178833264492116e-06,
      "loss": 0.83618867,
      "num_input_tokens_seen": 55910635,
      "step": 2602,
      "time_per_iteration": 2.736358404159546
    },
    {
      "auxiliary_loss_clip": 0.01147632,
      "auxiliary_loss_mlp": 0.01038815,
      "balance_loss_clip": 1.05597496,
      "balance_loss_mlp": 1.02635813,
      "epoch": 0.31299224433355377,
      "flos": 29896997800320.0,
      "grad_norm": 1.7950844541502249,
      "language_loss": 0.76400626,
      "learning_rate": 3.217265342086498e-06,
      "loss": 0.78587079,
      "num_input_tokens_seen": 55931125,
      "step": 2603,
      "time_per_iteration": 2.851270914077759
    },
    {
      "auxiliary_loss_clip": 0.01147158,
      "auxiliary_loss_mlp": 0.01036709,
      "balance_loss_clip": 1.05922413,
      "balance_loss_mlp": 1.02513397,
      "epoch": 0.3131124872241929,
      "flos": 11655886331520.0,
      "grad_norm": 2.229261270388559,
      "language_loss": 0.73333359,
      "learning_rate": 3.216647173065599e-06,
      "loss": 0.75517225,
      "num_input_tokens_seen": 55946590,
      "step": 2604,
      "time_per_iteration": 2.7294626235961914
    },
    {
      "auxiliary_loss_clip": 0.01152203,
      "auxiliary_loss_mlp": 0.01034593,
      "balance_loss_clip": 1.05282903,
      "balance_loss_mlp": 1.02412665,
      "epoch": 0.31323273011483194,
      "flos": 49848785470080.0,
      "grad_norm": 2.160280890464346,
      "language_loss": 0.73589933,
      "learning_rate": 3.216028819480292e-06,
      "loss": 0.75776732,
      "num_input_tokens_seen": 55967930,
      "step": 2605,
      "time_per_iteration": 2.9577367305755615
    },
    {
      "auxiliary_loss_clip": 0.01191249,
      "auxiliary_loss_mlp": 0.01033193,
      "balance_loss_clip": 1.06297135,
      "balance_loss_mlp": 1.022434,
      "epoch": 0.31335297300547105,
      "flos": 22601278667520.0,
      "grad_norm": 1.9786366986551338,
      "language_loss": 0.75646496,
      "learning_rate": 3.2154102814243793e-06,
      "loss": 0.77870935,
      "num_input_tokens_seen": 55987070,
      "step": 2606,
      "time_per_iteration": 2.686218738555908
    },
    {
      "auxiliary_loss_clip": 0.0115407,
      "auxiliary_loss_mlp": 0.00764816,
      "balance_loss_clip": 1.05944276,
      "balance_loss_mlp": 1.00067508,
      "epoch": 0.31347321589611016,
      "flos": 34710858708480.0,
      "grad_norm": 2.1219719270148523,
      "language_loss": 0.66790622,
      "learning_rate": 3.2147915589916937e-06,
      "loss": 0.68709505,
      "num_input_tokens_seen": 56008630,
      "step": 2607,
      "time_per_iteration": 2.8582942485809326
    },
    {
      "auxiliary_loss_clip": 0.01128954,
      "auxiliary_loss_mlp": 0.01038936,
      "balance_loss_clip": 1.05042052,
      "balance_loss_mlp": 1.02718186,
      "epoch": 0.3135934587867492,
      "flos": 19755789108480.0,
      "grad_norm": 1.8650590511643432,
      "language_loss": 0.82649493,
      "learning_rate": 3.2141726522760938e-06,
      "loss": 0.84817386,
      "num_input_tokens_seen": 56026690,
      "step": 2608,
      "time_per_iteration": 2.8083584308624268
    },
    {
      "auxiliary_loss_clip": 0.01059373,
      "auxiliary_loss_mlp": 0.01004856,
      "balance_loss_clip": 1.03211176,
      "balance_loss_mlp": 1.00319898,
      "epoch": 0.3137137016773883,
      "flos": 65815535583360.0,
      "grad_norm": 0.7018167349468984,
      "language_loss": 0.52702087,
      "learning_rate": 3.213553561371469e-06,
      "loss": 0.54766315,
      "num_input_tokens_seen": 56090425,
      "step": 2609,
      "time_per_iteration": 3.431980609893799
    },
    {
      "auxiliary_loss_clip": 0.01158165,
      "auxiliary_loss_mlp": 0.01035163,
      "balance_loss_clip": 1.05700469,
      "balance_loss_mlp": 1.02543533,
      "epoch": 0.31383394456802743,
      "flos": 16252739222400.0,
      "grad_norm": 2.6209811695190717,
      "language_loss": 0.9610914,
      "learning_rate": 3.212934286371733e-06,
      "loss": 0.98302466,
      "num_input_tokens_seen": 56107135,
      "step": 2610,
      "time_per_iteration": 2.697359561920166
    },
    {
      "auxiliary_loss_clip": 0.01117224,
      "auxiliary_loss_mlp": 0.0103968,
      "balance_loss_clip": 1.05064106,
      "balance_loss_mlp": 1.02857018,
      "epoch": 0.3139541874586665,
      "flos": 38795517613440.0,
      "grad_norm": 2.513078910669693,
      "language_loss": 0.83225518,
      "learning_rate": 3.2123148273708304e-06,
      "loss": 0.85382426,
      "num_input_tokens_seen": 56127325,
      "step": 2611,
      "time_per_iteration": 2.9515345096588135
    },
    {
      "auxiliary_loss_clip": 0.01131527,
      "auxiliary_loss_mlp": 0.01033914,
      "balance_loss_clip": 1.05365527,
      "balance_loss_mlp": 1.02361417,
      "epoch": 0.3140744303493056,
      "flos": 25046328430080.0,
      "grad_norm": 1.7710063081137248,
      "language_loss": 0.76660395,
      "learning_rate": 3.211695184462733e-06,
      "loss": 0.78825831,
      "num_input_tokens_seen": 56148500,
      "step": 2612,
      "time_per_iteration": 2.8122265338897705
    },
    {
      "auxiliary_loss_clip": 0.0106475,
      "auxiliary_loss_mlp": 0.010046,
      "balance_loss_clip": 1.04045582,
      "balance_loss_mlp": 1.00284767,
      "epoch": 0.3141946732399447,
      "flos": 72504254782080.0,
      "grad_norm": 0.8856967847789533,
      "language_loss": 0.60420346,
      "learning_rate": 3.2110753577414383e-06,
      "loss": 0.62489688,
      "num_input_tokens_seen": 56210080,
      "step": 2613,
      "time_per_iteration": 3.2766077518463135
    },
    {
      "auxiliary_loss_clip": 0.01205692,
      "auxiliary_loss_mlp": 0.00765142,
      "balance_loss_clip": 1.06187391,
      "balance_loss_mlp": 1.00078344,
      "epoch": 0.31431491613058377,
      "flos": 19239788280960.0,
      "grad_norm": 2.1754403192605394,
      "language_loss": 0.78757226,
      "learning_rate": 3.2104553473009757e-06,
      "loss": 0.8072806,
      "num_input_tokens_seen": 56228200,
      "step": 2614,
      "time_per_iteration": 2.6153109073638916
    },
    {
      "auxiliary_loss_clip": 0.01150584,
      "auxiliary_loss_mlp": 0.01035972,
      "balance_loss_clip": 1.05614424,
      "balance_loss_mlp": 1.02540398,
      "epoch": 0.3144351590212229,
      "flos": 36210596290560.0,
      "grad_norm": 1.7596435918739297,
      "language_loss": 0.67531085,
      "learning_rate": 3.209835153235399e-06,
      "loss": 0.6971764,
      "num_input_tokens_seen": 56249755,
      "step": 2615,
      "time_per_iteration": 2.95959210395813
    },
    {
      "auxiliary_loss_clip": 0.0114586,
      "auxiliary_loss_mlp": 0.01034607,
      "balance_loss_clip": 1.05279624,
      "balance_loss_mlp": 1.0244565,
      "epoch": 0.314555401911862,
      "flos": 18551740285440.0,
      "grad_norm": 1.8142942581872281,
      "language_loss": 0.67476237,
      "learning_rate": 3.2092147756387916e-06,
      "loss": 0.69656694,
      "num_input_tokens_seen": 56270080,
      "step": 2616,
      "time_per_iteration": 2.8962175846099854
    },
    {
      "auxiliary_loss_clip": 0.01181307,
      "auxiliary_loss_mlp": 0.01032883,
      "balance_loss_clip": 1.05643404,
      "balance_loss_mlp": 1.02158201,
      "epoch": 0.31467564480250104,
      "flos": 16362877299840.0,
      "grad_norm": 2.8408804350712584,
      "language_loss": 0.83199537,
      "learning_rate": 3.208594214605264e-06,
      "loss": 0.8541373,
      "num_input_tokens_seen": 56288625,
      "step": 2617,
      "time_per_iteration": 2.6973114013671875
    },
    {
      "auxiliary_loss_clip": 0.01116677,
      "auxiliary_loss_mlp": 0.01032841,
      "balance_loss_clip": 1.04999137,
      "balance_loss_mlp": 1.02276158,
      "epoch": 0.31479588769314015,
      "flos": 21652375127040.0,
      "grad_norm": 3.1048415072322384,
      "language_loss": 0.76957685,
      "learning_rate": 3.2079734702289553e-06,
      "loss": 0.79107201,
      "num_input_tokens_seen": 56307520,
      "step": 2618,
      "time_per_iteration": 2.9257256984710693
    },
    {
      "auxiliary_loss_clip": 0.01029282,
      "auxiliary_loss_mlp": 0.01001817,
      "balance_loss_clip": 1.02046168,
      "balance_loss_mlp": 1.00001717,
      "epoch": 0.3149161305837792,
      "flos": 66051072040320.0,
      "grad_norm": 0.8079838880455475,
      "language_loss": 0.6046235,
      "learning_rate": 3.207352542604031e-06,
      "loss": 0.62493449,
      "num_input_tokens_seen": 56369855,
      "step": 2619,
      "time_per_iteration": 3.5614724159240723
    },
    {
      "auxiliary_loss_clip": 0.01154358,
      "auxiliary_loss_mlp": 0.01035243,
      "balance_loss_clip": 1.05227613,
      "balance_loss_mlp": 1.0255754,
      "epoch": 0.3150363734744183,
      "flos": 28987201192320.0,
      "grad_norm": 1.644441781720434,
      "language_loss": 0.7857976,
      "learning_rate": 3.2067314318246864e-06,
      "loss": 0.8076936,
      "num_input_tokens_seen": 56390570,
      "step": 2620,
      "time_per_iteration": 3.21479868888855
    },
    {
      "auxiliary_loss_clip": 0.01149248,
      "auxiliary_loss_mlp": 0.01036235,
      "balance_loss_clip": 1.05728674,
      "balance_loss_mlp": 1.02602482,
      "epoch": 0.31515661636505743,
      "flos": 27636600879360.0,
      "grad_norm": 2.1221931607813267,
      "language_loss": 0.77995688,
      "learning_rate": 3.206110137985143e-06,
      "loss": 0.8018117,
      "num_input_tokens_seen": 56410775,
      "step": 2621,
      "time_per_iteration": 2.7997689247131348
    },
    {
      "auxiliary_loss_clip": 0.01141923,
      "auxiliary_loss_mlp": 0.01035602,
      "balance_loss_clip": 1.053087,
      "balance_loss_mlp": 1.02475953,
      "epoch": 0.3152768592556965,
      "flos": 24605632465920.0,
      "grad_norm": 2.7828704284986894,
      "language_loss": 0.92238426,
      "learning_rate": 3.2054886611796505e-06,
      "loss": 0.94415951,
      "num_input_tokens_seen": 56429770,
      "step": 2622,
      "time_per_iteration": 3.7489826679229736
    },
    {
      "auxiliary_loss_clip": 0.0109619,
      "auxiliary_loss_mlp": 0.01002221,
      "balance_loss_clip": 1.02330041,
      "balance_loss_mlp": 1.00072503,
      "epoch": 0.3153971021463356,
      "flos": 68476908026880.0,
      "grad_norm": 0.8919053788458435,
      "language_loss": 0.6355176,
      "learning_rate": 3.204867001502487e-06,
      "loss": 0.65650177,
      "num_input_tokens_seen": 56488425,
      "step": 2623,
      "time_per_iteration": 3.1708567142486572
    },
    {
      "auxiliary_loss_clip": 0.01130071,
      "auxiliary_loss_mlp": 0.01037808,
      "balance_loss_clip": 1.05488276,
      "balance_loss_mlp": 1.02719271,
      "epoch": 0.3155173450369747,
      "flos": 25593714766080.0,
      "grad_norm": 2.3211912947461038,
      "language_loss": 0.80884469,
      "learning_rate": 3.2042451590479567e-06,
      "loss": 0.83052343,
      "num_input_tokens_seen": 56508940,
      "step": 2624,
      "time_per_iteration": 4.66258430480957
    },
    {
      "auxiliary_loss_clip": 0.01099045,
      "auxiliary_loss_mlp": 0.01038759,
      "balance_loss_clip": 1.04689634,
      "balance_loss_mlp": 1.02845359,
      "epoch": 0.31563758792761376,
      "flos": 24309333175680.0,
      "grad_norm": 1.8036386629467265,
      "language_loss": 0.87206721,
      "learning_rate": 3.203623133910394e-06,
      "loss": 0.89344525,
      "num_input_tokens_seen": 56527245,
      "step": 2625,
      "time_per_iteration": 2.8865816593170166
    },
    {
      "auxiliary_loss_clip": 0.01169261,
      "auxiliary_loss_mlp": 0.01032119,
      "balance_loss_clip": 1.05731499,
      "balance_loss_mlp": 1.02228475,
      "epoch": 0.31575783081825287,
      "flos": 31903865550720.0,
      "grad_norm": 2.3673230916034305,
      "language_loss": 0.77262688,
      "learning_rate": 3.203000926184158e-06,
      "loss": 0.79464066,
      "num_input_tokens_seen": 56546170,
      "step": 2626,
      "time_per_iteration": 3.741044759750366
    },
    {
      "auxiliary_loss_clip": 0.01149322,
      "auxiliary_loss_mlp": 0.01030341,
      "balance_loss_clip": 1.05737185,
      "balance_loss_mlp": 1.02108443,
      "epoch": 0.315878073708892,
      "flos": 30810960385920.0,
      "grad_norm": 1.7749992963555488,
      "language_loss": 0.77834821,
      "learning_rate": 3.202378535963639e-06,
      "loss": 0.80014479,
      "num_input_tokens_seen": 56567085,
      "step": 2627,
      "time_per_iteration": 2.857830762863159
    },
    {
      "auxiliary_loss_clip": 0.01183114,
      "auxiliary_loss_mlp": 0.01037428,
      "balance_loss_clip": 1.06192589,
      "balance_loss_mlp": 1.02636492,
      "epoch": 0.31599831659953104,
      "flos": 22200264253440.0,
      "grad_norm": 1.674063839780192,
      "language_loss": 0.83752334,
      "learning_rate": 3.2017559633432516e-06,
      "loss": 0.85972881,
      "num_input_tokens_seen": 56586715,
      "step": 2628,
      "time_per_iteration": 2.7953591346740723
    },
    {
      "auxiliary_loss_clip": 0.01152026,
      "auxiliary_loss_mlp": 0.01032584,
      "balance_loss_clip": 1.05728066,
      "balance_loss_mlp": 1.02197397,
      "epoch": 0.31611855949017015,
      "flos": 25593463370880.0,
      "grad_norm": 1.976110895066697,
      "language_loss": 0.66290975,
      "learning_rate": 3.2011332084174398e-06,
      "loss": 0.6847558,
      "num_input_tokens_seen": 56607585,
      "step": 2629,
      "time_per_iteration": 2.86749267578125
    },
    {
      "auxiliary_loss_clip": 0.01187761,
      "auxiliary_loss_mlp": 0.01040036,
      "balance_loss_clip": 1.05483162,
      "balance_loss_mlp": 1.02932477,
      "epoch": 0.31623880238080926,
      "flos": 20594087694720.0,
      "grad_norm": 1.5806634222827325,
      "language_loss": 0.89101303,
      "learning_rate": 3.2005102712806756e-06,
      "loss": 0.91329098,
      "num_input_tokens_seen": 56626415,
      "step": 2630,
      "time_per_iteration": 2.6879448890686035
    },
    {
      "auxiliary_loss_clip": 0.01183459,
      "auxiliary_loss_mlp": 0.01038679,
      "balance_loss_clip": 1.0603441,
      "balance_loss_mlp": 1.02818835,
      "epoch": 0.3163590452714483,
      "flos": 12784917600000.0,
      "grad_norm": 2.3414976776423737,
      "language_loss": 0.73108268,
      "learning_rate": 3.1998871520274575e-06,
      "loss": 0.75330412,
      "num_input_tokens_seen": 56641750,
      "step": 2631,
      "time_per_iteration": 3.015660524368286
    },
    {
      "auxiliary_loss_clip": 0.01183761,
      "auxiliary_loss_mlp": 0.0103331,
      "balance_loss_clip": 1.05693316,
      "balance_loss_mlp": 1.02278984,
      "epoch": 0.3164792881620874,
      "flos": 23041292273280.0,
      "grad_norm": 1.8020399574771817,
      "language_loss": 0.84736037,
      "learning_rate": 3.199263850752312e-06,
      "loss": 0.86953104,
      "num_input_tokens_seen": 56662585,
      "step": 2632,
      "time_per_iteration": 2.6885972023010254
    },
    {
      "auxiliary_loss_clip": 0.01159977,
      "auxiliary_loss_mlp": 0.0103315,
      "balance_loss_clip": 1.05624497,
      "balance_loss_mlp": 1.02240944,
      "epoch": 0.31659953105272653,
      "flos": 18296271780480.0,
      "grad_norm": 2.037802227102436,
      "language_loss": 0.85752344,
      "learning_rate": 3.198640367549795e-06,
      "loss": 0.87945461,
      "num_input_tokens_seen": 56681480,
      "step": 2633,
      "time_per_iteration": 2.7385380268096924
    },
    {
      "auxiliary_loss_clip": 0.01136297,
      "auxiliary_loss_mlp": 0.01034715,
      "balance_loss_clip": 1.05134487,
      "balance_loss_mlp": 1.0250715,
      "epoch": 0.3167197739433656,
      "flos": 25703421880320.0,
      "grad_norm": 1.753448886336337,
      "language_loss": 0.85483211,
      "learning_rate": 3.198016702514487e-06,
      "loss": 0.87654221,
      "num_input_tokens_seen": 56701760,
      "step": 2634,
      "time_per_iteration": 2.8864128589630127
    },
    {
      "auxiliary_loss_clip": 0.01170476,
      "auxiliary_loss_mlp": 0.01030083,
      "balance_loss_clip": 1.05781221,
      "balance_loss_mlp": 1.02036774,
      "epoch": 0.3168400168340047,
      "flos": 23546016230400.0,
      "grad_norm": 1.6304079418645696,
      "language_loss": 0.84452778,
      "learning_rate": 3.1973928557409972e-06,
      "loss": 0.86653334,
      "num_input_tokens_seen": 56719800,
      "step": 2635,
      "time_per_iteration": 2.6880154609680176
    },
    {
      "auxiliary_loss_clip": 0.01154919,
      "auxiliary_loss_mlp": 0.01036453,
      "balance_loss_clip": 1.05432057,
      "balance_loss_mlp": 1.0268507,
      "epoch": 0.31696025972464376,
      "flos": 28366449327360.0,
      "grad_norm": 1.7697317975400118,
      "language_loss": 0.71458519,
      "learning_rate": 3.1967688273239636e-06,
      "loss": 0.73649883,
      "num_input_tokens_seen": 56739605,
      "step": 2636,
      "time_per_iteration": 2.798687219619751
    },
    {
      "auxiliary_loss_clip": 0.01144507,
      "auxiliary_loss_mlp": 0.0103372,
      "balance_loss_clip": 1.05144978,
      "balance_loss_mlp": 1.02339625,
      "epoch": 0.31708050261528287,
      "flos": 16399111144320.0,
      "grad_norm": 1.9309235319898805,
      "language_loss": 0.82278323,
      "learning_rate": 3.1961446173580503e-06,
      "loss": 0.84456551,
      "num_input_tokens_seen": 56756545,
      "step": 2637,
      "time_per_iteration": 2.800623655319214
    },
    {
      "auxiliary_loss_clip": 0.01126988,
      "auxiliary_loss_mlp": 0.01036753,
      "balance_loss_clip": 1.05120468,
      "balance_loss_mlp": 1.02628636,
      "epoch": 0.317200745505922,
      "flos": 26212347728640.0,
      "grad_norm": 1.8284552414224862,
      "language_loss": 0.77518821,
      "learning_rate": 3.1955202259379502e-06,
      "loss": 0.79682553,
      "num_input_tokens_seen": 56778275,
      "step": 2638,
      "time_per_iteration": 2.8392934799194336
    },
    {
      "auxiliary_loss_clip": 0.01115801,
      "auxiliary_loss_mlp": 0.01039205,
      "balance_loss_clip": 1.04855883,
      "balance_loss_mlp": 1.02827978,
      "epoch": 0.31732098839656103,
      "flos": 31350876693120.0,
      "grad_norm": 1.604950105609178,
      "language_loss": 0.83107495,
      "learning_rate": 3.194895653158381e-06,
      "loss": 0.85262501,
      "num_input_tokens_seen": 56797215,
      "step": 2639,
      "time_per_iteration": 2.878350257873535
    },
    {
      "auxiliary_loss_clip": 0.01068735,
      "auxiliary_loss_mlp": 0.01001062,
      "balance_loss_clip": 1.01966178,
      "balance_loss_mlp": 0.99952406,
      "epoch": 0.31744123128720014,
      "flos": 58989024835200.0,
      "grad_norm": 0.7647285348698556,
      "language_loss": 0.55522299,
      "learning_rate": 3.194270899114093e-06,
      "loss": 0.57592094,
      "num_input_tokens_seen": 56863010,
      "step": 2640,
      "time_per_iteration": 3.3480262756347656
    },
    {
      "auxiliary_loss_clip": 0.01177818,
      "auxiliary_loss_mlp": 0.00765185,
      "balance_loss_clip": 1.06198764,
      "balance_loss_mlp": 1.00084186,
      "epoch": 0.31756147417783925,
      "flos": 17417573372160.0,
      "grad_norm": 5.075816171307593,
      "language_loss": 0.82124954,
      "learning_rate": 3.193645963899858e-06,
      "loss": 0.84067965,
      "num_input_tokens_seen": 56880625,
      "step": 2641,
      "time_per_iteration": 2.749232053756714
    },
    {
      "auxiliary_loss_clip": 0.01130968,
      "auxiliary_loss_mlp": 0.01037138,
      "balance_loss_clip": 1.04944623,
      "balance_loss_mlp": 1.02658796,
      "epoch": 0.3176817170684783,
      "flos": 25481673267840.0,
      "grad_norm": 1.9251226896975802,
      "language_loss": 0.83762687,
      "learning_rate": 3.193020847610479e-06,
      "loss": 0.859308,
      "num_input_tokens_seen": 56900945,
      "step": 2642,
      "time_per_iteration": 2.8600611686706543
    },
    {
      "auxiliary_loss_clip": 0.01189636,
      "auxiliary_loss_mlp": 0.01037627,
      "balance_loss_clip": 1.06221962,
      "balance_loss_mlp": 1.02631974,
      "epoch": 0.3178019599591174,
      "flos": 24972603765120.0,
      "grad_norm": 2.3531815973076915,
      "language_loss": 0.70966947,
      "learning_rate": 3.192395550340787e-06,
      "loss": 0.73194218,
      "num_input_tokens_seen": 56918895,
      "step": 2643,
      "time_per_iteration": 2.6912248134613037
    },
    {
      "auxiliary_loss_clip": 0.01166525,
      "auxiliary_loss_mlp": 0.01035395,
      "balance_loss_clip": 1.05846715,
      "balance_loss_mlp": 1.02495277,
      "epoch": 0.31792220284975653,
      "flos": 12422220019200.0,
      "grad_norm": 1.7792400296360367,
      "language_loss": 0.76868808,
      "learning_rate": 3.191770072185638e-06,
      "loss": 0.79070723,
      "num_input_tokens_seen": 56935890,
      "step": 2644,
      "time_per_iteration": 2.6892318725585938
    },
    {
      "auxiliary_loss_clip": 0.01126997,
      "auxiliary_loss_mlp": 0.01032601,
      "balance_loss_clip": 1.04720068,
      "balance_loss_mlp": 1.02230752,
      "epoch": 0.3180424457403956,
      "flos": 15485759089920.0,
      "grad_norm": 3.1809181711122654,
      "language_loss": 0.72284889,
      "learning_rate": 3.191144413239916e-06,
      "loss": 0.74444479,
      "num_input_tokens_seen": 56952460,
      "step": 2645,
      "time_per_iteration": 2.8057258129119873
    },
    {
      "auxiliary_loss_clip": 0.01160054,
      "auxiliary_loss_mlp": 0.01031732,
      "balance_loss_clip": 1.05667114,
      "balance_loss_mlp": 1.02139664,
      "epoch": 0.3181626886310347,
      "flos": 26174964648960.0,
      "grad_norm": 2.383722218501749,
      "language_loss": 0.8818022,
      "learning_rate": 3.190518573598534e-06,
      "loss": 0.90372002,
      "num_input_tokens_seen": 56969065,
      "step": 2646,
      "time_per_iteration": 2.7128806114196777
    },
    {
      "auxiliary_loss_clip": 0.01176145,
      "auxiliary_loss_mlp": 0.01032893,
      "balance_loss_clip": 1.06022739,
      "balance_loss_mlp": 1.02209294,
      "epoch": 0.3182829315216738,
      "flos": 25483109811840.0,
      "grad_norm": 1.4281539167644035,
      "language_loss": 0.77492678,
      "learning_rate": 3.1898925533564308e-06,
      "loss": 0.79701722,
      "num_input_tokens_seen": 56990535,
      "step": 2647,
      "time_per_iteration": 2.770921230316162
    },
    {
      "auxiliary_loss_clip": 0.01192749,
      "auxiliary_loss_mlp": 0.01031468,
      "balance_loss_clip": 1.06094813,
      "balance_loss_mlp": 1.02100778,
      "epoch": 0.31840317441231286,
      "flos": 18113701927680.0,
      "grad_norm": 2.0056309930458904,
      "language_loss": 0.63894832,
      "learning_rate": 3.1892663526085733e-06,
      "loss": 0.66119051,
      "num_input_tokens_seen": 57008910,
      "step": 2648,
      "time_per_iteration": 3.6106903553009033
    },
    {
      "auxiliary_loss_clip": 0.0101515,
      "auxiliary_loss_mlp": 0.01007251,
      "balance_loss_clip": 1.01057863,
      "balance_loss_mlp": 1.00575471,
      "epoch": 0.31852341730295197,
      "flos": 64741948957440.0,
      "grad_norm": 0.7469214985192997,
      "language_loss": 0.56922656,
      "learning_rate": 3.188639971449956e-06,
      "loss": 0.58945054,
      "num_input_tokens_seen": 57074960,
      "step": 2649,
      "time_per_iteration": 3.3044304847717285
    },
    {
      "auxiliary_loss_clip": 0.01098391,
      "auxiliary_loss_mlp": 0.01034875,
      "balance_loss_clip": 1.04589629,
      "balance_loss_mlp": 1.02385437,
      "epoch": 0.318643660193591,
      "flos": 20668135582080.0,
      "grad_norm": 1.920207420625281,
      "language_loss": 0.72616339,
      "learning_rate": 3.1880134099756e-06,
      "loss": 0.74749607,
      "num_input_tokens_seen": 57094595,
      "step": 2650,
      "time_per_iteration": 3.798211097717285
    },
    {
      "auxiliary_loss_clip": 0.01137476,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.0539254,
      "balance_loss_mlp": 1.02159286,
      "epoch": 0.31876390308423014,
      "flos": 26943345411840.0,
      "grad_norm": 1.80017901784936,
      "language_loss": 0.70246953,
      "learning_rate": 3.1873866682805535e-06,
      "loss": 0.72417331,
      "num_input_tokens_seen": 57115290,
      "step": 2651,
      "time_per_iteration": 2.9509170055389404
    },
    {
      "auxiliary_loss_clip": 0.01135404,
      "auxiliary_loss_mlp": 0.01031413,
      "balance_loss_clip": 1.05363059,
      "balance_loss_mlp": 1.02119637,
      "epoch": 0.31888414597486925,
      "flos": 18041916597120.0,
      "grad_norm": 1.819218005486724,
      "language_loss": 0.88721895,
      "learning_rate": 3.186759746459894e-06,
      "loss": 0.90888715,
      "num_input_tokens_seen": 57134400,
      "step": 2652,
      "time_per_iteration": 3.6154284477233887
    },
    {
      "auxiliary_loss_clip": 0.01097612,
      "auxiliary_loss_mlp": 0.01030131,
      "balance_loss_clip": 1.05057871,
      "balance_loss_mlp": 1.01984954,
      "epoch": 0.3190043888655083,
      "flos": 25149319701120.0,
      "grad_norm": 2.0021805376373907,
      "language_loss": 0.79639041,
      "learning_rate": 3.1861326446087246e-06,
      "loss": 0.81766784,
      "num_input_tokens_seen": 57153140,
      "step": 2653,
      "time_per_iteration": 2.9717071056365967
    },
    {
      "auxiliary_loss_clip": 0.01139491,
      "auxiliary_loss_mlp": 0.00764559,
      "balance_loss_clip": 1.04862571,
      "balance_loss_mlp": 1.00076413,
      "epoch": 0.3191246317561474,
      "flos": 22053892331520.0,
      "grad_norm": 2.712050180158281,
      "language_loss": 0.72102326,
      "learning_rate": 3.1855053628221763e-06,
      "loss": 0.74006379,
      "num_input_tokens_seen": 57172395,
      "step": 2654,
      "time_per_iteration": 3.199678421020508
    },
    {
      "auxiliary_loss_clip": 0.01157657,
      "auxiliary_loss_mlp": 0.01033078,
      "balance_loss_clip": 1.05484891,
      "balance_loss_mlp": 1.02221179,
      "epoch": 0.3192448746467865,
      "flos": 14901815687040.0,
      "grad_norm": 2.9478939922267764,
      "language_loss": 0.899149,
      "learning_rate": 3.184877901195407e-06,
      "loss": 0.92105633,
      "num_input_tokens_seen": 57189090,
      "step": 2655,
      "time_per_iteration": 2.7161316871643066
    },
    {
      "auxiliary_loss_clip": 0.01082189,
      "auxiliary_loss_mlp": 0.01001386,
      "balance_loss_clip": 1.01975989,
      "balance_loss_mlp": 0.99986631,
      "epoch": 0.3193651175374256,
      "flos": 67234832657280.0,
      "grad_norm": 0.796084929001165,
      "language_loss": 0.62846261,
      "learning_rate": 3.184250259823602e-06,
      "loss": 0.64929837,
      "num_input_tokens_seen": 57251620,
      "step": 2656,
      "time_per_iteration": 3.314995288848877
    },
    {
      "auxiliary_loss_clip": 0.01163166,
      "auxiliary_loss_mlp": 0.01040314,
      "balance_loss_clip": 1.05650425,
      "balance_loss_mlp": 1.02941239,
      "epoch": 0.3194853604280647,
      "flos": 12233077977600.0,
      "grad_norm": 2.387992381235843,
      "language_loss": 0.81620145,
      "learning_rate": 3.183622438801974e-06,
      "loss": 0.83823621,
      "num_input_tokens_seen": 57266910,
      "step": 2657,
      "time_per_iteration": 2.692497491836548
    },
    {
      "auxiliary_loss_clip": 0.01139384,
      "auxiliary_loss_mlp": 0.00764458,
      "balance_loss_clip": 1.05528903,
      "balance_loss_mlp": 1.00070274,
      "epoch": 0.3196056033187038,
      "flos": 14939917038720.0,
      "grad_norm": 1.8024595696948589,
      "language_loss": 0.74870276,
      "learning_rate": 3.1829944382257637e-06,
      "loss": 0.76774114,
      "num_input_tokens_seen": 57285040,
      "step": 2658,
      "time_per_iteration": 2.7808139324188232
    },
    {
      "auxiliary_loss_clip": 0.01184351,
      "auxiliary_loss_mlp": 0.01035019,
      "balance_loss_clip": 1.05577064,
      "balance_loss_mlp": 1.02468348,
      "epoch": 0.31972584620934286,
      "flos": 23768878164480.0,
      "grad_norm": 2.166689987313028,
      "language_loss": 0.81184906,
      "learning_rate": 3.1823662581902373e-06,
      "loss": 0.83404279,
      "num_input_tokens_seen": 57302725,
      "step": 2659,
      "time_per_iteration": 2.6610963344573975
    },
    {
      "auxiliary_loss_clip": 0.01160048,
      "auxiliary_loss_mlp": 0.01031786,
      "balance_loss_clip": 1.05302083,
      "balance_loss_mlp": 1.02168298,
      "epoch": 0.31984608909998197,
      "flos": 21251540280960.0,
      "grad_norm": 2.055505605602787,
      "language_loss": 0.7463001,
      "learning_rate": 3.1817378987906896e-06,
      "loss": 0.76821846,
      "num_input_tokens_seen": 57322230,
      "step": 2660,
      "time_per_iteration": 2.7241907119750977
    },
    {
      "auxiliary_loss_clip": 0.01197156,
      "auxiliary_loss_mlp": 0.01033474,
      "balance_loss_clip": 1.06334233,
      "balance_loss_mlp": 1.0226438,
      "epoch": 0.3199663319906211,
      "flos": 18296235866880.0,
      "grad_norm": 1.9184711741953577,
      "language_loss": 0.80093718,
      "learning_rate": 3.181109360122442e-06,
      "loss": 0.82324344,
      "num_input_tokens_seen": 57339820,
      "step": 2661,
      "time_per_iteration": 2.6465511322021484
    },
    {
      "auxiliary_loss_clip": 0.01188258,
      "auxiliary_loss_mlp": 0.01034331,
      "balance_loss_clip": 1.06093264,
      "balance_loss_mlp": 1.02369165,
      "epoch": 0.32008657488126013,
      "flos": 18733627779840.0,
      "grad_norm": 2.942137869886852,
      "language_loss": 0.78487551,
      "learning_rate": 3.1804806422808445e-06,
      "loss": 0.80710149,
      "num_input_tokens_seen": 57356955,
      "step": 2662,
      "time_per_iteration": 2.6202805042266846
    },
    {
      "auxiliary_loss_clip": 0.01150951,
      "auxiliary_loss_mlp": 0.0104248,
      "balance_loss_clip": 1.0575757,
      "balance_loss_mlp": 1.03185892,
      "epoch": 0.32020681777189924,
      "flos": 20595344670720.0,
      "grad_norm": 2.0089761810689417,
      "language_loss": 0.73261964,
      "learning_rate": 3.1798517453612714e-06,
      "loss": 0.75455397,
      "num_input_tokens_seen": 57376760,
      "step": 2663,
      "time_per_iteration": 2.7821621894836426
    },
    {
      "auxiliary_loss_clip": 0.01143297,
      "auxiliary_loss_mlp": 0.01034774,
      "balance_loss_clip": 1.055161,
      "balance_loss_mlp": 1.02501035,
      "epoch": 0.32032706066253835,
      "flos": 35261692750080.0,
      "grad_norm": 15.916654875930082,
      "language_loss": 0.75232011,
      "learning_rate": 3.1792226694591265e-06,
      "loss": 0.77410084,
      "num_input_tokens_seen": 57398145,
      "step": 2664,
      "time_per_iteration": 2.8622398376464844
    },
    {
      "auxiliary_loss_clip": 0.01153028,
      "auxiliary_loss_mlp": 0.01030682,
      "balance_loss_clip": 1.05365586,
      "balance_loss_mlp": 1.020329,
      "epoch": 0.3204473035531774,
      "flos": 15304230731520.0,
      "grad_norm": 2.049975379475934,
      "language_loss": 0.80934703,
      "learning_rate": 3.178593414669841e-06,
      "loss": 0.83118415,
      "num_input_tokens_seen": 57416730,
      "step": 2665,
      "time_per_iteration": 2.6959972381591797
    },
    {
      "auxiliary_loss_clip": 0.01141419,
      "auxiliary_loss_mlp": 0.01041848,
      "balance_loss_clip": 1.04976439,
      "balance_loss_mlp": 1.03017759,
      "epoch": 0.3205675464438165,
      "flos": 24462564595200.0,
      "grad_norm": 2.022009316204648,
      "language_loss": 0.71112943,
      "learning_rate": 3.1779639810888707e-06,
      "loss": 0.73296207,
      "num_input_tokens_seen": 57436325,
      "step": 2666,
      "time_per_iteration": 2.728616237640381
    },
    {
      "auxiliary_loss_clip": 0.01156103,
      "auxiliary_loss_mlp": 0.01034191,
      "balance_loss_clip": 1.05659223,
      "balance_loss_mlp": 1.02411175,
      "epoch": 0.3206877893344556,
      "flos": 22456235548800.0,
      "grad_norm": 1.7795639584974283,
      "language_loss": 0.75951409,
      "learning_rate": 3.1773343688117013e-06,
      "loss": 0.78141707,
      "num_input_tokens_seen": 57457235,
      "step": 2667,
      "time_per_iteration": 2.7543845176696777
    },
    {
      "auxiliary_loss_clip": 0.01158813,
      "auxiliary_loss_mlp": 0.01034908,
      "balance_loss_clip": 1.05941057,
      "balance_loss_mlp": 1.02450156,
      "epoch": 0.3208080322250947,
      "flos": 20412236113920.0,
      "grad_norm": 2.5309388828642554,
      "language_loss": 0.8416459,
      "learning_rate": 3.1767045779338445e-06,
      "loss": 0.86358309,
      "num_input_tokens_seen": 57474895,
      "step": 2668,
      "time_per_iteration": 2.7291274070739746
    },
    {
      "auxiliary_loss_clip": 0.0117473,
      "auxiliary_loss_mlp": 0.01034825,
      "balance_loss_clip": 1.05825913,
      "balance_loss_mlp": 1.02468586,
      "epoch": 0.3209282751157338,
      "flos": 21762118154880.0,
      "grad_norm": 2.3495694691276747,
      "language_loss": 0.91482675,
      "learning_rate": 3.176074608550839e-06,
      "loss": 0.93692225,
      "num_input_tokens_seen": 57490715,
      "step": 2669,
      "time_per_iteration": 2.7087655067443848
    },
    {
      "auxiliary_loss_clip": 0.01179613,
      "auxiliary_loss_mlp": 0.01033827,
      "balance_loss_clip": 1.06290889,
      "balance_loss_mlp": 1.02405739,
      "epoch": 0.32104851800637285,
      "flos": 22055041566720.0,
      "grad_norm": 2.4049766077081367,
      "language_loss": 0.82417363,
      "learning_rate": 3.17544446075825e-06,
      "loss": 0.84630799,
      "num_input_tokens_seen": 57509880,
      "step": 2670,
      "time_per_iteration": 2.7029829025268555
    },
    {
      "auxiliary_loss_clip": 0.01171891,
      "auxiliary_loss_mlp": 0.01036292,
      "balance_loss_clip": 1.06099606,
      "balance_loss_mlp": 1.02682102,
      "epoch": 0.32116876089701196,
      "flos": 37012301896320.0,
      "grad_norm": 3.0064652901935185,
      "language_loss": 0.71062744,
      "learning_rate": 3.174814134651671e-06,
      "loss": 0.73270923,
      "num_input_tokens_seen": 57532430,
      "step": 2671,
      "time_per_iteration": 2.8488006591796875
    },
    {
      "auxiliary_loss_clip": 0.01104931,
      "auxiliary_loss_mlp": 0.01039258,
      "balance_loss_clip": 1.04756224,
      "balance_loss_mlp": 1.0289228,
      "epoch": 0.3212890037876511,
      "flos": 21979233912960.0,
      "grad_norm": 1.7704968527864358,
      "language_loss": 0.80203235,
      "learning_rate": 3.1741836303267215e-06,
      "loss": 0.82347429,
      "num_input_tokens_seen": 57551965,
      "step": 2672,
      "time_per_iteration": 2.878730058670044
    },
    {
      "auxiliary_loss_clip": 0.01118423,
      "auxiliary_loss_mlp": 0.00764006,
      "balance_loss_clip": 1.04770243,
      "balance_loss_mlp": 1.0007925,
      "epoch": 0.32140924667829013,
      "flos": 10342345875840.0,
      "grad_norm": 2.1689456009017944,
      "language_loss": 0.75219619,
      "learning_rate": 3.1735529478790496e-06,
      "loss": 0.77102047,
      "num_input_tokens_seen": 57569955,
      "step": 2673,
      "time_per_iteration": 3.7264208793640137
    },
    {
      "auxiliary_loss_clip": 0.01123833,
      "auxiliary_loss_mlp": 0.01035861,
      "balance_loss_clip": 1.04968846,
      "balance_loss_mlp": 1.02544177,
      "epoch": 0.32152948956892924,
      "flos": 50798910072960.0,
      "grad_norm": 1.77700641087299,
      "language_loss": 0.79212916,
      "learning_rate": 3.172922087404328e-06,
      "loss": 0.81372607,
      "num_input_tokens_seen": 57592215,
      "step": 2674,
      "time_per_iteration": 3.0380051136016846
    },
    {
      "auxiliary_loss_clip": 0.01057201,
      "auxiliary_loss_mlp": 0.0100299,
      "balance_loss_clip": 1.03230059,
      "balance_loss_mlp": 1.00141656,
      "epoch": 0.32164973245956835,
      "flos": 63863250549120.0,
      "grad_norm": 0.7660290681330119,
      "language_loss": 0.55221444,
      "learning_rate": 3.1722910489982586e-06,
      "loss": 0.57281637,
      "num_input_tokens_seen": 57652575,
      "step": 2675,
      "time_per_iteration": 5.52875828742981
    },
    {
      "auxiliary_loss_clip": 0.0118078,
      "auxiliary_loss_mlp": 0.01034117,
      "balance_loss_clip": 1.05782747,
      "balance_loss_mlp": 1.0243125,
      "epoch": 0.3217699753502074,
      "flos": 23513948363520.0,
      "grad_norm": 1.4672964069589565,
      "language_loss": 0.8005538,
      "learning_rate": 3.1716598327565694e-06,
      "loss": 0.82270277,
      "num_input_tokens_seen": 57672215,
      "step": 2676,
      "time_per_iteration": 2.7324140071868896
    },
    {
      "auxiliary_loss_clip": 0.01168355,
      "auxiliary_loss_mlp": 0.01034356,
      "balance_loss_clip": 1.05844116,
      "balance_loss_mlp": 1.02511704,
      "epoch": 0.3218902182408465,
      "flos": 19062533640960.0,
      "grad_norm": 1.4683499833321199,
      "language_loss": 0.83991235,
      "learning_rate": 3.171028438775015e-06,
      "loss": 0.86193943,
      "num_input_tokens_seen": 57691410,
      "step": 2677,
      "time_per_iteration": 3.830518960952759
    },
    {
      "auxiliary_loss_clip": 0.01200028,
      "auxiliary_loss_mlp": 0.01030852,
      "balance_loss_clip": 1.05896497,
      "balance_loss_mlp": 1.02150047,
      "epoch": 0.3220104611314856,
      "flos": 20375571306240.0,
      "grad_norm": 2.3322000207536493,
      "language_loss": 0.8391872,
      "learning_rate": 3.170396867149377e-06,
      "loss": 0.86149597,
      "num_input_tokens_seen": 57709415,
      "step": 2678,
      "time_per_iteration": 2.7006685733795166
    },
    {
      "auxiliary_loss_clip": 0.0116298,
      "auxiliary_loss_mlp": 0.0103521,
      "balance_loss_clip": 1.05620968,
      "balance_loss_mlp": 1.02542281,
      "epoch": 0.3221307040221247,
      "flos": 20117014231680.0,
      "grad_norm": 1.7663984192074103,
      "language_loss": 0.85842097,
      "learning_rate": 3.1697651179754653e-06,
      "loss": 0.88040286,
      "num_input_tokens_seen": 57728075,
      "step": 2679,
      "time_per_iteration": 2.7167065143585205
    },
    {
      "auxiliary_loss_clip": 0.01105213,
      "auxiliary_loss_mlp": 0.0076359,
      "balance_loss_clip": 1.0513854,
      "balance_loss_mlp": 1.00070512,
      "epoch": 0.3222509469127638,
      "flos": 23987789602560.0,
      "grad_norm": 2.0230678627963097,
      "language_loss": 0.73070705,
      "learning_rate": 3.1691331913491153e-06,
      "loss": 0.74939507,
      "num_input_tokens_seen": 57750645,
      "step": 2680,
      "time_per_iteration": 3.0187275409698486
    },
    {
      "auxiliary_loss_clip": 0.01190448,
      "auxiliary_loss_mlp": 0.0103762,
      "balance_loss_clip": 1.05898702,
      "balance_loss_mlp": 1.02850676,
      "epoch": 0.32237118980340285,
      "flos": 17675735397120.0,
      "grad_norm": 2.0026544122247514,
      "language_loss": 0.85075784,
      "learning_rate": 3.1685010873661898e-06,
      "loss": 0.87303853,
      "num_input_tokens_seen": 57769820,
      "step": 2681,
      "time_per_iteration": 2.8999247550964355
    },
    {
      "auxiliary_loss_clip": 0.01131389,
      "auxiliary_loss_mlp": 0.0103179,
      "balance_loss_clip": 1.05333722,
      "balance_loss_mlp": 1.02170491,
      "epoch": 0.32249143269404196,
      "flos": 23147982645120.0,
      "grad_norm": 2.618135054100831,
      "language_loss": 0.79548907,
      "learning_rate": 3.167868806122578e-06,
      "loss": 0.81712091,
      "num_input_tokens_seen": 57788870,
      "step": 2682,
      "time_per_iteration": 2.7531211376190186
    },
    {
      "auxiliary_loss_clip": 0.01174285,
      "auxiliary_loss_mlp": 0.01038858,
      "balance_loss_clip": 1.06109452,
      "balance_loss_mlp": 1.02865314,
      "epoch": 0.32261167558468107,
      "flos": 24422308427520.0,
      "grad_norm": 2.1255982632349175,
      "language_loss": 0.66451544,
      "learning_rate": 3.1672363477141968e-06,
      "loss": 0.68664688,
      "num_input_tokens_seen": 57808165,
      "step": 2683,
      "time_per_iteration": 0.03385138511657715
    },
    {
      "auxiliary_loss_clip": 0.01059508,
      "auxiliary_loss_mlp": 0.01036665,
      "balance_loss_clip": 1.03966594,
      "balance_loss_mlp": 1.02625847,
      "epoch": 0.3227319184753201,
      "flos": 30367175852160.0,
      "grad_norm": 1.7896558820404247,
      "language_loss": 0.85184127,
      "learning_rate": 3.1666037122369903e-06,
      "loss": 0.87280303,
      "num_input_tokens_seen": 57828825,
      "step": 2684,
      "time_per_iteration": 3.1017324924468994
    },
    {
      "auxiliary_loss_clip": 0.01181197,
      "auxiliary_loss_mlp": 0.01035997,
      "balance_loss_clip": 1.05595136,
      "balance_loss_mlp": 1.02597189,
      "epoch": 0.32285216136595923,
      "flos": 16946174257920.0,
      "grad_norm": 1.9645843162176555,
      "language_loss": 0.86796045,
      "learning_rate": 3.165970899786928e-06,
      "loss": 0.89013237,
      "num_input_tokens_seen": 57846740,
      "step": 2685,
      "time_per_iteration": 3.478851795196533
    },
    {
      "auxiliary_loss_clip": 0.0117882,
      "auxiliary_loss_mlp": 0.01031439,
      "balance_loss_clip": 1.05938482,
      "balance_loss_mlp": 1.02109218,
      "epoch": 0.32297240425659834,
      "flos": 21981532383360.0,
      "grad_norm": 2.3125977383204,
      "language_loss": 0.75032282,
      "learning_rate": 3.1653379104600067e-06,
      "loss": 0.77242541,
      "num_input_tokens_seen": 57866885,
      "step": 2686,
      "time_per_iteration": 2.774773359298706
    },
    {
      "auxiliary_loss_clip": 0.01144812,
      "auxiliary_loss_mlp": 0.01035249,
      "balance_loss_clip": 1.05518365,
      "balance_loss_mlp": 1.02454448,
      "epoch": 0.3230926471472374,
      "flos": 22748045639040.0,
      "grad_norm": 1.8672702015126401,
      "language_loss": 0.6930545,
      "learning_rate": 3.164704744352251e-06,
      "loss": 0.71485513,
      "num_input_tokens_seen": 57887690,
      "step": 2687,
      "time_per_iteration": 2.736060857772827
    },
    {
      "auxiliary_loss_clip": 0.01183088,
      "auxiliary_loss_mlp": 0.0103629,
      "balance_loss_clip": 1.05759478,
      "balance_loss_mlp": 1.02691984,
      "epoch": 0.3232128900378765,
      "flos": 16942977947520.0,
      "grad_norm": 1.8506155095833632,
      "language_loss": 0.80454189,
      "learning_rate": 3.164071401559713e-06,
      "loss": 0.82673562,
      "num_input_tokens_seen": 57905090,
      "step": 2688,
      "time_per_iteration": 2.669076442718506
    },
    {
      "auxiliary_loss_clip": 0.01151562,
      "auxiliary_loss_mlp": 0.01041655,
      "balance_loss_clip": 1.05323601,
      "balance_loss_mlp": 1.03125358,
      "epoch": 0.3233331329285156,
      "flos": 24023736138240.0,
      "grad_norm": 1.7227704568151005,
      "language_loss": 0.7109552,
      "learning_rate": 3.1634378821784674e-06,
      "loss": 0.73288739,
      "num_input_tokens_seen": 57925305,
      "step": 2689,
      "time_per_iteration": 2.7148468494415283
    },
    {
      "auxiliary_loss_clip": 0.01121886,
      "auxiliary_loss_mlp": 0.01036744,
      "balance_loss_clip": 1.04899168,
      "balance_loss_mlp": 1.02690887,
      "epoch": 0.3234533758191547,
      "flos": 18113845582080.0,
      "grad_norm": 2.193989091362534,
      "language_loss": 0.73850554,
      "learning_rate": 3.1628041863046208e-06,
      "loss": 0.76009184,
      "num_input_tokens_seen": 57942720,
      "step": 2690,
      "time_per_iteration": 2.7436776161193848
    },
    {
      "auxiliary_loss_clip": 0.01138287,
      "auxiliary_loss_mlp": 0.01038177,
      "balance_loss_clip": 1.05469871,
      "balance_loss_mlp": 1.0276984,
      "epoch": 0.3235736187097938,
      "flos": 16946138344320.0,
      "grad_norm": 2.1469963771200855,
      "language_loss": 0.91256571,
      "learning_rate": 3.162170314034304e-06,
      "loss": 0.93433028,
      "num_input_tokens_seen": 57960135,
      "step": 2691,
      "time_per_iteration": 2.724989652633667
    },
    {
      "auxiliary_loss_clip": 0.01124093,
      "auxiliary_loss_mlp": 0.01036239,
      "balance_loss_clip": 1.04731596,
      "balance_loss_mlp": 1.02552843,
      "epoch": 0.3236938616004329,
      "flos": 22127150119680.0,
      "grad_norm": 1.770747741391693,
      "language_loss": 0.80871463,
      "learning_rate": 3.1615362654636738e-06,
      "loss": 0.83031797,
      "num_input_tokens_seen": 57980875,
      "step": 2692,
      "time_per_iteration": 2.7853405475616455
    },
    {
      "auxiliary_loss_clip": 0.01190695,
      "auxiliary_loss_mlp": 0.01034925,
      "balance_loss_clip": 1.06172323,
      "balance_loss_mlp": 1.02622867,
      "epoch": 0.32381410449107195,
      "flos": 17164618819200.0,
      "grad_norm": 1.7836030498985593,
      "language_loss": 0.86946583,
      "learning_rate": 3.1609020406889163e-06,
      "loss": 0.89172208,
      "num_input_tokens_seen": 57998310,
      "step": 2693,
      "time_per_iteration": 2.634040355682373
    },
    {
      "auxiliary_loss_clip": 0.01134377,
      "auxiliary_loss_mlp": 0.01036918,
      "balance_loss_clip": 1.05179846,
      "balance_loss_mlp": 1.02793562,
      "epoch": 0.32393434738171106,
      "flos": 16578125550720.0,
      "grad_norm": 1.6594664282988438,
      "language_loss": 0.85079247,
      "learning_rate": 3.1602676398062416e-06,
      "loss": 0.87250543,
      "num_input_tokens_seen": 58017220,
      "step": 2694,
      "time_per_iteration": 2.7382123470306396
    },
    {
      "auxiliary_loss_clip": 0.01158222,
      "auxiliary_loss_mlp": 0.01034836,
      "balance_loss_clip": 1.05376697,
      "balance_loss_mlp": 1.02427411,
      "epoch": 0.3240545902723502,
      "flos": 25483612602240.0,
      "grad_norm": 2.425877664030455,
      "language_loss": 0.61685777,
      "learning_rate": 3.1596330629118886e-06,
      "loss": 0.63878834,
      "num_input_tokens_seen": 58037190,
      "step": 2695,
      "time_per_iteration": 2.7325398921966553
    },
    {
      "auxiliary_loss_clip": 0.01173647,
      "auxiliary_loss_mlp": 0.01034831,
      "balance_loss_clip": 1.05727017,
      "balance_loss_mlp": 1.02428699,
      "epoch": 0.32417483316298923,
      "flos": 35845851634560.0,
      "grad_norm": 3.542430969603266,
      "language_loss": 0.73377502,
      "learning_rate": 3.1589983101021223e-06,
      "loss": 0.75585973,
      "num_input_tokens_seen": 58055820,
      "step": 2696,
      "time_per_iteration": 2.816843271255493
    },
    {
      "auxiliary_loss_clip": 0.01162787,
      "auxiliary_loss_mlp": 0.01032646,
      "balance_loss_clip": 1.05254078,
      "balance_loss_mlp": 1.02213752,
      "epoch": 0.32429507605362834,
      "flos": 30080501406720.0,
      "grad_norm": 2.3243278846074484,
      "language_loss": 0.8443476,
      "learning_rate": 3.1583633814732337e-06,
      "loss": 0.86630189,
      "num_input_tokens_seen": 58075340,
      "step": 2697,
      "time_per_iteration": 2.738895893096924
    },
    {
      "auxiliary_loss_clip": 0.01157587,
      "auxiliary_loss_mlp": 0.01038741,
      "balance_loss_clip": 1.05392694,
      "balance_loss_mlp": 1.02925801,
      "epoch": 0.3244153189442674,
      "flos": 18223265387520.0,
      "grad_norm": 3.2396812411000466,
      "language_loss": 0.7148,
      "learning_rate": 3.157728277121541e-06,
      "loss": 0.7367633,
      "num_input_tokens_seen": 58093515,
      "step": 2698,
      "time_per_iteration": 2.6925761699676514
    },
    {
      "auxiliary_loss_clip": 0.01166227,
      "auxiliary_loss_mlp": 0.0103256,
      "balance_loss_clip": 1.0513761,
      "balance_loss_mlp": 1.02152729,
      "epoch": 0.3245355618349065,
      "flos": 17710317216000.0,
      "grad_norm": 2.600166513627574,
      "language_loss": 0.78327,
      "learning_rate": 3.1570929971433897e-06,
      "loss": 0.80525792,
      "num_input_tokens_seen": 58109300,
      "step": 2699,
      "time_per_iteration": 3.7759251594543457
    },
    {
      "auxiliary_loss_clip": 0.01194149,
      "auxiliary_loss_mlp": 0.01039665,
      "balance_loss_clip": 1.06445253,
      "balance_loss_mlp": 1.02974701,
      "epoch": 0.3246558047255456,
      "flos": 23440798316160.0,
      "grad_norm": 2.119658095703375,
      "language_loss": 0.83835375,
      "learning_rate": 3.1564575416351504e-06,
      "loss": 0.86069196,
      "num_input_tokens_seen": 58128000,
      "step": 2700,
      "time_per_iteration": 2.72519850730896
    },
    {
      "auxiliary_loss_clip": 0.01187303,
      "auxiliary_loss_mlp": 0.01037377,
      "balance_loss_clip": 1.06126571,
      "balance_loss_mlp": 1.02723837,
      "epoch": 0.32477604761618467,
      "flos": 21760861178880.0,
      "grad_norm": 1.9620876677540982,
      "language_loss": 0.73509538,
      "learning_rate": 3.155821910693221e-06,
      "loss": 0.75734216,
      "num_input_tokens_seen": 58147415,
      "step": 2701,
      "time_per_iteration": 2.66987681388855
    },
    {
      "auxiliary_loss_clip": 0.01171944,
      "auxiliary_loss_mlp": 0.01032186,
      "balance_loss_clip": 1.05736279,
      "balance_loss_mlp": 1.02253008,
      "epoch": 0.3248962905068238,
      "flos": 19828328624640.0,
      "grad_norm": 1.6930915378889333,
      "language_loss": 0.85888797,
      "learning_rate": 3.1551861044140275e-06,
      "loss": 0.88092923,
      "num_input_tokens_seen": 58167050,
      "step": 2702,
      "time_per_iteration": 5.462082147598267
    },
    {
      "auxiliary_loss_clip": 0.01153881,
      "auxiliary_loss_mlp": 0.01034699,
      "balance_loss_clip": 1.05875909,
      "balance_loss_mlp": 1.02453005,
      "epoch": 0.3250165333974629,
      "flos": 23948215793280.0,
      "grad_norm": 1.6837111488112508,
      "language_loss": 0.77488005,
      "learning_rate": 3.15455012289402e-06,
      "loss": 0.7967658,
      "num_input_tokens_seen": 58186695,
      "step": 2703,
      "time_per_iteration": 2.7482242584228516
    },
    {
      "auxiliary_loss_clip": 0.01189232,
      "auxiliary_loss_mlp": 0.01034478,
      "balance_loss_clip": 1.06308079,
      "balance_loss_mlp": 1.02390409,
      "epoch": 0.32513677628810195,
      "flos": 23989333887360.0,
      "grad_norm": 1.8243738938560636,
      "language_loss": 0.84128374,
      "learning_rate": 3.153913966229677e-06,
      "loss": 0.8635208,
      "num_input_tokens_seen": 58205815,
      "step": 2704,
      "time_per_iteration": 2.6687099933624268
    },
    {
      "auxiliary_loss_clip": 0.01083182,
      "auxiliary_loss_mlp": 0.01002551,
      "balance_loss_clip": 1.0285182,
      "balance_loss_mlp": 1.00088167,
      "epoch": 0.32525701917874106,
      "flos": 70655790009600.0,
      "grad_norm": 0.6540812740139549,
      "language_loss": 0.50300997,
      "learning_rate": 3.1532776345175027e-06,
      "loss": 0.52386731,
      "num_input_tokens_seen": 58270960,
      "step": 2705,
      "time_per_iteration": 3.2908310890197754
    },
    {
      "auxiliary_loss_clip": 0.01156249,
      "auxiliary_loss_mlp": 0.01030389,
      "balance_loss_clip": 1.0545547,
      "balance_loss_mlp": 1.0210669,
      "epoch": 0.32537726206938017,
      "flos": 19682639061120.0,
      "grad_norm": 2.1942419146962298,
      "language_loss": 0.78734583,
      "learning_rate": 3.1526411278540285e-06,
      "loss": 0.80921221,
      "num_input_tokens_seen": 58289390,
      "step": 2706,
      "time_per_iteration": 2.6745450496673584
    },
    {
      "auxiliary_loss_clip": 0.01189632,
      "auxiliary_loss_mlp": 0.01027341,
      "balance_loss_clip": 1.05786836,
      "balance_loss_mlp": 1.01767898,
      "epoch": 0.3254975049600192,
      "flos": 28760999293440.0,
      "grad_norm": 2.311095732971862,
      "language_loss": 0.81176817,
      "learning_rate": 3.1520044463358116e-06,
      "loss": 0.83393788,
      "num_input_tokens_seen": 58306120,
      "step": 2707,
      "time_per_iteration": 2.7011075019836426
    },
    {
      "auxiliary_loss_clip": 0.01122377,
      "auxiliary_loss_mlp": 0.00763657,
      "balance_loss_clip": 1.0481019,
      "balance_loss_mlp": 1.00080729,
      "epoch": 0.32561774785065833,
      "flos": 18877378008960.0,
      "grad_norm": 1.5328649040158253,
      "language_loss": 0.80227757,
      "learning_rate": 3.151367590059436e-06,
      "loss": 0.82113796,
      "num_input_tokens_seen": 58324545,
      "step": 2708,
      "time_per_iteration": 2.7073171138763428
    },
    {
      "auxiliary_loss_clip": 0.01172859,
      "auxiliary_loss_mlp": 0.01037194,
      "balance_loss_clip": 1.0604229,
      "balance_loss_mlp": 1.02688277,
      "epoch": 0.32573799074129745,
      "flos": 23112107936640.0,
      "grad_norm": 1.9786925179735553,
      "language_loss": 0.86531657,
      "learning_rate": 3.1507305591215117e-06,
      "loss": 0.88741714,
      "num_input_tokens_seen": 58342455,
      "step": 2709,
      "time_per_iteration": 2.7059824466705322
    },
    {
      "auxiliary_loss_clip": 0.01073783,
      "auxiliary_loss_mlp": 0.0075601,
      "balance_loss_clip": 1.0253495,
      "balance_loss_mlp": 1.00204778,
      "epoch": 0.3258582336319365,
      "flos": 71237650423680.0,
      "grad_norm": 0.9495456431479737,
      "language_loss": 0.55727649,
      "learning_rate": 3.150093353618677e-06,
      "loss": 0.5755744,
      "num_input_tokens_seen": 58407185,
      "step": 2710,
      "time_per_iteration": 3.3152828216552734
    },
    {
      "auxiliary_loss_clip": 0.01163568,
      "auxiliary_loss_mlp": 0.01036051,
      "balance_loss_clip": 1.05638766,
      "balance_loss_mlp": 1.02604926,
      "epoch": 0.3259784765225756,
      "flos": 22456020067200.0,
      "grad_norm": 2.2873325201685697,
      "language_loss": 0.88114768,
      "learning_rate": 3.149455973647596e-06,
      "loss": 0.90314382,
      "num_input_tokens_seen": 58425245,
      "step": 2711,
      "time_per_iteration": 2.7281270027160645
    },
    {
      "auxiliary_loss_clip": 0.01163559,
      "auxiliary_loss_mlp": 0.01031527,
      "balance_loss_clip": 1.0547806,
      "balance_loss_mlp": 1.02068496,
      "epoch": 0.32609871941321467,
      "flos": 20484811543680.0,
      "grad_norm": 1.93578574533033,
      "language_loss": 0.77160376,
      "learning_rate": 3.1488184193049563e-06,
      "loss": 0.79355466,
      "num_input_tokens_seen": 58444780,
      "step": 2712,
      "time_per_iteration": 2.739626884460449
    },
    {
      "auxiliary_loss_clip": 0.01189216,
      "auxiliary_loss_mlp": 0.01030749,
      "balance_loss_clip": 1.05949712,
      "balance_loss_mlp": 1.02176619,
      "epoch": 0.3262189623038538,
      "flos": 22416805393920.0,
      "grad_norm": 1.5653888498013038,
      "language_loss": 0.71987462,
      "learning_rate": 3.1481806906874767e-06,
      "loss": 0.74207419,
      "num_input_tokens_seen": 58466090,
      "step": 2713,
      "time_per_iteration": 2.7085092067718506
    },
    {
      "auxiliary_loss_clip": 0.01141189,
      "auxiliary_loss_mlp": 0.0103592,
      "balance_loss_clip": 1.05416405,
      "balance_loss_mlp": 1.0262996,
      "epoch": 0.3263392051944929,
      "flos": 20923496346240.0,
      "grad_norm": 2.0460807638423137,
      "language_loss": 0.87643129,
      "learning_rate": 3.147542787891899e-06,
      "loss": 0.89820242,
      "num_input_tokens_seen": 58485435,
      "step": 2714,
      "time_per_iteration": 2.712116003036499
    },
    {
      "auxiliary_loss_clip": 0.01126399,
      "auxiliary_loss_mlp": 0.01029233,
      "balance_loss_clip": 1.05230951,
      "balance_loss_mlp": 1.01957726,
      "epoch": 0.32645944808513194,
      "flos": 24025172682240.0,
      "grad_norm": 1.8927450847329983,
      "language_loss": 0.75152916,
      "learning_rate": 3.1469047110149926e-06,
      "loss": 0.77308547,
      "num_input_tokens_seen": 58504175,
      "step": 2715,
      "time_per_iteration": 2.8121836185455322
    },
    {
      "auxiliary_loss_clip": 0.01165499,
      "auxiliary_loss_mlp": 0.01034372,
      "balance_loss_clip": 1.05375731,
      "balance_loss_mlp": 1.02480567,
      "epoch": 0.32657969097577105,
      "flos": 21032413361280.0,
      "grad_norm": 1.9442474157577678,
      "language_loss": 0.85417449,
      "learning_rate": 3.146266460153554e-06,
      "loss": 0.8761732,
      "num_input_tokens_seen": 58523885,
      "step": 2716,
      "time_per_iteration": 2.769740581512451
    },
    {
      "auxiliary_loss_clip": 0.01123867,
      "auxiliary_loss_mlp": 0.00763853,
      "balance_loss_clip": 1.04776359,
      "balance_loss_mlp": 1.00065565,
      "epoch": 0.32669993386641016,
      "flos": 22710267509760.0,
      "grad_norm": 1.7213239012493176,
      "language_loss": 0.79925418,
      "learning_rate": 3.145628035404404e-06,
      "loss": 0.81813133,
      "num_input_tokens_seen": 58543085,
      "step": 2717,
      "time_per_iteration": 2.744945764541626
    },
    {
      "auxiliary_loss_clip": 0.01058258,
      "auxiliary_loss_mlp": 0.00755674,
      "balance_loss_clip": 1.02286911,
      "balance_loss_mlp": 1.00198746,
      "epoch": 0.3268201767570492,
      "flos": 72105718406400.0,
      "grad_norm": 0.8777350800247313,
      "language_loss": 0.57509482,
      "learning_rate": 3.1449894368643922e-06,
      "loss": 0.59323412,
      "num_input_tokens_seen": 58605400,
      "step": 2718,
      "time_per_iteration": 3.387526512145996
    },
    {
      "auxiliary_loss_clip": 0.01152991,
      "auxiliary_loss_mlp": 0.01029467,
      "balance_loss_clip": 1.05790043,
      "balance_loss_mlp": 1.02067578,
      "epoch": 0.32694041964768833,
      "flos": 24535175938560.0,
      "grad_norm": 1.4608667063282827,
      "language_loss": 0.71372825,
      "learning_rate": 3.1443506646303934e-06,
      "loss": 0.73555285,
      "num_input_tokens_seen": 58626700,
      "step": 2719,
      "time_per_iteration": 2.749171257019043
    },
    {
      "auxiliary_loss_clip": 0.01201487,
      "auxiliary_loss_mlp": 0.01031964,
      "balance_loss_clip": 1.06356263,
      "balance_loss_mlp": 1.02136683,
      "epoch": 0.32706066253832744,
      "flos": 33183003755520.0,
      "grad_norm": 2.0625890748120286,
      "language_loss": 0.67113936,
      "learning_rate": 3.1437117187993086e-06,
      "loss": 0.69347382,
      "num_input_tokens_seen": 58649020,
      "step": 2720,
      "time_per_iteration": 2.700389862060547
    },
    {
      "auxiliary_loss_clip": 0.01137397,
      "auxiliary_loss_mlp": 0.01034222,
      "balance_loss_clip": 1.05077493,
      "balance_loss_mlp": 1.02462566,
      "epoch": 0.3271809054289665,
      "flos": 24061622008320.0,
      "grad_norm": 1.5988473780181667,
      "language_loss": 0.79766983,
      "learning_rate": 3.143072599468065e-06,
      "loss": 0.81938601,
      "num_input_tokens_seen": 58668845,
      "step": 2721,
      "time_per_iteration": 2.832221269607544
    },
    {
      "auxiliary_loss_clip": 0.0112104,
      "auxiliary_loss_mlp": 0.01032715,
      "balance_loss_clip": 1.05576336,
      "balance_loss_mlp": 1.02370882,
      "epoch": 0.3273011483196056,
      "flos": 38253769712640.0,
      "grad_norm": 1.4441488496875816,
      "language_loss": 0.75761437,
      "learning_rate": 3.1424333067336174e-06,
      "loss": 0.77915192,
      "num_input_tokens_seen": 58691610,
      "step": 2722,
      "time_per_iteration": 2.885172128677368
    },
    {
      "auxiliary_loss_clip": 0.011791,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 1.05630076,
      "balance_loss_mlp": 1.01983666,
      "epoch": 0.3274213912102447,
      "flos": 29054389582080.0,
      "grad_norm": 1.7127675221231498,
      "language_loss": 0.78326368,
      "learning_rate": 3.141793840692945e-06,
      "loss": 0.80535322,
      "num_input_tokens_seen": 58712360,
      "step": 2723,
      "time_per_iteration": 2.769120931625366
    },
    {
      "auxiliary_loss_clip": 0.01203114,
      "auxiliary_loss_mlp": 0.01035722,
      "balance_loss_clip": 1.06178296,
      "balance_loss_mlp": 1.02564847,
      "epoch": 0.32754163410088377,
      "flos": 29133249891840.0,
      "grad_norm": 2.556681286985324,
      "language_loss": 0.60993195,
      "learning_rate": 3.1411542014430553e-06,
      "loss": 0.63232034,
      "num_input_tokens_seen": 58733440,
      "step": 2724,
      "time_per_iteration": 2.7009663581848145
    },
    {
      "auxiliary_loss_clip": 0.01158952,
      "auxiliary_loss_mlp": 0.01031966,
      "balance_loss_clip": 1.05389249,
      "balance_loss_mlp": 1.0231204,
      "epoch": 0.3276618769915229,
      "flos": 20631075724800.0,
      "grad_norm": 1.771498363854686,
      "language_loss": 0.81791633,
      "learning_rate": 3.1405143890809804e-06,
      "loss": 0.83982551,
      "num_input_tokens_seen": 58752735,
      "step": 2725,
      "time_per_iteration": 2.7265841960906982
    },
    {
      "auxiliary_loss_clip": 0.0113875,
      "auxiliary_loss_mlp": 0.01034808,
      "balance_loss_clip": 1.05413818,
      "balance_loss_mlp": 1.02552736,
      "epoch": 0.327782119882162,
      "flos": 18657425076480.0,
      "grad_norm": 2.0727223819204688,
      "language_loss": 0.7024442,
      "learning_rate": 3.1398744037037796e-06,
      "loss": 0.7241798,
      "num_input_tokens_seen": 58772070,
      "step": 2726,
      "time_per_iteration": 3.7133214473724365
    },
    {
      "auxiliary_loss_clip": 0.01187324,
      "auxiliary_loss_mlp": 0.01033541,
      "balance_loss_clip": 1.06175816,
      "balance_loss_mlp": 1.02469015,
      "epoch": 0.32790236277280105,
      "flos": 21795802133760.0,
      "grad_norm": 2.085447708115295,
      "language_loss": 0.83767861,
      "learning_rate": 3.139234245408538e-06,
      "loss": 0.85988724,
      "num_input_tokens_seen": 58790950,
      "step": 2727,
      "time_per_iteration": 4.5357348918914795
    },
    {
      "auxiliary_loss_clip": 0.01147207,
      "auxiliary_loss_mlp": 0.0103008,
      "balance_loss_clip": 1.05553341,
      "balance_loss_mlp": 1.0208894,
      "epoch": 0.32802260566344016,
      "flos": 23331414424320.0,
      "grad_norm": 1.3818413724671001,
      "language_loss": 0.76070035,
      "learning_rate": 3.1385939142923666e-06,
      "loss": 0.78247327,
      "num_input_tokens_seen": 58813340,
      "step": 2728,
      "time_per_iteration": 3.6608357429504395
    },
    {
      "auxiliary_loss_clip": 0.01171611,
      "auxiliary_loss_mlp": 0.01030159,
      "balance_loss_clip": 1.0590167,
      "balance_loss_mlp": 1.0201391,
      "epoch": 0.3281428485540792,
      "flos": 24206988349440.0,
      "grad_norm": 2.1952685342882075,
      "language_loss": 0.78428042,
      "learning_rate": 3.137953410452405e-06,
      "loss": 0.80629814,
      "num_input_tokens_seen": 58833610,
      "step": 2729,
      "time_per_iteration": 2.88061785697937
    },
    {
      "auxiliary_loss_clip": 0.0108581,
      "auxiliary_loss_mlp": 0.01034302,
      "balance_loss_clip": 1.04611492,
      "balance_loss_mlp": 1.02453327,
      "epoch": 0.3282630914447183,
      "flos": 34128962380800.0,
      "grad_norm": 1.5942769465364246,
      "language_loss": 0.74540782,
      "learning_rate": 3.1373127339858146e-06,
      "loss": 0.76660895,
      "num_input_tokens_seen": 58856210,
      "step": 2730,
      "time_per_iteration": 3.290015697479248
    },
    {
      "auxiliary_loss_clip": 0.01201977,
      "auxiliary_loss_mlp": 0.00762787,
      "balance_loss_clip": 1.06395674,
      "balance_loss_mlp": 1.00069654,
      "epoch": 0.32838333433535744,
      "flos": 27600726170880.0,
      "grad_norm": 1.8300071688566606,
      "language_loss": 0.74777377,
      "learning_rate": 3.136671884989787e-06,
      "loss": 0.76742148,
      "num_input_tokens_seen": 58876120,
      "step": 2731,
      "time_per_iteration": 2.892380475997925
    },
    {
      "auxiliary_loss_clip": 0.01154318,
      "auxiliary_loss_mlp": 0.01037864,
      "balance_loss_clip": 1.05805254,
      "balance_loss_mlp": 1.02788055,
      "epoch": 0.3285035772259965,
      "flos": 12349500935040.0,
      "grad_norm": 4.191038639392342,
      "language_loss": 0.87364781,
      "learning_rate": 3.1360308635615383e-06,
      "loss": 0.89556968,
      "num_input_tokens_seen": 58894660,
      "step": 2732,
      "time_per_iteration": 2.7408499717712402
    },
    {
      "auxiliary_loss_clip": 0.0112719,
      "auxiliary_loss_mlp": 0.0103996,
      "balance_loss_clip": 1.04987395,
      "balance_loss_mlp": 1.02861786,
      "epoch": 0.3286238201166356,
      "flos": 24316084932480.0,
      "grad_norm": 2.4128907350894693,
      "language_loss": 0.78583336,
      "learning_rate": 3.135389669798311e-06,
      "loss": 0.80750483,
      "num_input_tokens_seen": 58912720,
      "step": 2733,
      "time_per_iteration": 2.7849206924438477
    },
    {
      "auxiliary_loss_clip": 0.01128562,
      "auxiliary_loss_mlp": 0.0103126,
      "balance_loss_clip": 1.05610847,
      "balance_loss_mlp": 1.0224092,
      "epoch": 0.3287440630072747,
      "flos": 21392812471680.0,
      "grad_norm": 2.4663687096211984,
      "language_loss": 0.79947108,
      "learning_rate": 3.134748303797373e-06,
      "loss": 0.82106924,
      "num_input_tokens_seen": 58930090,
      "step": 2734,
      "time_per_iteration": 2.836294412612915
    },
    {
      "auxiliary_loss_clip": 0.01195117,
      "auxiliary_loss_mlp": 0.01036291,
      "balance_loss_clip": 1.06081116,
      "balance_loss_mlp": 1.02434635,
      "epoch": 0.32886430589791377,
      "flos": 23732536579200.0,
      "grad_norm": 2.1485507857276263,
      "language_loss": 0.81050074,
      "learning_rate": 3.1341067656560203e-06,
      "loss": 0.83281481,
      "num_input_tokens_seen": 58947935,
      "step": 2735,
      "time_per_iteration": 2.863785982131958
    },
    {
      "auxiliary_loss_clip": 0.01165142,
      "auxiliary_loss_mlp": 0.01032508,
      "balance_loss_clip": 1.05715573,
      "balance_loss_mlp": 1.02329278,
      "epoch": 0.3289845487885529,
      "flos": 22418708814720.0,
      "grad_norm": 1.9722160278472922,
      "language_loss": 0.85825694,
      "learning_rate": 3.133465055471572e-06,
      "loss": 0.88023347,
      "num_input_tokens_seen": 58967720,
      "step": 2736,
      "time_per_iteration": 3.343073606491089
    },
    {
      "auxiliary_loss_clip": 0.01204625,
      "auxiliary_loss_mlp": 0.01035966,
      "balance_loss_clip": 1.06420708,
      "balance_loss_mlp": 1.02616096,
      "epoch": 0.329104791679192,
      "flos": 19682603147520.0,
      "grad_norm": 2.4278049989134467,
      "language_loss": 0.66425276,
      "learning_rate": 3.1328231733413767e-06,
      "loss": 0.68665862,
      "num_input_tokens_seen": 58984360,
      "step": 2737,
      "time_per_iteration": 2.62835955619812
    },
    {
      "auxiliary_loss_clip": 0.01191543,
      "auxiliary_loss_mlp": 0.01029638,
      "balance_loss_clip": 1.06076539,
      "balance_loss_mlp": 1.02011299,
      "epoch": 0.32922503456983104,
      "flos": 15997234803840.0,
      "grad_norm": 2.1031319457050253,
      "language_loss": 0.91090828,
      "learning_rate": 3.1321811193628067e-06,
      "loss": 0.93312007,
      "num_input_tokens_seen": 59002505,
      "step": 2738,
      "time_per_iteration": 2.6220171451568604
    },
    {
      "auxiliary_loss_clip": 0.01178631,
      "auxiliary_loss_mlp": 0.0103639,
      "balance_loss_clip": 1.05887151,
      "balance_loss_mlp": 1.02612054,
      "epoch": 0.32934527746047015,
      "flos": 26834069260800.0,
      "grad_norm": 2.277391772292505,
      "language_loss": 0.70295632,
      "learning_rate": 3.131538893633261e-06,
      "loss": 0.72510648,
      "num_input_tokens_seen": 59022065,
      "step": 2739,
      "time_per_iteration": 2.7274158000946045
    },
    {
      "auxiliary_loss_clip": 0.01192307,
      "auxiliary_loss_mlp": 0.0103067,
      "balance_loss_clip": 1.06117845,
      "balance_loss_mlp": 1.02127016,
      "epoch": 0.32946552035110926,
      "flos": 23403774372480.0,
      "grad_norm": 2.276581107223923,
      "language_loss": 0.77831125,
      "learning_rate": 3.130896496250165e-06,
      "loss": 0.80054104,
      "num_input_tokens_seen": 59041890,
      "step": 2740,
      "time_per_iteration": 2.7996201515197754
    },
    {
      "auxiliary_loss_clip": 0.01175606,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.06002259,
      "balance_loss_mlp": 1.01901364,
      "epoch": 0.3295857632417483,
      "flos": 14172470029440.0,
      "grad_norm": 2.086432474989879,
      "language_loss": 0.86997694,
      "learning_rate": 3.1302539273109693e-06,
      "loss": 0.89202261,
      "num_input_tokens_seen": 59058715,
      "step": 2741,
      "time_per_iteration": 2.7003352642059326
    },
    {
      "auxiliary_loss_clip": 0.01181285,
      "auxiliary_loss_mlp": 0.01035001,
      "balance_loss_clip": 1.06242895,
      "balance_loss_mlp": 1.02470112,
      "epoch": 0.32970600613238743,
      "flos": 22196708807040.0,
      "grad_norm": 2.6578672884234953,
      "language_loss": 0.80388528,
      "learning_rate": 3.1296111869131513e-06,
      "loss": 0.8260482,
      "num_input_tokens_seen": 59076140,
      "step": 2742,
      "time_per_iteration": 2.6918816566467285
    },
    {
      "auxiliary_loss_clip": 0.01147787,
      "auxiliary_loss_mlp": 0.01032978,
      "balance_loss_clip": 1.04949665,
      "balance_loss_mlp": 1.02343571,
      "epoch": 0.32982624902302654,
      "flos": 22053784590720.0,
      "grad_norm": 1.8144225736554906,
      "language_loss": 0.85871363,
      "learning_rate": 3.1289682751542153e-06,
      "loss": 0.8805213,
      "num_input_tokens_seen": 59095700,
      "step": 2743,
      "time_per_iteration": 2.7215957641601562
    },
    {
      "auxiliary_loss_clip": 0.01175955,
      "auxiliary_loss_mlp": 0.01031925,
      "balance_loss_clip": 1.05730557,
      "balance_loss_mlp": 1.02259684,
      "epoch": 0.3299464919136656,
      "flos": 18661626967680.0,
      "grad_norm": 1.940026015628617,
      "language_loss": 0.71394604,
      "learning_rate": 3.1283251921316883e-06,
      "loss": 0.73602486,
      "num_input_tokens_seen": 59113445,
      "step": 2744,
      "time_per_iteration": 2.6753783226013184
    },
    {
      "auxiliary_loss_clip": 0.01162125,
      "auxiliary_loss_mlp": 0.01035923,
      "balance_loss_clip": 1.05813861,
      "balance_loss_mlp": 1.02681518,
      "epoch": 0.3300667348043047,
      "flos": 13407357404160.0,
      "grad_norm": 2.837907192147265,
      "language_loss": 0.81098545,
      "learning_rate": 3.1276819379431277e-06,
      "loss": 0.83296597,
      "num_input_tokens_seen": 59131535,
      "step": 2745,
      "time_per_iteration": 2.763732433319092
    },
    {
      "auxiliary_loss_clip": 0.01174587,
      "auxiliary_loss_mlp": 0.01033273,
      "balance_loss_clip": 1.05605829,
      "balance_loss_mlp": 1.02327132,
      "epoch": 0.33018697769494376,
      "flos": 15742556398080.0,
      "grad_norm": 2.1081337089801533,
      "language_loss": 0.75579238,
      "learning_rate": 3.1270385126861134e-06,
      "loss": 0.77787101,
      "num_input_tokens_seen": 59149520,
      "step": 2746,
      "time_per_iteration": 2.693553924560547
    },
    {
      "auxiliary_loss_clip": 0.0115113,
      "auxiliary_loss_mlp": 0.01034702,
      "balance_loss_clip": 1.05499518,
      "balance_loss_mlp": 1.02413428,
      "epoch": 0.3303072205855829,
      "flos": 18258601392000.0,
      "grad_norm": 1.8961244720096087,
      "language_loss": 0.82066607,
      "learning_rate": 3.1263949164582533e-06,
      "loss": 0.84252441,
      "num_input_tokens_seen": 59169170,
      "step": 2747,
      "time_per_iteration": 2.7697436809539795
    },
    {
      "auxiliary_loss_clip": 0.01176303,
      "auxiliary_loss_mlp": 0.01033111,
      "balance_loss_clip": 1.05285478,
      "balance_loss_mlp": 1.02382469,
      "epoch": 0.330427463476222,
      "flos": 17749424148480.0,
      "grad_norm": 1.8495410984277931,
      "language_loss": 0.7820099,
      "learning_rate": 3.1257511493571797e-06,
      "loss": 0.80410403,
      "num_input_tokens_seen": 59187675,
      "step": 2748,
      "time_per_iteration": 2.6858386993408203
    },
    {
      "auxiliary_loss_clip": 0.01134607,
      "auxiliary_loss_mlp": 0.01030667,
      "balance_loss_clip": 1.05472755,
      "balance_loss_mlp": 1.02071309,
      "epoch": 0.33054770636686104,
      "flos": 27162580072320.0,
      "grad_norm": 1.9545900852512816,
      "language_loss": 0.7860415,
      "learning_rate": 3.125107211480552e-06,
      "loss": 0.80769432,
      "num_input_tokens_seen": 59207610,
      "step": 2749,
      "time_per_iteration": 2.791935682296753
    },
    {
      "auxiliary_loss_clip": 0.01144528,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.05701053,
      "balance_loss_mlp": 1.01818466,
      "epoch": 0.33066794925750015,
      "flos": 20117193799680.0,
      "grad_norm": 1.7197610336289608,
      "language_loss": 0.79688144,
      "learning_rate": 3.124463102926054e-06,
      "loss": 0.81861079,
      "num_input_tokens_seen": 59226945,
      "step": 2750,
      "time_per_iteration": 2.7545199394226074
    },
    {
      "auxiliary_loss_clip": 0.01107772,
      "auxiliary_loss_mlp": 0.00755674,
      "balance_loss_clip": 1.03594291,
      "balance_loss_mlp": 1.00213134,
      "epoch": 0.33078819214813926,
      "flos": 70642609718400.0,
      "grad_norm": 0.7905962516814257,
      "language_loss": 0.61582541,
      "learning_rate": 3.1238188237913984e-06,
      "loss": 0.63445985,
      "num_input_tokens_seen": 59291485,
      "step": 2751,
      "time_per_iteration": 4.430990934371948
    },
    {
      "auxiliary_loss_clip": 0.01169786,
      "auxiliary_loss_mlp": 0.01038614,
      "balance_loss_clip": 1.05749393,
      "balance_loss_mlp": 1.02792144,
      "epoch": 0.3309084350387783,
      "flos": 21141940907520.0,
      "grad_norm": 1.8837800484110858,
      "language_loss": 0.76555634,
      "learning_rate": 3.1231743741743202e-06,
      "loss": 0.78764039,
      "num_input_tokens_seen": 59310990,
      "step": 2752,
      "time_per_iteration": 2.660719156265259
    },
    {
      "auxiliary_loss_clip": 0.01161739,
      "auxiliary_loss_mlp": 0.01031488,
      "balance_loss_clip": 1.0560863,
      "balance_loss_mlp": 1.02167082,
      "epoch": 0.3310286779294174,
      "flos": 14209350318720.0,
      "grad_norm": 2.2211829668743057,
      "language_loss": 0.83753043,
      "learning_rate": 3.122529754172582e-06,
      "loss": 0.85946262,
      "num_input_tokens_seen": 59327875,
      "step": 2753,
      "time_per_iteration": 4.885593891143799
    },
    {
      "auxiliary_loss_clip": 0.0117721,
      "auxiliary_loss_mlp": 0.01030765,
      "balance_loss_clip": 1.05896378,
      "balance_loss_mlp": 1.02070951,
      "epoch": 0.33114892082005654,
      "flos": 20778130005120.0,
      "grad_norm": 1.9672482953770185,
      "language_loss": 0.72251034,
      "learning_rate": 3.1218849638839736e-06,
      "loss": 0.74459004,
      "num_input_tokens_seen": 59347135,
      "step": 2754,
      "time_per_iteration": 3.598261833190918
    },
    {
      "auxiliary_loss_clip": 0.0112949,
      "auxiliary_loss_mlp": 0.01033706,
      "balance_loss_clip": 1.05010962,
      "balance_loss_mlp": 1.02302444,
      "epoch": 0.3312691637106956,
      "flos": 17090750499840.0,
      "grad_norm": 2.0789254774896593,
      "language_loss": 0.78360552,
      "learning_rate": 3.121240003406307e-06,
      "loss": 0.80523741,
      "num_input_tokens_seen": 59365985,
      "step": 2755,
      "time_per_iteration": 2.8872978687286377
    },
    {
      "auxiliary_loss_clip": 0.01176662,
      "auxiliary_loss_mlp": 0.0103799,
      "balance_loss_clip": 1.05991375,
      "balance_loss_mlp": 1.02779794,
      "epoch": 0.3313894066013347,
      "flos": 29456230008960.0,
      "grad_norm": 2.025483380613737,
      "language_loss": 0.72481108,
      "learning_rate": 3.120594872837425e-06,
      "loss": 0.74695754,
      "num_input_tokens_seen": 59384655,
      "step": 2756,
      "time_per_iteration": 2.75400972366333
    },
    {
      "auxiliary_loss_clip": 0.01086757,
      "auxiliary_loss_mlp": 0.01001887,
      "balance_loss_clip": 1.03350186,
      "balance_loss_mlp": 1.00019395,
      "epoch": 0.3315096494919738,
      "flos": 61419242280960.0,
      "grad_norm": 0.825847317856505,
      "language_loss": 0.62359059,
      "learning_rate": 3.1199495722751906e-06,
      "loss": 0.64447713,
      "num_input_tokens_seen": 59444185,
      "step": 2757,
      "time_per_iteration": 3.3078575134277344
    },
    {
      "auxiliary_loss_clip": 0.0113999,
      "auxiliary_loss_mlp": 0.0103601,
      "balance_loss_clip": 1.05580115,
      "balance_loss_mlp": 1.02605581,
      "epoch": 0.33162989238261287,
      "flos": 21653057485440.0,
      "grad_norm": 1.7116156893307286,
      "language_loss": 0.83823133,
      "learning_rate": 3.1193041018174972e-06,
      "loss": 0.85999131,
      "num_input_tokens_seen": 59464900,
      "step": 2758,
      "time_per_iteration": 2.7458205223083496
    },
    {
      "auxiliary_loss_clip": 0.01113919,
      "auxiliary_loss_mlp": 0.01032535,
      "balance_loss_clip": 1.05404341,
      "balance_loss_mlp": 1.02243209,
      "epoch": 0.331750135273252,
      "flos": 22674787850880.0,
      "grad_norm": 2.0044549378775263,
      "language_loss": 0.94767302,
      "learning_rate": 3.118658461562261e-06,
      "loss": 0.96913755,
      "num_input_tokens_seen": 59481000,
      "step": 2759,
      "time_per_iteration": 3.2181718349456787
    },
    {
      "auxiliary_loss_clip": 0.01205117,
      "auxiliary_loss_mlp": 0.0103851,
      "balance_loss_clip": 1.06386089,
      "balance_loss_mlp": 1.02829957,
      "epoch": 0.33187037816389103,
      "flos": 22746896403840.0,
      "grad_norm": 1.416941328312977,
      "language_loss": 0.85101998,
      "learning_rate": 3.118012651607426e-06,
      "loss": 0.87345624,
      "num_input_tokens_seen": 59502605,
      "step": 2760,
      "time_per_iteration": 2.901244878768921
    },
    {
      "auxiliary_loss_clip": 0.01194127,
      "auxiliary_loss_mlp": 0.00763609,
      "balance_loss_clip": 1.06307483,
      "balance_loss_mlp": 1.00083816,
      "epoch": 0.33199062105453014,
      "flos": 19203769918080.0,
      "grad_norm": 2.0882256543368882,
      "language_loss": 0.83872724,
      "learning_rate": 3.1173666720509603e-06,
      "loss": 0.85830456,
      "num_input_tokens_seen": 59519540,
      "step": 2761,
      "time_per_iteration": 2.7161660194396973
    },
    {
      "auxiliary_loss_clip": 0.01135092,
      "auxiliary_loss_mlp": 0.01031803,
      "balance_loss_clip": 1.0543108,
      "balance_loss_mlp": 1.02249897,
      "epoch": 0.33211086394516925,
      "flos": 31577006764800.0,
      "grad_norm": 1.9907412194641234,
      "language_loss": 0.68467075,
      "learning_rate": 3.116720522990859e-06,
      "loss": 0.70633972,
      "num_input_tokens_seen": 59540415,
      "step": 2762,
      "time_per_iteration": 2.941807746887207
    },
    {
      "auxiliary_loss_clip": 0.01118982,
      "auxiliary_loss_mlp": 0.01038591,
      "balance_loss_clip": 1.04507279,
      "balance_loss_mlp": 1.02833343,
      "epoch": 0.3322311068358083,
      "flos": 17932496791680.0,
      "grad_norm": 2.193067890422915,
      "language_loss": 0.62079728,
      "learning_rate": 3.116074204525142e-06,
      "loss": 0.64237297,
      "num_input_tokens_seen": 59558590,
      "step": 2763,
      "time_per_iteration": 2.7810754776000977
    },
    {
      "auxiliary_loss_clip": 0.01144188,
      "auxiliary_loss_mlp": 0.01032341,
      "balance_loss_clip": 1.058007,
      "balance_loss_mlp": 1.02191615,
      "epoch": 0.3323513497264474,
      "flos": 32269831269120.0,
      "grad_norm": 1.8126222920216037,
      "language_loss": 0.83704293,
      "learning_rate": 3.1154277167518553e-06,
      "loss": 0.85880828,
      "num_input_tokens_seen": 59580205,
      "step": 2764,
      "time_per_iteration": 2.8833417892456055
    },
    {
      "auxiliary_loss_clip": 0.01091996,
      "auxiliary_loss_mlp": 0.01002122,
      "balance_loss_clip": 1.02794862,
      "balance_loss_mlp": 1.00048912,
      "epoch": 0.33247159261708653,
      "flos": 52668674588160.0,
      "grad_norm": 0.7757618527591361,
      "language_loss": 0.59512389,
      "learning_rate": 3.114781059769072e-06,
      "loss": 0.61606503,
      "num_input_tokens_seen": 59631530,
      "step": 2765,
      "time_per_iteration": 3.069768190383911
    },
    {
      "auxiliary_loss_clip": 0.01159897,
      "auxiliary_loss_mlp": 0.01034157,
      "balance_loss_clip": 1.05504704,
      "balance_loss_mlp": 1.02372015,
      "epoch": 0.3325918355077256,
      "flos": 27125232906240.0,
      "grad_norm": 4.40609425388848,
      "language_loss": 0.67514539,
      "learning_rate": 3.1141342336748874e-06,
      "loss": 0.69708598,
      "num_input_tokens_seen": 59651090,
      "step": 2766,
      "time_per_iteration": 2.839026927947998
    },
    {
      "auxiliary_loss_clip": 0.01171009,
      "auxiliary_loss_mlp": 0.00763703,
      "balance_loss_clip": 1.05796933,
      "balance_loss_mlp": 1.00090361,
      "epoch": 0.3327120783983647,
      "flos": 23664414435840.0,
      "grad_norm": 1.4733567876066043,
      "language_loss": 0.82120669,
      "learning_rate": 3.1134872385674253e-06,
      "loss": 0.84055376,
      "num_input_tokens_seen": 59675245,
      "step": 2767,
      "time_per_iteration": 2.756342649459839
    },
    {
      "auxiliary_loss_clip": 0.01121844,
      "auxiliary_loss_mlp": 0.01033831,
      "balance_loss_clip": 1.05103052,
      "balance_loss_mlp": 1.02347803,
      "epoch": 0.3328323212890038,
      "flos": 19171378828800.0,
      "grad_norm": 1.9655156587481875,
      "language_loss": 0.85711503,
      "learning_rate": 3.1128400745448353e-06,
      "loss": 0.87867177,
      "num_input_tokens_seen": 59694625,
      "step": 2768,
      "time_per_iteration": 2.794199228286743
    },
    {
      "auxiliary_loss_clip": 0.01160137,
      "auxiliary_loss_mlp": 0.01038069,
      "balance_loss_clip": 1.05739355,
      "balance_loss_mlp": 1.02808571,
      "epoch": 0.33295256417964286,
      "flos": 37706347463040.0,
      "grad_norm": 2.698225910964282,
      "language_loss": 0.63039494,
      "learning_rate": 3.11219274170529e-06,
      "loss": 0.65237701,
      "num_input_tokens_seen": 59716435,
      "step": 2769,
      "time_per_iteration": 2.878419876098633
    },
    {
      "auxiliary_loss_clip": 0.01187486,
      "auxiliary_loss_mlp": 0.01031265,
      "balance_loss_clip": 1.05843115,
      "balance_loss_mlp": 1.02163267,
      "epoch": 0.333072807070282,
      "flos": 26505989412480.0,
      "grad_norm": 1.7768202981787804,
      "language_loss": 0.81639922,
      "learning_rate": 3.1115452401469903e-06,
      "loss": 0.83858681,
      "num_input_tokens_seen": 59736835,
      "step": 2770,
      "time_per_iteration": 2.68346905708313
    },
    {
      "auxiliary_loss_clip": 0.01164163,
      "auxiliary_loss_mlp": 0.0103834,
      "balance_loss_clip": 1.05813408,
      "balance_loss_mlp": 1.02867186,
      "epoch": 0.3331930499609211,
      "flos": 21430913823360.0,
      "grad_norm": 2.119895850451059,
      "language_loss": 0.86796808,
      "learning_rate": 3.1108975699681613e-06,
      "loss": 0.88999307,
      "num_input_tokens_seen": 59754230,
      "step": 2771,
      "time_per_iteration": 2.710498332977295
    },
    {
      "auxiliary_loss_clip": 0.01139368,
      "auxiliary_loss_mlp": 0.01036614,
      "balance_loss_clip": 1.05188549,
      "balance_loss_mlp": 1.0279355,
      "epoch": 0.33331329285156014,
      "flos": 20659947281280.0,
      "grad_norm": 1.663895166459772,
      "language_loss": 0.71414733,
      "learning_rate": 3.1102497312670542e-06,
      "loss": 0.73590708,
      "num_input_tokens_seen": 59772235,
      "step": 2772,
      "time_per_iteration": 2.7743780612945557
    },
    {
      "auxiliary_loss_clip": 0.01151568,
      "auxiliary_loss_mlp": 0.01033217,
      "balance_loss_clip": 1.05627346,
      "balance_loss_mlp": 1.02305436,
      "epoch": 0.33343353574219925,
      "flos": 28001596930560.0,
      "grad_norm": 1.7575585952496033,
      "language_loss": 0.80667138,
      "learning_rate": 3.109601724141946e-06,
      "loss": 0.82851928,
      "num_input_tokens_seen": 59791230,
      "step": 2773,
      "time_per_iteration": 2.8551828861236572
    },
    {
      "auxiliary_loss_clip": 0.01177348,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.06119573,
      "balance_loss_mlp": 1.024405,
      "epoch": 0.33355377863283836,
      "flos": 23764963582080.0,
      "grad_norm": 1.9192082473551306,
      "language_loss": 0.68149745,
      "learning_rate": 3.108953548691138e-06,
      "loss": 0.70360535,
      "num_input_tokens_seen": 59811315,
      "step": 2774,
      "time_per_iteration": 2.7625956535339355
    },
    {
      "auxiliary_loss_clip": 0.01145306,
      "auxiliary_loss_mlp": 0.01028345,
      "balance_loss_clip": 1.05390644,
      "balance_loss_mlp": 1.01795006,
      "epoch": 0.3336740215234774,
      "flos": 37779677078400.0,
      "grad_norm": 2.121908653016425,
      "language_loss": 0.7247473,
      "learning_rate": 3.108305205012959e-06,
      "loss": 0.7464838,
      "num_input_tokens_seen": 59832010,
      "step": 2775,
      "time_per_iteration": 2.8587732315063477
    },
    {
      "auxiliary_loss_clip": 0.01186969,
      "auxiliary_loss_mlp": 0.01027776,
      "balance_loss_clip": 1.05860949,
      "balance_loss_mlp": 1.01851916,
      "epoch": 0.3337942644141165,
      "flos": 25519056347520.0,
      "grad_norm": 2.0704038616744627,
      "language_loss": 0.87406015,
      "learning_rate": 3.107656693205761e-06,
      "loss": 0.89620757,
      "num_input_tokens_seen": 59851450,
      "step": 2776,
      "time_per_iteration": 3.713768243789673
    },
    {
      "auxiliary_loss_clip": 0.01113586,
      "auxiliary_loss_mlp": 0.01034023,
      "balance_loss_clip": 1.04729319,
      "balance_loss_mlp": 1.02250123,
      "epoch": 0.3339145073047556,
      "flos": 25989844930560.0,
      "grad_norm": 2.2449982142069365,
      "language_loss": 0.70493835,
      "learning_rate": 3.107008013367924e-06,
      "loss": 0.72641444,
      "num_input_tokens_seen": 59870245,
      "step": 2777,
      "time_per_iteration": 2.8532674312591553
    },
    {
      "auxiliary_loss_clip": 0.01155974,
      "auxiliary_loss_mlp": 0.00764219,
      "balance_loss_clip": 1.05528998,
      "balance_loss_mlp": 1.00093043,
      "epoch": 0.3340347501953947,
      "flos": 19062569554560.0,
      "grad_norm": 2.424610162014106,
      "language_loss": 0.86883277,
      "learning_rate": 3.1063591655978507e-06,
      "loss": 0.8880347,
      "num_input_tokens_seen": 59886195,
      "step": 2778,
      "time_per_iteration": 2.7376818656921387
    },
    {
      "auxiliary_loss_clip": 0.01133631,
      "auxiliary_loss_mlp": 0.01032593,
      "balance_loss_clip": 1.05246258,
      "balance_loss_mlp": 1.02232957,
      "epoch": 0.3341549930860338,
      "flos": 18109715518080.0,
      "grad_norm": 1.7728846244204814,
      "language_loss": 0.79515612,
      "learning_rate": 3.105710149993972e-06,
      "loss": 0.81681836,
      "num_input_tokens_seen": 59905525,
      "step": 2779,
      "time_per_iteration": 4.6579155921936035
    },
    {
      "auxiliary_loss_clip": 0.01178623,
      "auxiliary_loss_mlp": 0.01034598,
      "balance_loss_clip": 1.05626953,
      "balance_loss_mlp": 1.02580047,
      "epoch": 0.33427523597667286,
      "flos": 22674967418880.0,
      "grad_norm": 2.188810540769781,
      "language_loss": 0.85647237,
      "learning_rate": 3.1050609666547427e-06,
      "loss": 0.87860459,
      "num_input_tokens_seen": 59925085,
      "step": 2780,
      "time_per_iteration": 3.6456565856933594
    },
    {
      "auxiliary_loss_clip": 0.0119994,
      "auxiliary_loss_mlp": 0.00764281,
      "balance_loss_clip": 1.06068814,
      "balance_loss_mlp": 1.00105703,
      "epoch": 0.33439547886731197,
      "flos": 22638338524800.0,
      "grad_norm": 1.7864318907450074,
      "language_loss": 0.76986414,
      "learning_rate": 3.104411615678644e-06,
      "loss": 0.78950638,
      "num_input_tokens_seen": 59943935,
      "step": 2781,
      "time_per_iteration": 2.667062759399414
    },
    {
      "auxiliary_loss_clip": 0.01189864,
      "auxiliary_loss_mlp": 0.010302,
      "balance_loss_clip": 1.05992842,
      "balance_loss_mlp": 1.01960218,
      "epoch": 0.3345157217579511,
      "flos": 24096383395200.0,
      "grad_norm": 2.950940684870054,
      "language_loss": 0.73443997,
      "learning_rate": 3.1037620971641803e-06,
      "loss": 0.75664055,
      "num_input_tokens_seen": 59963725,
      "step": 2782,
      "time_per_iteration": 2.6526947021484375
    },
    {
      "auxiliary_loss_clip": 0.01144138,
      "auxiliary_loss_mlp": 0.00764961,
      "balance_loss_clip": 1.04787564,
      "balance_loss_mlp": 1.00096273,
      "epoch": 0.33463596464859013,
      "flos": 18989491334400.0,
      "grad_norm": 2.2524151918320148,
      "language_loss": 0.65063423,
      "learning_rate": 3.1031124112098844e-06,
      "loss": 0.66972518,
      "num_input_tokens_seen": 59981935,
      "step": 2783,
      "time_per_iteration": 2.7720224857330322
    },
    {
      "auxiliary_loss_clip": 0.01144027,
      "auxiliary_loss_mlp": 0.01035874,
      "balance_loss_clip": 1.05425107,
      "balance_loss_mlp": 1.02596164,
      "epoch": 0.33475620753922924,
      "flos": 20375607219840.0,
      "grad_norm": 1.8530573317526469,
      "language_loss": 0.71918869,
      "learning_rate": 3.1024625579143127e-06,
      "loss": 0.74098766,
      "num_input_tokens_seen": 59999455,
      "step": 2784,
      "time_per_iteration": 2.793689727783203
    },
    {
      "auxiliary_loss_clip": 0.01133937,
      "auxiliary_loss_mlp": 0.01036137,
      "balance_loss_clip": 1.04952168,
      "balance_loss_mlp": 1.02534282,
      "epoch": 0.33487645042986836,
      "flos": 18182578256640.0,
      "grad_norm": 1.964873489894207,
      "language_loss": 0.73080766,
      "learning_rate": 3.101812537376048e-06,
      "loss": 0.7525084,
      "num_input_tokens_seen": 60018475,
      "step": 2785,
      "time_per_iteration": 2.728863000869751
    },
    {
      "auxiliary_loss_clip": 0.01200376,
      "auxiliary_loss_mlp": 0.01033724,
      "balance_loss_clip": 1.06228423,
      "balance_loss_mlp": 1.02381146,
      "epoch": 0.3349966933205074,
      "flos": 25848824135040.0,
      "grad_norm": 2.2692919403020713,
      "language_loss": 0.84251851,
      "learning_rate": 3.1011623496936973e-06,
      "loss": 0.86485958,
      "num_input_tokens_seen": 60036770,
      "step": 2786,
      "time_per_iteration": 2.7097837924957275
    },
    {
      "auxiliary_loss_clip": 0.01199257,
      "auxiliary_loss_mlp": 0.01033925,
      "balance_loss_clip": 1.06103563,
      "balance_loss_mlp": 1.02413797,
      "epoch": 0.3351169362111465,
      "flos": 28111447699200.0,
      "grad_norm": 1.8092171331017906,
      "language_loss": 0.70069057,
      "learning_rate": 3.100511994965893e-06,
      "loss": 0.72302234,
      "num_input_tokens_seen": 60056725,
      "step": 2787,
      "time_per_iteration": 2.754690408706665
    },
    {
      "auxiliary_loss_clip": 0.01200456,
      "auxiliary_loss_mlp": 0.01031497,
      "balance_loss_clip": 1.06230593,
      "balance_loss_mlp": 1.02259183,
      "epoch": 0.33523717910178563,
      "flos": 22673315393280.0,
      "grad_norm": 1.7088312923188722,
      "language_loss": 0.83885682,
      "learning_rate": 3.0998614732912947e-06,
      "loss": 0.86117637,
      "num_input_tokens_seen": 60076100,
      "step": 2788,
      "time_per_iteration": 2.6635866165161133
    },
    {
      "auxiliary_loss_clip": 0.01203013,
      "auxiliary_loss_mlp": 0.01036332,
      "balance_loss_clip": 1.06314802,
      "balance_loss_mlp": 1.02606809,
      "epoch": 0.3353574219924247,
      "flos": 15669801400320.0,
      "grad_norm": 1.9181476684069085,
      "language_loss": 0.67978275,
      "learning_rate": 3.0992107847685855e-06,
      "loss": 0.70217621,
      "num_input_tokens_seen": 60093815,
      "step": 2789,
      "time_per_iteration": 2.5979931354522705
    },
    {
      "auxiliary_loss_clip": 0.01202487,
      "auxiliary_loss_mlp": 0.01033522,
      "balance_loss_clip": 1.06280744,
      "balance_loss_mlp": 1.02321672,
      "epoch": 0.3354776648830638,
      "flos": 24790644443520.0,
      "grad_norm": 1.762173128548306,
      "language_loss": 0.79537457,
      "learning_rate": 3.0985599294964736e-06,
      "loss": 0.81773472,
      "num_input_tokens_seen": 60113370,
      "step": 2790,
      "time_per_iteration": 2.6746225357055664
    },
    {
      "auxiliary_loss_clip": 0.01205297,
      "auxiliary_loss_mlp": 0.01031979,
      "balance_loss_clip": 1.06265569,
      "balance_loss_mlp": 1.02229333,
      "epoch": 0.33559790777370285,
      "flos": 28694852398080.0,
      "grad_norm": 2.555341917961168,
      "language_loss": 0.69793719,
      "learning_rate": 3.097908907573695e-06,
      "loss": 0.72030997,
      "num_input_tokens_seen": 60131350,
      "step": 2791,
      "time_per_iteration": 2.668006658554077
    },
    {
      "auxiliary_loss_clip": 0.01202301,
      "auxiliary_loss_mlp": 0.01030364,
      "balance_loss_clip": 1.06249714,
      "balance_loss_mlp": 1.02157211,
      "epoch": 0.33571815066434196,
      "flos": 22235779825920.0,
      "grad_norm": 2.2199724520556186,
      "language_loss": 0.89550644,
      "learning_rate": 3.0972577190990067e-06,
      "loss": 0.91783309,
      "num_input_tokens_seen": 60149830,
      "step": 2792,
      "time_per_iteration": 2.6125240325927734
    },
    {
      "auxiliary_loss_clip": 0.01203344,
      "auxiliary_loss_mlp": 0.01038586,
      "balance_loss_clip": 1.06406474,
      "balance_loss_mlp": 1.02900791,
      "epoch": 0.3358383935549811,
      "flos": 23842279607040.0,
      "grad_norm": 1.7774939154706337,
      "language_loss": 0.80260432,
      "learning_rate": 3.096606364171196e-06,
      "loss": 0.82502359,
      "num_input_tokens_seen": 60169620,
      "step": 2793,
      "time_per_iteration": 2.649911642074585
    },
    {
      "auxiliary_loss_clip": 0.01203682,
      "auxiliary_loss_mlp": 0.01030622,
      "balance_loss_clip": 1.06453109,
      "balance_loss_mlp": 1.02079916,
      "epoch": 0.33595863644562013,
      "flos": 22267308988800.0,
      "grad_norm": 2.4856897897753125,
      "language_loss": 0.84933853,
      "learning_rate": 3.0959548428890703e-06,
      "loss": 0.87168157,
      "num_input_tokens_seen": 60188490,
      "step": 2794,
      "time_per_iteration": 2.672675609588623
    },
    {
      "auxiliary_loss_clip": 0.01201588,
      "auxiliary_loss_mlp": 0.01031278,
      "balance_loss_clip": 1.06275392,
      "balance_loss_mlp": 1.02170503,
      "epoch": 0.33607887933625924,
      "flos": 20119779578880.0,
      "grad_norm": 1.636000081969542,
      "language_loss": 0.84111768,
      "learning_rate": 3.095303155351468e-06,
      "loss": 0.86344635,
      "num_input_tokens_seen": 60208695,
      "step": 2795,
      "time_per_iteration": 2.5808515548706055
    },
    {
      "auxiliary_loss_clip": 0.01200657,
      "auxiliary_loss_mlp": 0.010449,
      "balance_loss_clip": 1.06194663,
      "balance_loss_mlp": 1.03570354,
      "epoch": 0.33619912222689835,
      "flos": 19318109886720.0,
      "grad_norm": 2.2817069705988438,
      "language_loss": 0.79001939,
      "learning_rate": 3.0946513016572464e-06,
      "loss": 0.81247497,
      "num_input_tokens_seen": 60227600,
      "step": 2796,
      "time_per_iteration": 2.596431016921997
    },
    {
      "auxiliary_loss_clip": 0.01202819,
      "auxiliary_loss_mlp": 0.01035517,
      "balance_loss_clip": 1.06010604,
      "balance_loss_mlp": 1.0247643,
      "epoch": 0.3363193651175374,
      "flos": 16800664262400.0,
      "grad_norm": 2.035370265268154,
      "language_loss": 0.76884544,
      "learning_rate": 3.0939992819052938e-06,
      "loss": 0.79122883,
      "num_input_tokens_seen": 60245110,
      "step": 2797,
      "time_per_iteration": 2.6102781295776367
    },
    {
      "auxiliary_loss_clip": 0.01204683,
      "auxiliary_loss_mlp": 0.01037266,
      "balance_loss_clip": 1.06534624,
      "balance_loss_mlp": 1.02741909,
      "epoch": 0.3364396080081765,
      "flos": 23550289948800.0,
      "grad_norm": 2.206800309146081,
      "language_loss": 0.80914432,
      "learning_rate": 3.0933470961945193e-06,
      "loss": 0.83156383,
      "num_input_tokens_seen": 60263405,
      "step": 2798,
      "time_per_iteration": 2.6220896244049072
    },
    {
      "auxiliary_loss_clip": 0.01198271,
      "auxiliary_loss_mlp": 0.01031908,
      "balance_loss_clip": 1.06048048,
      "balance_loss_mlp": 1.02312231,
      "epoch": 0.3365598508988156,
      "flos": 28037902602240.0,
      "grad_norm": 1.7195600567393647,
      "language_loss": 0.68137604,
      "learning_rate": 3.0926947446238597e-06,
      "loss": 0.70367789,
      "num_input_tokens_seen": 60282975,
      "step": 2799,
      "time_per_iteration": 2.7089009284973145
    },
    {
      "auxiliary_loss_clip": 0.01200421,
      "auxiliary_loss_mlp": 0.01034352,
      "balance_loss_clip": 1.05898178,
      "balance_loss_mlp": 1.02344465,
      "epoch": 0.3366800937894547,
      "flos": 16982767238400.0,
      "grad_norm": 2.836801630520301,
      "language_loss": 0.82651275,
      "learning_rate": 3.092042227292276e-06,
      "loss": 0.8488605,
      "num_input_tokens_seen": 60299810,
      "step": 2800,
      "time_per_iteration": 2.639106035232544
    },
    {
      "auxiliary_loss_clip": 0.01197102,
      "auxiliary_loss_mlp": 0.01038807,
      "balance_loss_clip": 1.06035995,
      "balance_loss_mlp": 1.0296514,
      "epoch": 0.3368003366800938,
      "flos": 23915321913600.0,
      "grad_norm": 2.0698729680549386,
      "language_loss": 0.88255358,
      "learning_rate": 3.0913895442987557e-06,
      "loss": 0.90491259,
      "num_input_tokens_seen": 60320775,
      "step": 2801,
      "time_per_iteration": 2.6744508743286133
    },
    {
      "auxiliary_loss_clip": 0.012007,
      "auxiliary_loss_mlp": 0.01028278,
      "balance_loss_clip": 1.06090415,
      "balance_loss_mlp": 1.01947498,
      "epoch": 0.3369205795707329,
      "flos": 24791219061120.0,
      "grad_norm": 1.667425608181132,
      "language_loss": 0.85673797,
      "learning_rate": 3.090736695742308e-06,
      "loss": 0.87902772,
      "num_input_tokens_seen": 60341905,
      "step": 2802,
      "time_per_iteration": 3.7991013526916504
    },
    {
      "auxiliary_loss_clip": 0.01199013,
      "auxiliary_loss_mlp": 0.01029289,
      "balance_loss_clip": 1.06027234,
      "balance_loss_mlp": 1.01978207,
      "epoch": 0.33704082246137196,
      "flos": 17931096161280.0,
      "grad_norm": 2.400747818378506,
      "language_loss": 0.51776958,
      "learning_rate": 3.0900836817219713e-06,
      "loss": 0.54005259,
      "num_input_tokens_seen": 60358335,
      "step": 2803,
      "time_per_iteration": 2.6806678771972656
    },
    {
      "auxiliary_loss_clip": 0.0119672,
      "auxiliary_loss_mlp": 0.01033343,
      "balance_loss_clip": 1.05951178,
      "balance_loss_mlp": 1.02399099,
      "epoch": 0.33716106535201107,
      "flos": 21286517149440.0,
      "grad_norm": 1.7054739329765047,
      "language_loss": 0.83998883,
      "learning_rate": 3.089430502336807e-06,
      "loss": 0.86228943,
      "num_input_tokens_seen": 60378305,
      "step": 2804,
      "time_per_iteration": 2.602208375930786
    },
    {
      "auxiliary_loss_clip": 0.01201807,
      "auxiliary_loss_mlp": 0.01035244,
      "balance_loss_clip": 1.0612911,
      "balance_loss_mlp": 1.02477717,
      "epoch": 0.3372813082426502,
      "flos": 18402962152320.0,
      "grad_norm": 3.439584782363286,
      "language_loss": 0.89713085,
      "learning_rate": 3.088777157685902e-06,
      "loss": 0.9195013,
      "num_input_tokens_seen": 60393895,
      "step": 2805,
      "time_per_iteration": 4.433128356933594
    },
    {
      "auxiliary_loss_clip": 0.01199079,
      "auxiliary_loss_mlp": 0.01037506,
      "balance_loss_clip": 1.06084144,
      "balance_loss_mlp": 1.02813029,
      "epoch": 0.33740155113328923,
      "flos": 17201391367680.0,
      "grad_norm": 1.929115254507908,
      "language_loss": 0.85581452,
      "learning_rate": 3.088123647868367e-06,
      "loss": 0.87818038,
      "num_input_tokens_seen": 60410445,
      "step": 2806,
      "time_per_iteration": 3.5586414337158203
    },
    {
      "auxiliary_loss_clip": 0.01201367,
      "auxiliary_loss_mlp": 0.0103553,
      "balance_loss_clip": 1.0618012,
      "balance_loss_mlp": 1.02582049,
      "epoch": 0.33752179402392835,
      "flos": 29058950609280.0,
      "grad_norm": 2.210647968246168,
      "language_loss": 0.81405365,
      "learning_rate": 3.0874699729833405e-06,
      "loss": 0.83642268,
      "num_input_tokens_seen": 60431815,
      "step": 2807,
      "time_per_iteration": 2.697756052017212
    },
    {
      "auxiliary_loss_clip": 0.01197926,
      "auxiliary_loss_mlp": 0.01031185,
      "balance_loss_clip": 1.05870605,
      "balance_loss_mlp": 1.02172577,
      "epoch": 0.3376420369145674,
      "flos": 25080730680960.0,
      "grad_norm": 1.5864179856618943,
      "language_loss": 0.79654819,
      "learning_rate": 3.086816133129983e-06,
      "loss": 0.81883931,
      "num_input_tokens_seen": 60452075,
      "step": 2808,
      "time_per_iteration": 2.6233651638031006
    },
    {
      "auxiliary_loss_clip": 0.01199799,
      "auxiliary_loss_mlp": 0.01040846,
      "balance_loss_clip": 1.06165504,
      "balance_loss_mlp": 1.03101683,
      "epoch": 0.3377622798052065,
      "flos": 27490624007040.0,
      "grad_norm": 1.9434858633957006,
      "language_loss": 0.76217031,
      "learning_rate": 3.0861621284074826e-06,
      "loss": 0.78457677,
      "num_input_tokens_seen": 60472600,
      "step": 2809,
      "time_per_iteration": 2.6612370014190674
    },
    {
      "auxiliary_loss_clip": 0.01197879,
      "auxiliary_loss_mlp": 0.01036824,
      "balance_loss_clip": 1.06106353,
      "balance_loss_mlp": 1.02810431,
      "epoch": 0.3378825226958456,
      "flos": 21975211589760.0,
      "grad_norm": 1.5177111676290194,
      "language_loss": 0.72959316,
      "learning_rate": 3.085507958915051e-06,
      "loss": 0.75194013,
      "num_input_tokens_seen": 60491030,
      "step": 2810,
      "time_per_iteration": 2.6565232276916504
    },
    {
      "auxiliary_loss_clip": 0.01199768,
      "auxiliary_loss_mlp": 0.01035024,
      "balance_loss_clip": 1.05955386,
      "balance_loss_mlp": 1.025177,
      "epoch": 0.3380027655864847,
      "flos": 42523189200000.0,
      "grad_norm": 2.4562132318653145,
      "language_loss": 0.71436489,
      "learning_rate": 3.084853624751925e-06,
      "loss": 0.73671281,
      "num_input_tokens_seen": 60512615,
      "step": 2811,
      "time_per_iteration": 2.783097505569458
    },
    {
      "auxiliary_loss_clip": 0.01202469,
      "auxiliary_loss_mlp": 0.01037852,
      "balance_loss_clip": 1.06325042,
      "balance_loss_mlp": 1.02776098,
      "epoch": 0.3381230084771238,
      "flos": 26725080418560.0,
      "grad_norm": 1.7506207018692184,
      "language_loss": 0.85945636,
      "learning_rate": 3.0841991260173668e-06,
      "loss": 0.8818596,
      "num_input_tokens_seen": 60532520,
      "step": 2812,
      "time_per_iteration": 2.6677143573760986
    },
    {
      "auxiliary_loss_clip": 0.01197883,
      "auxiliary_loss_mlp": 0.01037313,
      "balance_loss_clip": 1.05971968,
      "balance_loss_mlp": 1.02732944,
      "epoch": 0.3382432513677629,
      "flos": 22710375250560.0,
      "grad_norm": 2.1431810057887932,
      "language_loss": 0.79885966,
      "learning_rate": 3.0835444628106634e-06,
      "loss": 0.82121164,
      "num_input_tokens_seen": 60551500,
      "step": 2813,
      "time_per_iteration": 2.639927387237549
    },
    {
      "auxiliary_loss_clip": 0.01198159,
      "auxiliary_loss_mlp": 0.01034863,
      "balance_loss_clip": 1.06062472,
      "balance_loss_mlp": 1.02557039,
      "epoch": 0.33836349425840195,
      "flos": 22122409524480.0,
      "grad_norm": 1.7815877198061947,
      "language_loss": 0.82917958,
      "learning_rate": 3.082889635231126e-06,
      "loss": 0.85150981,
      "num_input_tokens_seen": 60570160,
      "step": 2814,
      "time_per_iteration": 2.611776828765869
    },
    {
      "auxiliary_loss_clip": 0.01201732,
      "auxiliary_loss_mlp": 0.01035133,
      "balance_loss_clip": 1.06153905,
      "balance_loss_mlp": 1.02567375,
      "epoch": 0.33848373714904106,
      "flos": 27308090067840.0,
      "grad_norm": 2.360434601263472,
      "language_loss": 0.76195115,
      "learning_rate": 3.0822346433780925e-06,
      "loss": 0.78431976,
      "num_input_tokens_seen": 60590885,
      "step": 2815,
      "time_per_iteration": 2.6786949634552
    },
    {
      "auxiliary_loss_clip": 0.0120084,
      "auxiliary_loss_mlp": 0.01035534,
      "balance_loss_clip": 1.05931616,
      "balance_loss_mlp": 1.02458429,
      "epoch": 0.3386039800396802,
      "flos": 25848716394240.0,
      "grad_norm": 4.850977230214622,
      "language_loss": 0.86529851,
      "learning_rate": 3.0815794873509237e-06,
      "loss": 0.88766217,
      "num_input_tokens_seen": 60609170,
      "step": 2816,
      "time_per_iteration": 2.6359221935272217
    },
    {
      "auxiliary_loss_clip": 0.01200439,
      "auxiliary_loss_mlp": 0.01035321,
      "balance_loss_clip": 1.06199193,
      "balance_loss_mlp": 1.02596283,
      "epoch": 0.33872422293031923,
      "flos": 18880646146560.0,
      "grad_norm": 5.561355805404444,
      "language_loss": 0.72336137,
      "learning_rate": 3.0809241672490066e-06,
      "loss": 0.74571896,
      "num_input_tokens_seen": 60627340,
      "step": 2817,
      "time_per_iteration": 2.652111768722534
    },
    {
      "auxiliary_loss_clip": 0.01198446,
      "auxiliary_loss_mlp": 0.01027277,
      "balance_loss_clip": 1.0596869,
      "balance_loss_mlp": 1.01895678,
      "epoch": 0.33884446582095834,
      "flos": 23146977064320.0,
      "grad_norm": 1.8660187376522026,
      "language_loss": 0.85190684,
      "learning_rate": 3.080268683171753e-06,
      "loss": 0.87416404,
      "num_input_tokens_seen": 60647630,
      "step": 2818,
      "time_per_iteration": 2.6273646354675293
    },
    {
      "auxiliary_loss_clip": 0.01197573,
      "auxiliary_loss_mlp": 0.01033811,
      "balance_loss_clip": 1.05915058,
      "balance_loss_mlp": 1.02479839,
      "epoch": 0.33896470871159745,
      "flos": 15997342544640.0,
      "grad_norm": 2.083292367878838,
      "language_loss": 0.8953681,
      "learning_rate": 3.0796130352185985e-06,
      "loss": 0.91768193,
      "num_input_tokens_seen": 60664485,
      "step": 2819,
      "time_per_iteration": 2.749847173690796
    },
    {
      "auxiliary_loss_clip": 0.01201208,
      "auxiliary_loss_mlp": 0.01033339,
      "balance_loss_clip": 1.0608052,
      "balance_loss_mlp": 1.02378988,
      "epoch": 0.3390849516022365,
      "flos": 34495754112000.0,
      "grad_norm": 1.9306920235235787,
      "language_loss": 0.66820699,
      "learning_rate": 3.0789572234890057e-06,
      "loss": 0.69055247,
      "num_input_tokens_seen": 60686125,
      "step": 2820,
      "time_per_iteration": 2.7518410682678223
    },
    {
      "auxiliary_loss_clip": 0.01202479,
      "auxiliary_loss_mlp": 0.0103884,
      "balance_loss_clip": 1.06247687,
      "balance_loss_mlp": 1.02929711,
      "epoch": 0.3392051944928756,
      "flos": 16180307447040.0,
      "grad_norm": 2.2658713151752905,
      "language_loss": 0.77186656,
      "learning_rate": 3.0783012480824596e-06,
      "loss": 0.79427981,
      "num_input_tokens_seen": 60705270,
      "step": 2821,
      "time_per_iteration": 2.570250988006592
    },
    {
      "auxiliary_loss_clip": 0.01196822,
      "auxiliary_loss_mlp": 0.01031202,
      "balance_loss_clip": 1.05732131,
      "balance_loss_mlp": 1.02129602,
      "epoch": 0.33932543738351467,
      "flos": 17086656349440.0,
      "grad_norm": 1.9854631574518473,
      "language_loss": 0.73915875,
      "learning_rate": 3.077645109098471e-06,
      "loss": 0.76143903,
      "num_input_tokens_seen": 60721540,
      "step": 2822,
      "time_per_iteration": 2.6431915760040283
    },
    {
      "auxiliary_loss_clip": 0.01197066,
      "auxiliary_loss_mlp": 0.01032519,
      "balance_loss_clip": 1.059829,
      "balance_loss_mlp": 1.02323294,
      "epoch": 0.3394456802741538,
      "flos": 22126970551680.0,
      "grad_norm": 1.7536668798090291,
      "language_loss": 0.72255039,
      "learning_rate": 3.076988806636577e-06,
      "loss": 0.74484628,
      "num_input_tokens_seen": 60739300,
      "step": 2823,
      "time_per_iteration": 2.6293251514434814
    },
    {
      "auxiliary_loss_clip": 0.01201595,
      "auxiliary_loss_mlp": 0.01031356,
      "balance_loss_clip": 1.06127417,
      "balance_loss_mlp": 1.02056742,
      "epoch": 0.3395659231647929,
      "flos": 25226887121280.0,
      "grad_norm": 2.161952601068167,
      "language_loss": 0.89049989,
      "learning_rate": 3.0763323407963377e-06,
      "loss": 0.9128294,
      "num_input_tokens_seen": 60758910,
      "step": 2824,
      "time_per_iteration": 2.657714605331421
    },
    {
      "auxiliary_loss_clip": 0.0120063,
      "auxiliary_loss_mlp": 0.01033046,
      "balance_loss_clip": 1.06125903,
      "balance_loss_mlp": 1.02361059,
      "epoch": 0.33968616605543195,
      "flos": 29096477343360.0,
      "grad_norm": 1.8264755349110926,
      "language_loss": 0.80194223,
      "learning_rate": 3.075675711677337e-06,
      "loss": 0.82427907,
      "num_input_tokens_seen": 60779005,
      "step": 2825,
      "time_per_iteration": 2.6802570819854736
    },
    {
      "auxiliary_loss_clip": 0.01198392,
      "auxiliary_loss_mlp": 0.01029116,
      "balance_loss_clip": 1.06053054,
      "balance_loss_mlp": 1.02016926,
      "epoch": 0.33980640894607106,
      "flos": 21433966479360.0,
      "grad_norm": 2.1804950385840947,
      "language_loss": 0.78350842,
      "learning_rate": 3.0750189193791865e-06,
      "loss": 0.80578351,
      "num_input_tokens_seen": 60798590,
      "step": 2826,
      "time_per_iteration": 2.6459853649139404
    },
    {
      "auxiliary_loss_clip": 0.01200392,
      "auxiliary_loss_mlp": 0.01036252,
      "balance_loss_clip": 1.06099832,
      "balance_loss_mlp": 1.02563071,
      "epoch": 0.33992665183671017,
      "flos": 32490035596800.0,
      "grad_norm": 4.557605704856933,
      "language_loss": 0.70284212,
      "learning_rate": 3.0743619640015203e-06,
      "loss": 0.72520852,
      "num_input_tokens_seen": 60818840,
      "step": 2827,
      "time_per_iteration": 2.700941324234009
    },
    {
      "auxiliary_loss_clip": 0.01200237,
      "auxiliary_loss_mlp": 0.01032936,
      "balance_loss_clip": 1.06056929,
      "balance_loss_mlp": 1.02319646,
      "epoch": 0.3400468947273492,
      "flos": 17055414495360.0,
      "grad_norm": 2.3380451888490907,
      "language_loss": 0.92130423,
      "learning_rate": 3.073704845643999e-06,
      "loss": 0.94363594,
      "num_input_tokens_seen": 60835965,
      "step": 2828,
      "time_per_iteration": 3.4884843826293945
    },
    {
      "auxiliary_loss_clip": 0.01199373,
      "auxiliary_loss_mlp": 0.01032562,
      "balance_loss_clip": 1.05948782,
      "balance_loss_mlp": 1.02180314,
      "epoch": 0.34016713761798834,
      "flos": 16872988296960.0,
      "grad_norm": 3.217382618385608,
      "language_loss": 0.77488351,
      "learning_rate": 3.0730475644063063e-06,
      "loss": 0.79720294,
      "num_input_tokens_seen": 60851065,
      "step": 2829,
      "time_per_iteration": 2.579181671142578
    },
    {
      "auxiliary_loss_clip": 0.01195493,
      "auxiliary_loss_mlp": 0.01035562,
      "balance_loss_clip": 1.05860615,
      "balance_loss_mlp": 1.02620995,
      "epoch": 0.34028738050862745,
      "flos": 21907161273600.0,
      "grad_norm": 1.8937581798187515,
      "language_loss": 0.65036798,
      "learning_rate": 3.072390120388151e-06,
      "loss": 0.67267859,
      "num_input_tokens_seen": 60869390,
      "step": 2830,
      "time_per_iteration": 2.6751534938812256
    },
    {
      "auxiliary_loss_clip": 0.01201371,
      "auxiliary_loss_mlp": 0.01030666,
      "balance_loss_clip": 1.05937314,
      "balance_loss_mlp": 1.02050984,
      "epoch": 0.3404076233992665,
      "flos": 22746034477440.0,
      "grad_norm": 2.0238878227220973,
      "language_loss": 0.70710909,
      "learning_rate": 3.071732513689267e-06,
      "loss": 0.72942942,
      "num_input_tokens_seen": 60887925,
      "step": 2831,
      "time_per_iteration": 3.5724923610687256
    },
    {
      "auxiliary_loss_clip": 0.01203977,
      "auxiliary_loss_mlp": 0.01045372,
      "balance_loss_clip": 1.06332564,
      "balance_loss_mlp": 1.03484619,
      "epoch": 0.3405278662899056,
      "flos": 17052361839360.0,
      "grad_norm": 2.1829031551520903,
      "language_loss": 0.67005599,
      "learning_rate": 3.0710747444094134e-06,
      "loss": 0.69254947,
      "num_input_tokens_seen": 60905955,
      "step": 2832,
      "time_per_iteration": 4.74679708480835
    },
    {
      "auxiliary_loss_clip": 0.01201819,
      "auxiliary_loss_mlp": 0.01034525,
      "balance_loss_clip": 1.06157708,
      "balance_loss_mlp": 1.02332497,
      "epoch": 0.3406481091805447,
      "flos": 42813131783040.0,
      "grad_norm": 2.256739194573391,
      "language_loss": 0.65064436,
      "learning_rate": 3.070416812648372e-06,
      "loss": 0.67300779,
      "num_input_tokens_seen": 60929405,
      "step": 2833,
      "time_per_iteration": 2.811544179916382
    },
    {
      "auxiliary_loss_clip": 0.01203793,
      "auxiliary_loss_mlp": 0.01029635,
      "balance_loss_clip": 1.06449199,
      "balance_loss_mlp": 1.02042651,
      "epoch": 0.3407683520711838,
      "flos": 26761457917440.0,
      "grad_norm": 2.2985612197116962,
      "language_loss": 0.65145087,
      "learning_rate": 3.069758718505951e-06,
      "loss": 0.67378509,
      "num_input_tokens_seen": 60951145,
      "step": 2834,
      "time_per_iteration": 2.6898856163024902
    },
    {
      "auxiliary_loss_clip": 0.01199126,
      "auxiliary_loss_mlp": 0.01037964,
      "balance_loss_clip": 1.06142545,
      "balance_loss_mlp": 1.02824283,
      "epoch": 0.3408885949618229,
      "flos": 28767643309440.0,
      "grad_norm": 1.817315114145487,
      "language_loss": 0.80158585,
      "learning_rate": 3.0691004620819836e-06,
      "loss": 0.82395673,
      "num_input_tokens_seen": 60971275,
      "step": 2835,
      "time_per_iteration": 2.6747326850891113
    },
    {
      "auxiliary_loss_clip": 0.01100173,
      "auxiliary_loss_mlp": 0.01001915,
      "balance_loss_clip": 1.02865183,
      "balance_loss_mlp": 1.00024605,
      "epoch": 0.341008837852462,
      "flos": 63576252881280.0,
      "grad_norm": 0.7965773855218521,
      "language_loss": 0.60149747,
      "learning_rate": 3.0684420434763254e-06,
      "loss": 0.6225183,
      "num_input_tokens_seen": 61037460,
      "step": 2836,
      "time_per_iteration": 3.224681854248047
    },
    {
      "auxiliary_loss_clip": 0.01198743,
      "auxiliary_loss_mlp": 0.01037547,
      "balance_loss_clip": 1.06247842,
      "balance_loss_mlp": 1.0286839,
      "epoch": 0.34112908074310105,
      "flos": 20812173120000.0,
      "grad_norm": 2.3024040450172474,
      "language_loss": 0.7719804,
      "learning_rate": 3.06778346278886e-06,
      "loss": 0.79434323,
      "num_input_tokens_seen": 61056295,
      "step": 2837,
      "time_per_iteration": 2.566783905029297
    },
    {
      "auxiliary_loss_clip": 0.01200211,
      "auxiliary_loss_mlp": 0.01037616,
      "balance_loss_clip": 1.06228328,
      "balance_loss_mlp": 1.02702475,
      "epoch": 0.34124932363374016,
      "flos": 24976446520320.0,
      "grad_norm": 2.8783695487143226,
      "language_loss": 0.78949672,
      "learning_rate": 3.0671247201194906e-06,
      "loss": 0.81187499,
      "num_input_tokens_seen": 61078430,
      "step": 2838,
      "time_per_iteration": 2.7033841609954834
    },
    {
      "auxiliary_loss_clip": 0.01202899,
      "auxiliary_loss_mlp": 0.01036814,
      "balance_loss_clip": 1.06181502,
      "balance_loss_mlp": 1.02596641,
      "epoch": 0.3413695665243792,
      "flos": 28402970480640.0,
      "grad_norm": 1.7329228718266732,
      "language_loss": 0.75320059,
      "learning_rate": 3.066465815568151e-06,
      "loss": 0.77559775,
      "num_input_tokens_seen": 61099260,
      "step": 2839,
      "time_per_iteration": 2.7210652828216553
    },
    {
      "auxiliary_loss_clip": 0.01199413,
      "auxiliary_loss_mlp": 0.01037116,
      "balance_loss_clip": 1.06036592,
      "balance_loss_mlp": 1.02776957,
      "epoch": 0.34148980941501833,
      "flos": 25302012416640.0,
      "grad_norm": 1.8647478787410658,
      "language_loss": 0.68782049,
      "learning_rate": 3.0658067492347947e-06,
      "loss": 0.71018583,
      "num_input_tokens_seen": 61121900,
      "step": 2840,
      "time_per_iteration": 2.734476327896118
    },
    {
      "auxiliary_loss_clip": 0.01199896,
      "auxiliary_loss_mlp": 0.01037501,
      "balance_loss_clip": 1.06062341,
      "balance_loss_mlp": 1.02760696,
      "epoch": 0.34161005230565744,
      "flos": 17530081747200.0,
      "grad_norm": 2.156258278031593,
      "language_loss": 0.66858381,
      "learning_rate": 3.065147521219402e-06,
      "loss": 0.69095778,
      "num_input_tokens_seen": 61141155,
      "step": 2841,
      "time_per_iteration": 2.6282715797424316
    },
    {
      "auxiliary_loss_clip": 0.01204555,
      "auxiliary_loss_mlp": 0.01033412,
      "balance_loss_clip": 1.06459689,
      "balance_loss_mlp": 1.02431679,
      "epoch": 0.3417302951962965,
      "flos": 43650101566080.0,
      "grad_norm": 1.6323534564730822,
      "language_loss": 0.74642652,
      "learning_rate": 3.064488131621977e-06,
      "loss": 0.76880622,
      "num_input_tokens_seen": 61164480,
      "step": 2842,
      "time_per_iteration": 2.8221473693847656
    },
    {
      "auxiliary_loss_clip": 0.01197797,
      "auxiliary_loss_mlp": 0.01034315,
      "balance_loss_clip": 1.05949306,
      "balance_loss_mlp": 1.02472425,
      "epoch": 0.3418505380869356,
      "flos": 30882207012480.0,
      "grad_norm": 1.8314397871809565,
      "language_loss": 0.73880327,
      "learning_rate": 3.063828580542549e-06,
      "loss": 0.76112437,
      "num_input_tokens_seen": 61185675,
      "step": 2843,
      "time_per_iteration": 2.778714179992676
    },
    {
      "auxiliary_loss_clip": 0.01198353,
      "auxiliary_loss_mlp": 0.01035132,
      "balance_loss_clip": 1.06166899,
      "balance_loss_mlp": 1.02616704,
      "epoch": 0.3419707809775747,
      "flos": 19463871277440.0,
      "grad_norm": 1.9229058466737974,
      "language_loss": 0.73371661,
      "learning_rate": 3.0631688680811706e-06,
      "loss": 0.75605142,
      "num_input_tokens_seen": 61205300,
      "step": 2844,
      "time_per_iteration": 2.639674425125122
    },
    {
      "auxiliary_loss_clip": 0.01201786,
      "auxiliary_loss_mlp": 0.01029517,
      "balance_loss_clip": 1.06115317,
      "balance_loss_mlp": 1.02018893,
      "epoch": 0.3420910238682138,
      "flos": 28727818104960.0,
      "grad_norm": 2.1684759495150545,
      "language_loss": 0.75888801,
      "learning_rate": 3.062508994337921e-06,
      "loss": 0.78120106,
      "num_input_tokens_seen": 61224905,
      "step": 2845,
      "time_per_iteration": 2.714994430541992
    },
    {
      "auxiliary_loss_clip": 0.01199382,
      "auxiliary_loss_mlp": 0.01036847,
      "balance_loss_clip": 1.05823874,
      "balance_loss_mlp": 1.02679729,
      "epoch": 0.3422112667588529,
      "flos": 21397265758080.0,
      "grad_norm": 2.2183289467921705,
      "language_loss": 0.7938484,
      "learning_rate": 3.0618489594129013e-06,
      "loss": 0.81621063,
      "num_input_tokens_seen": 61243045,
      "step": 2846,
      "time_per_iteration": 2.6085548400878906
    },
    {
      "auxiliary_loss_clip": 0.01200405,
      "auxiliary_loss_mlp": 0.01029485,
      "balance_loss_clip": 1.06070375,
      "balance_loss_mlp": 1.01950693,
      "epoch": 0.342331509649492,
      "flos": 13881450038400.0,
      "grad_norm": 2.175547620265624,
      "language_loss": 0.71140355,
      "learning_rate": 3.061188763406239e-06,
      "loss": 0.73370242,
      "num_input_tokens_seen": 61259190,
      "step": 2847,
      "time_per_iteration": 2.612950086593628
    },
    {
      "auxiliary_loss_clip": 0.01201448,
      "auxiliary_loss_mlp": 0.01035833,
      "balance_loss_clip": 1.06281304,
      "balance_loss_mlp": 1.02562869,
      "epoch": 0.34245175254013105,
      "flos": 28621450955520.0,
      "grad_norm": 2.0586914369936484,
      "language_loss": 0.82213378,
      "learning_rate": 3.060528406418085e-06,
      "loss": 0.84450662,
      "num_input_tokens_seen": 61279040,
      "step": 2848,
      "time_per_iteration": 2.7271366119384766
    },
    {
      "auxiliary_loss_clip": 0.01199216,
      "auxiliary_loss_mlp": 0.01031534,
      "balance_loss_clip": 1.0605104,
      "balance_loss_mlp": 1.02172351,
      "epoch": 0.34257199543077016,
      "flos": 34127058960000.0,
      "grad_norm": 1.8042920725624827,
      "language_loss": 0.61942112,
      "learning_rate": 3.0598678885486145e-06,
      "loss": 0.64172864,
      "num_input_tokens_seen": 61301580,
      "step": 2849,
      "time_per_iteration": 2.7214198112487793
    },
    {
      "auxiliary_loss_clip": 0.01196534,
      "auxiliary_loss_mlp": 0.01029977,
      "balance_loss_clip": 1.05789542,
      "balance_loss_mlp": 1.02025521,
      "epoch": 0.34269223832140927,
      "flos": 19974018188160.0,
      "grad_norm": 1.980634171381577,
      "language_loss": 0.74429452,
      "learning_rate": 3.0592072098980282e-06,
      "loss": 0.7665596,
      "num_input_tokens_seen": 61321240,
      "step": 2850,
      "time_per_iteration": 2.58499813079834
    },
    {
      "auxiliary_loss_clip": 0.01198114,
      "auxiliary_loss_mlp": 0.0103524,
      "balance_loss_clip": 1.06019664,
      "balance_loss_mlp": 1.02572155,
      "epoch": 0.3428124812120483,
      "flos": 27235658292480.0,
      "grad_norm": 1.9818790074810804,
      "language_loss": 0.7310884,
      "learning_rate": 3.0585463705665514e-06,
      "loss": 0.7534219,
      "num_input_tokens_seen": 61341615,
      "step": 2851,
      "time_per_iteration": 2.7004964351654053
    },
    {
      "auxiliary_loss_clip": 0.01199552,
      "auxiliary_loss_mlp": 0.01033149,
      "balance_loss_clip": 1.06081772,
      "balance_loss_mlp": 1.02279556,
      "epoch": 0.34293272410268744,
      "flos": 24570871079040.0,
      "grad_norm": 2.173108139443355,
      "language_loss": 0.70398474,
      "learning_rate": 3.0578853706544304e-06,
      "loss": 0.7263118,
      "num_input_tokens_seen": 61359005,
      "step": 2852,
      "time_per_iteration": 2.6654322147369385
    },
    {
      "auxiliary_loss_clip": 0.01203061,
      "auxiliary_loss_mlp": 0.01031269,
      "balance_loss_clip": 1.06249726,
      "balance_loss_mlp": 1.02123725,
      "epoch": 0.34305296699332655,
      "flos": 21506865131520.0,
      "grad_norm": 2.0609226864943513,
      "language_loss": 0.65862548,
      "learning_rate": 3.0572242102619404e-06,
      "loss": 0.68096882,
      "num_input_tokens_seen": 61376160,
      "step": 2853,
      "time_per_iteration": 2.617591381072998
    },
    {
      "auxiliary_loss_clip": 0.01200568,
      "auxiliary_loss_mlp": 0.0103606,
      "balance_loss_clip": 1.06267548,
      "balance_loss_mlp": 1.02624345,
      "epoch": 0.3431732098839656,
      "flos": 24056665931520.0,
      "grad_norm": 1.8819710219038264,
      "language_loss": 0.80164969,
      "learning_rate": 3.0565628894893784e-06,
      "loss": 0.82401597,
      "num_input_tokens_seen": 61396795,
      "step": 2854,
      "time_per_iteration": 2.846308708190918
    },
    {
      "auxiliary_loss_clip": 0.01201097,
      "auxiliary_loss_mlp": 0.01032264,
      "balance_loss_clip": 1.06203854,
      "balance_loss_mlp": 1.02266204,
      "epoch": 0.3432934527746047,
      "flos": 16800879744000.0,
      "grad_norm": 1.7106132253144555,
      "language_loss": 0.74691921,
      "learning_rate": 3.0559014084370655e-06,
      "loss": 0.76925278,
      "num_input_tokens_seen": 61415320,
      "step": 2855,
      "time_per_iteration": 3.5948619842529297
    },
    {
      "auxiliary_loss_clip": 0.01201029,
      "auxiliary_loss_mlp": 0.01030155,
      "balance_loss_clip": 1.06093621,
      "balance_loss_mlp": 1.01948535,
      "epoch": 0.34341369566524377,
      "flos": 23439720908160.0,
      "grad_norm": 2.045340096002887,
      "language_loss": 0.78637981,
      "learning_rate": 3.055239767205349e-06,
      "loss": 0.80869162,
      "num_input_tokens_seen": 61437070,
      "step": 2856,
      "time_per_iteration": 2.6648643016815186
    },
    {
      "auxiliary_loss_clip": 0.01201863,
      "auxiliary_loss_mlp": 0.01039911,
      "balance_loss_clip": 1.06451869,
      "balance_loss_mlp": 1.03013039,
      "epoch": 0.3435339385558829,
      "flos": 17267466435840.0,
      "grad_norm": 1.8548848029885696,
      "language_loss": 0.78262305,
      "learning_rate": 3.054577965894599e-06,
      "loss": 0.80504078,
      "num_input_tokens_seen": 61453215,
      "step": 2857,
      "time_per_iteration": 2.5774219036102295
    },
    {
      "auxiliary_loss_clip": 0.01208028,
      "auxiliary_loss_mlp": 0.01038826,
      "balance_loss_clip": 1.06474328,
      "balance_loss_mlp": 1.02748299,
      "epoch": 0.343654181446522,
      "flos": 22199366413440.0,
      "grad_norm": 1.7019895383596977,
      "language_loss": 0.7003212,
      "learning_rate": 3.0539160046052094e-06,
      "loss": 0.72278976,
      "num_input_tokens_seen": 61472915,
      "step": 2858,
      "time_per_iteration": 4.503403902053833
    },
    {
      "auxiliary_loss_clip": 0.01204719,
      "auxiliary_loss_mlp": 0.01027382,
      "balance_loss_clip": 1.06158233,
      "balance_loss_mlp": 1.01637316,
      "epoch": 0.34377442433716104,
      "flos": 19901801894400.0,
      "grad_norm": 2.0303657249396165,
      "language_loss": 0.70502549,
      "learning_rate": 3.0532538834376003e-06,
      "loss": 0.72734648,
      "num_input_tokens_seen": 61492475,
      "step": 2859,
      "time_per_iteration": 3.566575288772583
    },
    {
      "auxiliary_loss_clip": 0.01203113,
      "auxiliary_loss_mlp": 0.01033198,
      "balance_loss_clip": 1.06227589,
      "balance_loss_mlp": 1.02313089,
      "epoch": 0.34389466722780015,
      "flos": 22197678474240.0,
      "grad_norm": 2.6762755829771847,
      "language_loss": 0.78513044,
      "learning_rate": 3.0525916024922143e-06,
      "loss": 0.80749351,
      "num_input_tokens_seen": 61511660,
      "step": 2860,
      "time_per_iteration": 2.6275596618652344
    },
    {
      "auxiliary_loss_clip": 0.01202789,
      "auxiliary_loss_mlp": 0.01037077,
      "balance_loss_clip": 1.06283176,
      "balance_loss_mlp": 1.02675366,
      "epoch": 0.34401491011843927,
      "flos": 18624567110400.0,
      "grad_norm": 3.8705964865250726,
      "language_loss": 0.84366643,
      "learning_rate": 3.0519291618695193e-06,
      "loss": 0.86606514,
      "num_input_tokens_seen": 61529060,
      "step": 2861,
      "time_per_iteration": 2.619967222213745
    },
    {
      "auxiliary_loss_clip": 0.01198646,
      "auxiliary_loss_mlp": 0.01040188,
      "balance_loss_clip": 1.06115007,
      "balance_loss_mlp": 1.03120518,
      "epoch": 0.3441351530090783,
      "flos": 17858197509120.0,
      "grad_norm": 1.7724329474155451,
      "language_loss": 0.75860655,
      "learning_rate": 3.0512665616700065e-06,
      "loss": 0.78099489,
      "num_input_tokens_seen": 61548125,
      "step": 2862,
      "time_per_iteration": 2.6046226024627686
    },
    {
      "auxiliary_loss_clip": 0.01199189,
      "auxiliary_loss_mlp": 0.0103973,
      "balance_loss_clip": 1.06120384,
      "balance_loss_mlp": 1.02919197,
      "epoch": 0.34425539589971743,
      "flos": 23112754381440.0,
      "grad_norm": 1.8326047108045649,
      "language_loss": 0.89287579,
      "learning_rate": 3.0506038019941933e-06,
      "loss": 0.91526496,
      "num_input_tokens_seen": 61568135,
      "step": 2863,
      "time_per_iteration": 2.6526458263397217
    },
    {
      "auxiliary_loss_clip": 0.01203219,
      "auxiliary_loss_mlp": 0.01033685,
      "balance_loss_clip": 1.0611372,
      "balance_loss_mlp": 1.02253854,
      "epoch": 0.34437563879035654,
      "flos": 21907699977600.0,
      "grad_norm": 2.978088032101895,
      "language_loss": 0.67819101,
      "learning_rate": 3.049940882942617e-06,
      "loss": 0.70056009,
      "num_input_tokens_seen": 61586920,
      "step": 2864,
      "time_per_iteration": 2.662815570831299
    },
    {
      "auxiliary_loss_clip": 0.01203723,
      "auxiliary_loss_mlp": 0.01034687,
      "balance_loss_clip": 1.0627439,
      "balance_loss_mlp": 1.02370799,
      "epoch": 0.3444958816809956,
      "flos": 23076915586560.0,
      "grad_norm": 2.8512816800618794,
      "language_loss": 0.80650783,
      "learning_rate": 3.0492778046158448e-06,
      "loss": 0.82889193,
      "num_input_tokens_seen": 61608340,
      "step": 2865,
      "time_per_iteration": 2.659628391265869
    },
    {
      "auxiliary_loss_clip": 0.01198556,
      "auxiliary_loss_mlp": 0.01035561,
      "balance_loss_clip": 1.06234753,
      "balance_loss_mlp": 1.02579832,
      "epoch": 0.3446161245716347,
      "flos": 21908633731200.0,
      "grad_norm": 3.1779327494743113,
      "language_loss": 0.77002639,
      "learning_rate": 3.0486145671144633e-06,
      "loss": 0.79236758,
      "num_input_tokens_seen": 61628130,
      "step": 2866,
      "time_per_iteration": 2.7151241302490234
    },
    {
      "auxiliary_loss_clip": 0.01205589,
      "auxiliary_loss_mlp": 0.01033502,
      "balance_loss_clip": 1.06530583,
      "balance_loss_mlp": 1.02225435,
      "epoch": 0.3447363674622738,
      "flos": 25112834461440.0,
      "grad_norm": 2.923321219930132,
      "language_loss": 0.76857817,
      "learning_rate": 3.047951170539086e-06,
      "loss": 0.79096907,
      "num_input_tokens_seen": 61647755,
      "step": 2867,
      "time_per_iteration": 2.6398415565490723
    },
    {
      "auxiliary_loss_clip": 0.01195372,
      "auxiliary_loss_mlp": 0.01036681,
      "balance_loss_clip": 1.06181347,
      "balance_loss_mlp": 1.02763891,
      "epoch": 0.3448566103529129,
      "flos": 11984684451840.0,
      "grad_norm": 1.8988403974093255,
      "language_loss": 0.84184128,
      "learning_rate": 3.047287614990349e-06,
      "loss": 0.86416185,
      "num_input_tokens_seen": 61665675,
      "step": 2868,
      "time_per_iteration": 2.6306684017181396
    },
    {
      "auxiliary_loss_clip": 0.01206313,
      "auxiliary_loss_mlp": 0.01040262,
      "balance_loss_clip": 1.06427538,
      "balance_loss_mlp": 1.02962208,
      "epoch": 0.344976853243552,
      "flos": 40187882465280.0,
      "grad_norm": 2.5025704887046105,
      "language_loss": 0.61868382,
      "learning_rate": 3.046623900568914e-06,
      "loss": 0.64114952,
      "num_input_tokens_seen": 61688240,
      "step": 2869,
      "time_per_iteration": 2.784860372543335
    },
    {
      "auxiliary_loss_clip": 0.01204411,
      "auxiliary_loss_mlp": 0.01037159,
      "balance_loss_clip": 1.06274903,
      "balance_loss_mlp": 1.02698421,
      "epoch": 0.34509709613419104,
      "flos": 28723652127360.0,
      "grad_norm": 2.232722365602426,
      "language_loss": 0.70281744,
      "learning_rate": 3.045960027375465e-06,
      "loss": 0.72523308,
      "num_input_tokens_seen": 61706075,
      "step": 2870,
      "time_per_iteration": 2.6315391063690186
    },
    {
      "auxiliary_loss_clip": 0.01200038,
      "auxiliary_loss_mlp": 0.01033269,
      "balance_loss_clip": 1.05943608,
      "balance_loss_mlp": 1.0219382,
      "epoch": 0.34521733902483015,
      "flos": 29967597982080.0,
      "grad_norm": 3.230341481289415,
      "language_loss": 0.8279922,
      "learning_rate": 3.045295995510711e-06,
      "loss": 0.85032523,
      "num_input_tokens_seen": 61723045,
      "step": 2871,
      "time_per_iteration": 2.6694352626800537
    },
    {
      "auxiliary_loss_clip": 0.01196811,
      "auxiliary_loss_mlp": 0.01031188,
      "balance_loss_clip": 1.061867,
      "balance_loss_mlp": 1.02119803,
      "epoch": 0.34533758191546926,
      "flos": 27923059843200.0,
      "grad_norm": 2.4505635291705787,
      "language_loss": 0.74375528,
      "learning_rate": 3.0446318050753865e-06,
      "loss": 0.76603532,
      "num_input_tokens_seen": 61743525,
      "step": 2872,
      "time_per_iteration": 2.6949565410614014
    },
    {
      "auxiliary_loss_clip": 0.01198262,
      "auxiliary_loss_mlp": 0.01029466,
      "balance_loss_clip": 1.06044436,
      "balance_loss_mlp": 1.02023911,
      "epoch": 0.3454578248061083,
      "flos": 27125879351040.0,
      "grad_norm": 2.0875647881032284,
      "language_loss": 0.77526343,
      "learning_rate": 3.0439674561702474e-06,
      "loss": 0.79754072,
      "num_input_tokens_seen": 61763025,
      "step": 2873,
      "time_per_iteration": 2.5962073802948
    },
    {
      "auxiliary_loss_clip": 0.01199592,
      "auxiliary_loss_mlp": 0.01033865,
      "balance_loss_clip": 1.06270981,
      "balance_loss_mlp": 1.0242033,
      "epoch": 0.3455780676967474,
      "flos": 19024899166080.0,
      "grad_norm": 5.02503608911005,
      "language_loss": 0.8760727,
      "learning_rate": 3.043302948896076e-06,
      "loss": 0.89840734,
      "num_input_tokens_seen": 61781630,
      "step": 2874,
      "time_per_iteration": 2.5883278846740723
    },
    {
      "auxiliary_loss_clip": 0.01201307,
      "auxiliary_loss_mlp": 0.01033943,
      "balance_loss_clip": 1.06316304,
      "balance_loss_mlp": 1.02336943,
      "epoch": 0.34569831058738654,
      "flos": 34496005507200.0,
      "grad_norm": 3.3466123319735726,
      "language_loss": 0.60522878,
      "learning_rate": 3.0426382833536756e-06,
      "loss": 0.62758136,
      "num_input_tokens_seen": 61804985,
      "step": 2875,
      "time_per_iteration": 2.752133369445801
    },
    {
      "auxiliary_loss_clip": 0.01199115,
      "auxiliary_loss_mlp": 0.01035863,
      "balance_loss_clip": 1.05976582,
      "balance_loss_mlp": 1.02616501,
      "epoch": 0.3458185534780256,
      "flos": 31138681098240.0,
      "grad_norm": 2.2191992406792473,
      "language_loss": 0.78258801,
      "learning_rate": 3.041973459643877e-06,
      "loss": 0.80493784,
      "num_input_tokens_seen": 61824440,
      "step": 2876,
      "time_per_iteration": 2.665708065032959
    },
    {
      "auxiliary_loss_clip": 0.01197078,
      "auxiliary_loss_mlp": 0.01031456,
      "balance_loss_clip": 1.05853844,
      "balance_loss_mlp": 1.02221751,
      "epoch": 0.3459387963686647,
      "flos": 32452508862720.0,
      "grad_norm": 2.0412075061122086,
      "language_loss": 0.66843486,
      "learning_rate": 3.0413084778675334e-06,
      "loss": 0.6907202,
      "num_input_tokens_seen": 61845690,
      "step": 2877,
      "time_per_iteration": 2.7427618503570557
    },
    {
      "auxiliary_loss_clip": 0.01196845,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.06006646,
      "balance_loss_mlp": 1.01946092,
      "epoch": 0.3460590392593038,
      "flos": 24675658030080.0,
      "grad_norm": 2.271527976173135,
      "language_loss": 0.84092659,
      "learning_rate": 3.0406433381255214e-06,
      "loss": 0.86317909,
      "num_input_tokens_seen": 61863725,
      "step": 2878,
      "time_per_iteration": 2.7313735485076904
    },
    {
      "auxiliary_loss_clip": 0.01202897,
      "auxiliary_loss_mlp": 0.0103115,
      "balance_loss_clip": 1.06443763,
      "balance_loss_mlp": 1.02062333,
      "epoch": 0.34617928214994287,
      "flos": 18807316531200.0,
      "grad_norm": 2.279456432431382,
      "language_loss": 0.82033831,
      "learning_rate": 3.0399780405187425e-06,
      "loss": 0.84267879,
      "num_input_tokens_seen": 61882720,
      "step": 2879,
      "time_per_iteration": 2.6329195499420166
    },
    {
      "auxiliary_loss_clip": 0.01194203,
      "auxiliary_loss_mlp": 0.01036136,
      "balance_loss_clip": 1.05903673,
      "balance_loss_mlp": 1.02696288,
      "epoch": 0.346299525040582,
      "flos": 24857653265280.0,
      "grad_norm": 2.2145663995812908,
      "language_loss": 0.78631115,
      "learning_rate": 3.0393125851481216e-06,
      "loss": 0.80861449,
      "num_input_tokens_seen": 61902595,
      "step": 2880,
      "time_per_iteration": 2.673861265182495
    },
    {
      "auxiliary_loss_clip": 0.01198479,
      "auxiliary_loss_mlp": 0.01035824,
      "balance_loss_clip": 1.06046486,
      "balance_loss_mlp": 1.0268954,
      "epoch": 0.3464197679312211,
      "flos": 16434914025600.0,
      "grad_norm": 2.279317205095691,
      "language_loss": 0.86346352,
      "learning_rate": 3.038646972114608e-06,
      "loss": 0.88580656,
      "num_input_tokens_seen": 61918920,
      "step": 2881,
      "time_per_iteration": 2.580656051635742
    },
    {
      "auxiliary_loss_clip": 0.01195634,
      "auxiliary_loss_mlp": 0.01034869,
      "balance_loss_clip": 1.05922389,
      "balance_loss_mlp": 1.02480221,
      "epoch": 0.34654001082186014,
      "flos": 22382474970240.0,
      "grad_norm": 2.139290616481118,
      "language_loss": 0.67400849,
      "learning_rate": 3.037981201519174e-06,
      "loss": 0.6963135,
      "num_input_tokens_seen": 61939520,
      "step": 2882,
      "time_per_iteration": 3.493849754333496
    },
    {
      "auxiliary_loss_clip": 0.01201907,
      "auxiliary_loss_mlp": 0.01036717,
      "balance_loss_clip": 1.06394219,
      "balance_loss_mlp": 1.02694786,
      "epoch": 0.34666025371249926,
      "flos": 19573901614080.0,
      "grad_norm": 4.357420572711419,
      "language_loss": 0.71344638,
      "learning_rate": 3.0373152734628175e-06,
      "loss": 0.73583257,
      "num_input_tokens_seen": 61957800,
      "step": 2883,
      "time_per_iteration": 2.6217851638793945
    },
    {
      "auxiliary_loss_clip": 0.01193874,
      "auxiliary_loss_mlp": 0.0103132,
      "balance_loss_clip": 1.05662274,
      "balance_loss_mlp": 1.02183652,
      "epoch": 0.34678049660313837,
      "flos": 15267637751040.0,
      "grad_norm": 2.181476176401931,
      "language_loss": 0.75893611,
      "learning_rate": 3.0366491880465584e-06,
      "loss": 0.78118801,
      "num_input_tokens_seen": 61975820,
      "step": 2884,
      "time_per_iteration": 2.590257167816162
    },
    {
      "auxiliary_loss_clip": 0.01205327,
      "auxiliary_loss_mlp": 0.0103248,
      "balance_loss_clip": 1.06415987,
      "balance_loss_mlp": 1.02237082,
      "epoch": 0.3469007394937774,
      "flos": 21181550630400.0,
      "grad_norm": 1.6307965742309947,
      "language_loss": 0.82059038,
      "learning_rate": 3.035982945371443e-06,
      "loss": 0.84296846,
      "num_input_tokens_seen": 61997515,
      "step": 2885,
      "time_per_iteration": 4.455843925476074
    },
    {
      "auxiliary_loss_clip": 0.01201363,
      "auxiliary_loss_mlp": 0.01033394,
      "balance_loss_clip": 1.06381881,
      "balance_loss_mlp": 1.02413785,
      "epoch": 0.34702098238441653,
      "flos": 22375471818240.0,
      "grad_norm": 3.4669426924064455,
      "language_loss": 0.85286152,
      "learning_rate": 3.035316545538537e-06,
      "loss": 0.87520903,
      "num_input_tokens_seen": 62016310,
      "step": 2886,
      "time_per_iteration": 2.647310733795166
    },
    {
      "auxiliary_loss_clip": 0.01197768,
      "auxiliary_loss_mlp": 0.01030881,
      "balance_loss_clip": 1.06078112,
      "balance_loss_mlp": 1.02165449,
      "epoch": 0.3471412252750556,
      "flos": 22929430343040.0,
      "grad_norm": 2.084154778505068,
      "language_loss": 0.79232669,
      "learning_rate": 3.034649988648935e-06,
      "loss": 0.81461316,
      "num_input_tokens_seen": 62036075,
      "step": 2887,
      "time_per_iteration": 2.5900917053222656
    },
    {
      "auxiliary_loss_clip": 0.01195933,
      "auxiliary_loss_mlp": 0.01031102,
      "balance_loss_clip": 1.05853033,
      "balance_loss_mlp": 1.0213809,
      "epoch": 0.3472614681656947,
      "flos": 21324259365120.0,
      "grad_norm": 2.1653083804098885,
      "language_loss": 0.80717224,
      "learning_rate": 3.033983274803752e-06,
      "loss": 0.82944262,
      "num_input_tokens_seen": 62055865,
      "step": 2888,
      "time_per_iteration": 2.6461007595062256
    },
    {
      "auxiliary_loss_clip": 0.01196836,
      "auxiliary_loss_mlp": 0.01034211,
      "balance_loss_clip": 1.06057858,
      "balance_loss_mlp": 1.02502012,
      "epoch": 0.3473817110563338,
      "flos": 23475739271040.0,
      "grad_norm": 2.2380680952563576,
      "language_loss": 0.72680402,
      "learning_rate": 3.0333164041041283e-06,
      "loss": 0.74911445,
      "num_input_tokens_seen": 62072180,
      "step": 2889,
      "time_per_iteration": 2.6975185871124268
    },
    {
      "auxiliary_loss_clip": 0.01195487,
      "auxiliary_loss_mlp": 0.01031996,
      "balance_loss_clip": 1.06138659,
      "balance_loss_mlp": 1.02305579,
      "epoch": 0.34750195394697286,
      "flos": 22346025644160.0,
      "grad_norm": 1.96945936040373,
      "language_loss": 0.71647716,
      "learning_rate": 3.032649376651228e-06,
      "loss": 0.73875201,
      "num_input_tokens_seen": 62091600,
      "step": 2890,
      "time_per_iteration": 2.638101100921631
    },
    {
      "auxiliary_loss_clip": 0.01199995,
      "auxiliary_loss_mlp": 0.01031714,
      "balance_loss_clip": 1.06046832,
      "balance_loss_mlp": 1.02199805,
      "epoch": 0.347622196837612,
      "flos": 29095004885760.0,
      "grad_norm": 1.8137606986322754,
      "language_loss": 0.75666857,
      "learning_rate": 3.031982192546238e-06,
      "loss": 0.77898568,
      "num_input_tokens_seen": 62114695,
      "step": 2891,
      "time_per_iteration": 2.6570799350738525
    },
    {
      "auxiliary_loss_clip": 0.0119695,
      "auxiliary_loss_mlp": 0.01031421,
      "balance_loss_clip": 1.06075609,
      "balance_loss_mlp": 1.02218795,
      "epoch": 0.3477424397282511,
      "flos": 22455732758400.0,
      "grad_norm": 2.0683516455244395,
      "language_loss": 0.94670492,
      "learning_rate": 3.0313148518903696e-06,
      "loss": 0.96898872,
      "num_input_tokens_seen": 62134520,
      "step": 2892,
      "time_per_iteration": 2.6315999031066895
    },
    {
      "auxiliary_loss_clip": 0.01196948,
      "auxiliary_loss_mlp": 0.01035434,
      "balance_loss_clip": 1.05918407,
      "balance_loss_mlp": 1.02605247,
      "epoch": 0.34786268261889014,
      "flos": 15778790242560.0,
      "grad_norm": 2.774885691743459,
      "language_loss": 0.81566393,
      "learning_rate": 3.030647354784859e-06,
      "loss": 0.83798778,
      "num_input_tokens_seen": 62151560,
      "step": 2893,
      "time_per_iteration": 2.60758638381958
    },
    {
      "auxiliary_loss_clip": 0.01192191,
      "auxiliary_loss_mlp": 0.01039182,
      "balance_loss_clip": 1.05767918,
      "balance_loss_mlp": 1.02962744,
      "epoch": 0.34798292550952925,
      "flos": 20777627214720.0,
      "grad_norm": 2.102856436565301,
      "language_loss": 0.77327049,
      "learning_rate": 3.029979701330964e-06,
      "loss": 0.7955842,
      "num_input_tokens_seen": 62170985,
      "step": 2894,
      "time_per_iteration": 2.6363208293914795
    },
    {
      "auxiliary_loss_clip": 0.01198324,
      "auxiliary_loss_mlp": 0.01031182,
      "balance_loss_clip": 1.06259918,
      "balance_loss_mlp": 1.02204442,
      "epoch": 0.34810316840016836,
      "flos": 19937820257280.0,
      "grad_norm": 2.9822530941017336,
      "language_loss": 0.80251592,
      "learning_rate": 3.029311891629966e-06,
      "loss": 0.82481098,
      "num_input_tokens_seen": 62189440,
      "step": 2895,
      "time_per_iteration": 2.6248183250427246
    },
    {
      "auxiliary_loss_clip": 0.01196788,
      "auxiliary_loss_mlp": 0.0103426,
      "balance_loss_clip": 1.0599525,
      "balance_loss_mlp": 1.02527213,
      "epoch": 0.3482234112908074,
      "flos": 23623296341760.0,
      "grad_norm": 1.9030460185043043,
      "language_loss": 0.7451787,
      "learning_rate": 3.0286439257831744e-06,
      "loss": 0.76748919,
      "num_input_tokens_seen": 62208910,
      "step": 2896,
      "time_per_iteration": 2.595308303833008
    },
    {
      "auxiliary_loss_clip": 0.01200877,
      "auxiliary_loss_mlp": 0.01034514,
      "balance_loss_clip": 1.05995524,
      "balance_loss_mlp": 1.02368939,
      "epoch": 0.3483436541814465,
      "flos": 23986712194560.0,
      "grad_norm": 2.144535339448348,
      "language_loss": 0.71866965,
      "learning_rate": 3.0279758038919156e-06,
      "loss": 0.74102354,
      "num_input_tokens_seen": 62227135,
      "step": 2897,
      "time_per_iteration": 2.663020372390747
    },
    {
      "auxiliary_loss_clip": 0.01198003,
      "auxiliary_loss_mlp": 0.01032587,
      "balance_loss_clip": 1.06022382,
      "balance_loss_mlp": 1.02331257,
      "epoch": 0.34846389707208564,
      "flos": 22638338524800.0,
      "grad_norm": 1.8484947209488132,
      "language_loss": 0.78296542,
      "learning_rate": 3.0273075260575455e-06,
      "loss": 0.80527127,
      "num_input_tokens_seen": 62246035,
      "step": 2898,
      "time_per_iteration": 2.6042163372039795
    },
    {
      "auxiliary_loss_clip": 0.01202411,
      "auxiliary_loss_mlp": 0.01038043,
      "balance_loss_clip": 1.06159699,
      "balance_loss_mlp": 1.02796364,
      "epoch": 0.3485841399627247,
      "flos": 21792857218560.0,
      "grad_norm": 2.5511699549573503,
      "language_loss": 0.80916035,
      "learning_rate": 3.0266390923814396e-06,
      "loss": 0.8315649,
      "num_input_tokens_seen": 62264095,
      "step": 2899,
      "time_per_iteration": 2.672903299331665
    },
    {
      "auxiliary_loss_clip": 0.01200677,
      "auxiliary_loss_mlp": 0.01037527,
      "balance_loss_clip": 1.06289136,
      "balance_loss_mlp": 1.02781117,
      "epoch": 0.3487043828533638,
      "flos": 17019036996480.0,
      "grad_norm": 3.157022507472144,
      "language_loss": 0.82144547,
      "learning_rate": 3.0259705029650008e-06,
      "loss": 0.84382749,
      "num_input_tokens_seen": 62282025,
      "step": 2900,
      "time_per_iteration": 2.6058592796325684
    },
    {
      "auxiliary_loss_clip": 0.01197212,
      "auxiliary_loss_mlp": 0.01033381,
      "balance_loss_clip": 1.0600934,
      "balance_loss_mlp": 1.02374315,
      "epoch": 0.34882462574400286,
      "flos": 22601135013120.0,
      "grad_norm": 3.4811920589975935,
      "language_loss": 0.72957063,
      "learning_rate": 3.025301757909652e-06,
      "loss": 0.75187659,
      "num_input_tokens_seen": 62302220,
      "step": 2901,
      "time_per_iteration": 2.6563990116119385
    },
    {
      "auxiliary_loss_clip": 0.01199356,
      "auxiliary_loss_mlp": 0.01034698,
      "balance_loss_clip": 1.05940247,
      "balance_loss_mlp": 1.02485669,
      "epoch": 0.34894486863464197,
      "flos": 29861518141440.0,
      "grad_norm": 1.4907933281498547,
      "language_loss": 0.8044976,
      "learning_rate": 3.024632857316842e-06,
      "loss": 0.82683814,
      "num_input_tokens_seen": 62323535,
      "step": 2902,
      "time_per_iteration": 2.6939163208007812
    },
    {
      "auxiliary_loss_clip": 0.01200125,
      "auxiliary_loss_mlp": 0.0103374,
      "balance_loss_clip": 1.0618155,
      "balance_loss_mlp": 1.02358961,
      "epoch": 0.3490651115252811,
      "flos": 22122265870080.0,
      "grad_norm": 2.174610545449714,
      "language_loss": 0.77259207,
      "learning_rate": 3.0239638012880412e-06,
      "loss": 0.7949307,
      "num_input_tokens_seen": 62343430,
      "step": 2903,
      "time_per_iteration": 2.661583185195923
    },
    {
      "auxiliary_loss_clip": 0.0119835,
      "auxiliary_loss_mlp": 0.01029496,
      "balance_loss_clip": 1.05924511,
      "balance_loss_mlp": 1.01907682,
      "epoch": 0.34918535441592014,
      "flos": 12676682943360.0,
      "grad_norm": 2.6128206406428323,
      "language_loss": 0.8128432,
      "learning_rate": 3.0232945899247466e-06,
      "loss": 0.83512163,
      "num_input_tokens_seen": 62360365,
      "step": 2904,
      "time_per_iteration": 2.5679454803466797
    },
    {
      "auxiliary_loss_clip": 0.01201873,
      "auxiliary_loss_mlp": 0.01035174,
      "balance_loss_clip": 1.06131542,
      "balance_loss_mlp": 1.02484465,
      "epoch": 0.34930559730655925,
      "flos": 23185617120000.0,
      "grad_norm": 1.9199817837179294,
      "language_loss": 0.77466309,
      "learning_rate": 3.022625223328476e-06,
      "loss": 0.79703355,
      "num_input_tokens_seen": 62382105,
      "step": 2905,
      "time_per_iteration": 2.67364239692688
    },
    {
      "auxiliary_loss_clip": 0.01202281,
      "auxiliary_loss_mlp": 0.01031887,
      "balance_loss_clip": 1.06256998,
      "balance_loss_mlp": 1.02097964,
      "epoch": 0.34942584019719836,
      "flos": 22855023319680.0,
      "grad_norm": 4.118199912850016,
      "language_loss": 0.691607,
      "learning_rate": 3.0219557016007723e-06,
      "loss": 0.71394873,
      "num_input_tokens_seen": 62402235,
      "step": 2906,
      "time_per_iteration": 2.6054859161376953
    },
    {
      "auxiliary_loss_clip": 0.01194374,
      "auxiliary_loss_mlp": 0.01035807,
      "balance_loss_clip": 1.05806017,
      "balance_loss_mlp": 1.0260613,
      "epoch": 0.3495460830878374,
      "flos": 24426043441920.0,
      "grad_norm": 1.9213636645644792,
      "language_loss": 0.70160669,
      "learning_rate": 3.021286024843202e-06,
      "loss": 0.72390854,
      "num_input_tokens_seen": 62420430,
      "step": 2907,
      "time_per_iteration": 2.6603660583496094
    },
    {
      "auxiliary_loss_clip": 0.01094226,
      "auxiliary_loss_mlp": 0.01006428,
      "balance_loss_clip": 1.02318609,
      "balance_loss_mlp": 1.00481915,
      "epoch": 0.3496663259784765,
      "flos": 70008749389440.0,
      "grad_norm": 1.0577417984271162,
      "language_loss": 0.64786768,
      "learning_rate": 3.0206161931573526e-06,
      "loss": 0.66887426,
      "num_input_tokens_seen": 62472980,
      "step": 2908,
      "time_per_iteration": 3.973095178604126
    },
    {
      "auxiliary_loss_clip": 0.01194846,
      "auxiliary_loss_mlp": 0.0103386,
      "balance_loss_clip": 1.05710816,
      "balance_loss_mlp": 1.0250262,
      "epoch": 0.34978656886911563,
      "flos": 28692805322880.0,
      "grad_norm": 3.1475476923309036,
      "language_loss": 0.92896962,
      "learning_rate": 3.0199462066448388e-06,
      "loss": 0.95125675,
      "num_input_tokens_seen": 62495175,
      "step": 2909,
      "time_per_iteration": 2.671382188796997
    },
    {
      "auxiliary_loss_clip": 0.01197664,
      "auxiliary_loss_mlp": 0.01033112,
      "balance_loss_clip": 1.0595181,
      "balance_loss_mlp": 1.02315247,
      "epoch": 0.3499068117597547,
      "flos": 21142156389120.0,
      "grad_norm": 1.9623037019081473,
      "language_loss": 0.69302309,
      "learning_rate": 3.019276065407296e-06,
      "loss": 0.71533084,
      "num_input_tokens_seen": 62514295,
      "step": 2910,
      "time_per_iteration": 2.6610350608825684
    },
    {
      "auxiliary_loss_clip": 0.01202959,
      "auxiliary_loss_mlp": 0.01039274,
      "balance_loss_clip": 1.06269968,
      "balance_loss_mlp": 1.02901614,
      "epoch": 0.3500270546503938,
      "flos": 22782699285120.0,
      "grad_norm": 1.8695853600358672,
      "language_loss": 0.80591267,
      "learning_rate": 3.018605769546385e-06,
      "loss": 0.82833499,
      "num_input_tokens_seen": 62534850,
      "step": 2911,
      "time_per_iteration": 2.628981113433838
    },
    {
      "auxiliary_loss_clip": 0.01200288,
      "auxiliary_loss_mlp": 0.01033533,
      "balance_loss_clip": 1.06023955,
      "balance_loss_mlp": 1.02317369,
      "epoch": 0.3501472975410329,
      "flos": 22894058424960.0,
      "grad_norm": 1.8420358334201405,
      "language_loss": 0.79061782,
      "learning_rate": 3.017935319163788e-06,
      "loss": 0.81295598,
      "num_input_tokens_seen": 62553810,
      "step": 2912,
      "time_per_iteration": 4.483952522277832
    },
    {
      "auxiliary_loss_clip": 0.01198149,
      "auxiliary_loss_mlp": 0.01039243,
      "balance_loss_clip": 1.05805695,
      "balance_loss_mlp": 1.02776313,
      "epoch": 0.35026754043167196,
      "flos": 25446588658560.0,
      "grad_norm": 1.9630361566387706,
      "language_loss": 0.7086069,
      "learning_rate": 3.017264714361213e-06,
      "loss": 0.73098075,
      "num_input_tokens_seen": 62573460,
      "step": 2913,
      "time_per_iteration": 2.7099130153656006
    },
    {
      "auxiliary_loss_clip": 0.01198885,
      "auxiliary_loss_mlp": 0.01041013,
      "balance_loss_clip": 1.05929267,
      "balance_loss_mlp": 1.03009343,
      "epoch": 0.3503877833223111,
      "flos": 19573757959680.0,
      "grad_norm": 2.069037463136162,
      "language_loss": 0.82112598,
      "learning_rate": 3.016593955240389e-06,
      "loss": 0.84352493,
      "num_input_tokens_seen": 62592150,
      "step": 2914,
      "time_per_iteration": 2.6201696395874023
    },
    {
      "auxiliary_loss_clip": 0.01093194,
      "auxiliary_loss_mlp": 0.01005978,
      "balance_loss_clip": 1.02235556,
      "balance_loss_mlp": 1.00427294,
      "epoch": 0.3505080262129502,
      "flos": 65072075880960.0,
      "grad_norm": 0.8262256650207871,
      "language_loss": 0.63663614,
      "learning_rate": 3.015923041903071e-06,
      "loss": 0.65762794,
      "num_input_tokens_seen": 62658275,
      "step": 2915,
      "time_per_iteration": 3.2363197803497314
    },
    {
      "auxiliary_loss_clip": 0.01198073,
      "auxiliary_loss_mlp": 0.0103263,
      "balance_loss_clip": 1.06030071,
      "balance_loss_mlp": 1.02339137,
      "epoch": 0.35062826910358924,
      "flos": 29314562768640.0,
      "grad_norm": 1.8864020929208851,
      "language_loss": 0.83894539,
      "learning_rate": 3.0152519744510347e-06,
      "loss": 0.86125243,
      "num_input_tokens_seen": 62678075,
      "step": 2916,
      "time_per_iteration": 2.691889524459839
    },
    {
      "auxiliary_loss_clip": 0.01197299,
      "auxiliary_loss_mlp": 0.01039258,
      "balance_loss_clip": 1.05826557,
      "balance_loss_mlp": 1.02900577,
      "epoch": 0.35074851199422835,
      "flos": 23987717775360.0,
      "grad_norm": 1.74589274633142,
      "language_loss": 0.82839429,
      "learning_rate": 3.014580752986081e-06,
      "loss": 0.85075986,
      "num_input_tokens_seen": 62696950,
      "step": 2917,
      "time_per_iteration": 2.631910562515259
    },
    {
      "auxiliary_loss_clip": 0.01199322,
      "auxiliary_loss_mlp": 0.01030791,
      "balance_loss_clip": 1.06130075,
      "balance_loss_mlp": 1.02119434,
      "epoch": 0.3508687548848674,
      "flos": 15224436668160.0,
      "grad_norm": 2.0054412236269967,
      "language_loss": 0.7853049,
      "learning_rate": 3.0139093776100345e-06,
      "loss": 0.8076061,
      "num_input_tokens_seen": 62713540,
      "step": 2918,
      "time_per_iteration": 2.604372262954712
    },
    {
      "auxiliary_loss_clip": 0.0119387,
      "auxiliary_loss_mlp": 0.0103049,
      "balance_loss_clip": 1.05649614,
      "balance_loss_mlp": 1.02102435,
      "epoch": 0.3509889977755065,
      "flos": 21361750185600.0,
      "grad_norm": 1.676894165131111,
      "language_loss": 0.75340879,
      "learning_rate": 3.013237848424741e-06,
      "loss": 0.77565241,
      "num_input_tokens_seen": 62732925,
      "step": 2919,
      "time_per_iteration": 2.6157186031341553
    },
    {
      "auxiliary_loss_clip": 0.01200313,
      "auxiliary_loss_mlp": 0.01037545,
      "balance_loss_clip": 1.06086564,
      "balance_loss_mlp": 1.02735221,
      "epoch": 0.35110924066614563,
      "flos": 19135360465920.0,
      "grad_norm": 2.2235793725149886,
      "language_loss": 0.75042403,
      "learning_rate": 3.012566165532072e-06,
      "loss": 0.77280259,
      "num_input_tokens_seen": 62751715,
      "step": 2920,
      "time_per_iteration": 2.5954649448394775
    },
    {
      "auxiliary_loss_clip": 0.01197975,
      "auxiliary_loss_mlp": 0.01027544,
      "balance_loss_clip": 1.05875969,
      "balance_loss_mlp": 1.01783407,
      "epoch": 0.3512294835567847,
      "flos": 21980885938560.0,
      "grad_norm": 3.260613344574021,
      "language_loss": 0.76474607,
      "learning_rate": 3.0118943290339207e-06,
      "loss": 0.78700131,
      "num_input_tokens_seen": 62771925,
      "step": 2921,
      "time_per_iteration": 2.5841262340545654
    },
    {
      "auxiliary_loss_clip": 0.01196938,
      "auxiliary_loss_mlp": 0.01034477,
      "balance_loss_clip": 1.05775273,
      "balance_loss_mlp": 1.02424288,
      "epoch": 0.3513497264474238,
      "flos": 17817294896640.0,
      "grad_norm": 2.037431189335455,
      "language_loss": 0.67966324,
      "learning_rate": 3.011222339032204e-06,
      "loss": 0.70197737,
      "num_input_tokens_seen": 62790075,
      "step": 2922,
      "time_per_iteration": 2.6082212924957275
    },
    {
      "auxiliary_loss_clip": 0.01197837,
      "auxiliary_loss_mlp": 0.01033348,
      "balance_loss_clip": 1.06029582,
      "balance_loss_mlp": 1.02371633,
      "epoch": 0.3514699693380629,
      "flos": 26943417239040.0,
      "grad_norm": 1.9444078179324955,
      "language_loss": 0.69718498,
      "learning_rate": 3.0105501956288626e-06,
      "loss": 0.71949685,
      "num_input_tokens_seen": 62810545,
      "step": 2923,
      "time_per_iteration": 2.6671411991119385
    },
    {
      "auxiliary_loss_clip": 0.01202039,
      "auxiliary_loss_mlp": 0.0102975,
      "balance_loss_clip": 1.0613296,
      "balance_loss_mlp": 1.02006388,
      "epoch": 0.35159021222870196,
      "flos": 15267565923840.0,
      "grad_norm": 1.9278579069588242,
      "language_loss": 0.7308073,
      "learning_rate": 3.0098778989258602e-06,
      "loss": 0.75312519,
      "num_input_tokens_seen": 62829155,
      "step": 2924,
      "time_per_iteration": 2.67402982711792
    },
    {
      "auxiliary_loss_clip": 0.0119968,
      "auxiliary_loss_mlp": 0.01037679,
      "balance_loss_clip": 1.06111336,
      "balance_loss_mlp": 1.02791619,
      "epoch": 0.35171045511934107,
      "flos": 13984154000640.0,
      "grad_norm": 2.0626413032590083,
      "language_loss": 0.88290787,
      "learning_rate": 3.009205449025183e-06,
      "loss": 0.90528142,
      "num_input_tokens_seen": 62845350,
      "step": 2925,
      "time_per_iteration": 2.6147639751434326
    },
    {
      "auxiliary_loss_clip": 0.0119576,
      "auxiliary_loss_mlp": 0.01037785,
      "balance_loss_clip": 1.05937648,
      "balance_loss_mlp": 1.02857637,
      "epoch": 0.3518306980099802,
      "flos": 14283434119680.0,
      "grad_norm": 1.892794479483914,
      "language_loss": 0.63485146,
      "learning_rate": 3.008532846028842e-06,
      "loss": 0.65718687,
      "num_input_tokens_seen": 62862110,
      "step": 2926,
      "time_per_iteration": 2.646303176879883
    },
    {
      "auxiliary_loss_clip": 0.01200606,
      "auxiliary_loss_mlp": 0.01038243,
      "balance_loss_clip": 1.06052995,
      "balance_loss_mlp": 1.02790177,
      "epoch": 0.35195094090061924,
      "flos": 27052872958080.0,
      "grad_norm": 2.8417562144492248,
      "language_loss": 0.7243098,
      "learning_rate": 3.0078600900388694e-06,
      "loss": 0.74669826,
      "num_input_tokens_seen": 62882415,
      "step": 2927,
      "time_per_iteration": 2.7676169872283936
    },
    {
      "auxiliary_loss_clip": 0.01196877,
      "auxiliary_loss_mlp": 0.01033253,
      "balance_loss_clip": 1.05852854,
      "balance_loss_mlp": 1.02385902,
      "epoch": 0.35207118379125835,
      "flos": 25629266252160.0,
      "grad_norm": 2.009809211727915,
      "language_loss": 0.73959816,
      "learning_rate": 3.007187181157323e-06,
      "loss": 0.76189947,
      "num_input_tokens_seen": 62902425,
      "step": 2928,
      "time_per_iteration": 2.7482028007507324
    },
    {
      "auxiliary_loss_clip": 0.01194993,
      "auxiliary_loss_mlp": 0.01033543,
      "balance_loss_clip": 1.05728269,
      "balance_loss_mlp": 1.02389348,
      "epoch": 0.35219142668189746,
      "flos": 18004713085440.0,
      "grad_norm": 3.392227087574368,
      "language_loss": 0.680089,
      "learning_rate": 3.006514119486282e-06,
      "loss": 0.7023744,
      "num_input_tokens_seen": 62919255,
      "step": 2929,
      "time_per_iteration": 2.640655994415283
    },
    {
      "auxiliary_loss_clip": 0.0119515,
      "auxiliary_loss_mlp": 0.01027965,
      "balance_loss_clip": 1.057935,
      "balance_loss_mlp": 1.0181601,
      "epoch": 0.3523116695725365,
      "flos": 14028109269120.0,
      "grad_norm": 3.0478417840258154,
      "language_loss": 0.69539785,
      "learning_rate": 3.005840905127849e-06,
      "loss": 0.71762908,
      "num_input_tokens_seen": 62936160,
      "step": 2930,
      "time_per_iteration": 2.6601767539978027
    },
    {
      "auxiliary_loss_clip": 0.01198464,
      "auxiliary_loss_mlp": 0.01036001,
      "balance_loss_clip": 1.06128395,
      "balance_loss_mlp": 1.02630341,
      "epoch": 0.3524319124631756,
      "flos": 21433966479360.0,
      "grad_norm": 2.1659692326901347,
      "language_loss": 0.86152935,
      "learning_rate": 3.0051675381841516e-06,
      "loss": 0.88387394,
      "num_input_tokens_seen": 62953470,
      "step": 2931,
      "time_per_iteration": 2.593573808670044
    },
    {
      "auxiliary_loss_clip": 0.01197543,
      "auxiliary_loss_mlp": 0.01037272,
      "balance_loss_clip": 1.05843914,
      "balance_loss_mlp": 1.02731252,
      "epoch": 0.3525521553538147,
      "flos": 26322773114880.0,
      "grad_norm": 1.788036653782403,
      "language_loss": 0.76623201,
      "learning_rate": 3.0044940187573363e-06,
      "loss": 0.78858018,
      "num_input_tokens_seen": 62974480,
      "step": 2932,
      "time_per_iteration": 2.647615671157837
    },
    {
      "auxiliary_loss_clip": 0.01197834,
      "auxiliary_loss_mlp": 0.01040249,
      "balance_loss_clip": 1.06065893,
      "balance_loss_mlp": 1.03079569,
      "epoch": 0.3526723982444538,
      "flos": 21543314457600.0,
      "grad_norm": 11.72856671640214,
      "language_loss": 0.65234679,
      "learning_rate": 3.003820346949578e-06,
      "loss": 0.67472762,
      "num_input_tokens_seen": 62992560,
      "step": 2933,
      "time_per_iteration": 2.6691172122955322
    },
    {
      "auxiliary_loss_clip": 0.01196839,
      "auxiliary_loss_mlp": 0.01036155,
      "balance_loss_clip": 1.05840778,
      "balance_loss_mlp": 1.02606404,
      "epoch": 0.3527926411350929,
      "flos": 23733649900800.0,
      "grad_norm": 3.202531507483283,
      "language_loss": 0.79566574,
      "learning_rate": 3.003146522863071e-06,
      "loss": 0.81799567,
      "num_input_tokens_seen": 63013445,
      "step": 2934,
      "time_per_iteration": 3.604076862335205
    },
    {
      "auxiliary_loss_clip": 0.01196129,
      "auxiliary_loss_mlp": 0.01031478,
      "balance_loss_clip": 1.05832803,
      "balance_loss_mlp": 1.02151847,
      "epoch": 0.35291288402573195,
      "flos": 30445461544320.0,
      "grad_norm": 2.4354483007747993,
      "language_loss": 0.85646141,
      "learning_rate": 3.0024725466000345e-06,
      "loss": 0.87873745,
      "num_input_tokens_seen": 63033400,
      "step": 2935,
      "time_per_iteration": 2.714303970336914
    },
    {
      "auxiliary_loss_clip": 0.0119762,
      "auxiliary_loss_mlp": 0.01028431,
      "balance_loss_clip": 1.06052768,
      "balance_loss_mlp": 1.01882887,
      "epoch": 0.35303312691637107,
      "flos": 23112179763840.0,
      "grad_norm": 2.1017759708985984,
      "language_loss": 0.79006314,
      "learning_rate": 3.0017984182627087e-06,
      "loss": 0.81232357,
      "num_input_tokens_seen": 63052725,
      "step": 2936,
      "time_per_iteration": 2.602527618408203
    },
    {
      "auxiliary_loss_clip": 0.01197454,
      "auxiliary_loss_mlp": 0.01034348,
      "balance_loss_clip": 1.05893564,
      "balance_loss_mlp": 1.02420342,
      "epoch": 0.3531533698070102,
      "flos": 21835699165440.0,
      "grad_norm": 2.2453761092458926,
      "language_loss": 0.82233834,
      "learning_rate": 3.00112413795336e-06,
      "loss": 0.84465635,
      "num_input_tokens_seen": 63072560,
      "step": 2937,
      "time_per_iteration": 2.669358491897583
    },
    {
      "auxiliary_loss_clip": 0.01196195,
      "auxiliary_loss_mlp": 0.01033402,
      "balance_loss_clip": 1.05774188,
      "balance_loss_mlp": 1.02348375,
      "epoch": 0.35327361269764923,
      "flos": 15778969810560.0,
      "grad_norm": 2.094883371987424,
      "language_loss": 0.80218542,
      "learning_rate": 3.000449705774275e-06,
      "loss": 0.82448137,
      "num_input_tokens_seen": 63090800,
      "step": 2938,
      "time_per_iteration": 4.458374500274658
    },
    {
      "auxiliary_loss_clip": 0.01195437,
      "auxiliary_loss_mlp": 0.01034941,
      "balance_loss_clip": 1.05773032,
      "balance_loss_mlp": 1.02506423,
      "epoch": 0.35339385558828834,
      "flos": 22090413484800.0,
      "grad_norm": 2.0225030507466433,
      "language_loss": 0.71736753,
      "learning_rate": 2.9997751218277654e-06,
      "loss": 0.73967129,
      "num_input_tokens_seen": 63108955,
      "step": 2939,
      "time_per_iteration": 3.5352847576141357
    },
    {
      "auxiliary_loss_clip": 0.01195961,
      "auxiliary_loss_mlp": 0.01033125,
      "balance_loss_clip": 1.05941808,
      "balance_loss_mlp": 1.02263486,
      "epoch": 0.35351409847892745,
      "flos": 24165008328960.0,
      "grad_norm": 2.1734346230810337,
      "language_loss": 0.77624357,
      "learning_rate": 2.999100386216166e-06,
      "loss": 0.79853445,
      "num_input_tokens_seen": 63127895,
      "step": 2940,
      "time_per_iteration": 2.6905977725982666
    },
    {
      "auxiliary_loss_clip": 0.01197508,
      "auxiliary_loss_mlp": 0.01034062,
      "balance_loss_clip": 1.05985022,
      "balance_loss_mlp": 1.02450132,
      "epoch": 0.3536343413695665,
      "flos": 27052298340480.0,
      "grad_norm": 1.9184598817339988,
      "language_loss": 0.74350059,
      "learning_rate": 2.998425499041831e-06,
      "loss": 0.76581633,
      "num_input_tokens_seen": 63148410,
      "step": 2941,
      "time_per_iteration": 2.6752243041992188
    },
    {
      "auxiliary_loss_clip": 0.01092425,
      "auxiliary_loss_mlp": 0.01006965,
      "balance_loss_clip": 1.02250624,
      "balance_loss_mlp": 1.00541496,
      "epoch": 0.3537545842602056,
      "flos": 65991066370560.0,
      "grad_norm": 1.2744163732683715,
      "language_loss": 0.64534676,
      "learning_rate": 2.997750460407142e-06,
      "loss": 0.66634059,
      "num_input_tokens_seen": 63209765,
      "step": 2942,
      "time_per_iteration": 3.2775206565856934
    },
    {
      "auxiliary_loss_clip": 0.01200045,
      "auxiliary_loss_mlp": 0.01038659,
      "balance_loss_clip": 1.0585593,
      "balance_loss_mlp": 1.02838945,
      "epoch": 0.35387482715084473,
      "flos": 18436897526400.0,
      "grad_norm": 1.8894256320501268,
      "language_loss": 0.69884789,
      "learning_rate": 2.997075270414501e-06,
      "loss": 0.72123492,
      "num_input_tokens_seen": 63226980,
      "step": 2943,
      "time_per_iteration": 2.619469404220581
    },
    {
      "auxiliary_loss_clip": 0.01092093,
      "auxiliary_loss_mlp": 0.01004399,
      "balance_loss_clip": 1.02240968,
      "balance_loss_mlp": 1.00281322,
      "epoch": 0.3539950700414838,
      "flos": 65588579498880.0,
      "grad_norm": 0.7056707976549332,
      "language_loss": 0.57747787,
      "learning_rate": 2.9963999291663347e-06,
      "loss": 0.59844279,
      "num_input_tokens_seen": 63292760,
      "step": 2944,
      "time_per_iteration": 3.211517810821533
    },
    {
      "auxiliary_loss_clip": 0.01201813,
      "auxiliary_loss_mlp": 0.01036521,
      "balance_loss_clip": 1.06258535,
      "balance_loss_mlp": 1.02680564,
      "epoch": 0.3541153129321229,
      "flos": 20521655919360.0,
      "grad_norm": 2.82502955935749,
      "language_loss": 0.73735976,
      "learning_rate": 2.9957244367650915e-06,
      "loss": 0.75974309,
      "num_input_tokens_seen": 63309005,
      "step": 2945,
      "time_per_iteration": 2.642988920211792
    },
    {
      "auxiliary_loss_clip": 0.01198058,
      "auxiliary_loss_mlp": 0.01029267,
      "balance_loss_clip": 1.06209791,
      "balance_loss_mlp": 1.02008152,
      "epoch": 0.354235555822762,
      "flos": 19573578391680.0,
      "grad_norm": 2.3446780549004385,
      "language_loss": 0.83773172,
      "learning_rate": 2.9950487933132425e-06,
      "loss": 0.8600049,
      "num_input_tokens_seen": 63326420,
      "step": 2946,
      "time_per_iteration": 2.6229772567749023
    },
    {
      "auxiliary_loss_clip": 0.011973,
      "auxiliary_loss_mlp": 0.01034461,
      "balance_loss_clip": 1.05895877,
      "balance_loss_mlp": 1.02414942,
      "epoch": 0.35435579871340106,
      "flos": 20777268078720.0,
      "grad_norm": 2.539185613294466,
      "language_loss": 0.71064687,
      "learning_rate": 2.994372998913283e-06,
      "loss": 0.73296446,
      "num_input_tokens_seen": 63344925,
      "step": 2947,
      "time_per_iteration": 2.6319568157196045
    },
    {
      "auxiliary_loss_clip": 0.01199711,
      "auxiliary_loss_mlp": 0.01036987,
      "balance_loss_clip": 1.06094146,
      "balance_loss_mlp": 1.02737904,
      "epoch": 0.35447604160404017,
      "flos": 23951807153280.0,
      "grad_norm": 2.4474211503055696,
      "language_loss": 0.62552649,
      "learning_rate": 2.99369705366773e-06,
      "loss": 0.64789349,
      "num_input_tokens_seen": 63365170,
      "step": 2948,
      "time_per_iteration": 2.6175413131713867
    },
    {
      "auxiliary_loss_clip": 0.01197128,
      "auxiliary_loss_mlp": 0.01034494,
      "balance_loss_clip": 1.05892706,
      "balance_loss_mlp": 1.0249511,
      "epoch": 0.3545962844946792,
      "flos": 23435662671360.0,
      "grad_norm": 2.065606954521431,
      "language_loss": 0.82017028,
      "learning_rate": 2.9930209576791244e-06,
      "loss": 0.84248656,
      "num_input_tokens_seen": 63383645,
      "step": 2949,
      "time_per_iteration": 2.592679023742676
    },
    {
      "auxiliary_loss_clip": 0.01192271,
      "auxiliary_loss_mlp": 0.01041815,
      "balance_loss_clip": 1.05574536,
      "balance_loss_mlp": 1.03187323,
      "epoch": 0.35471652738531834,
      "flos": 22085134185600.0,
      "grad_norm": 1.847729161049237,
      "language_loss": 0.6311807,
      "learning_rate": 2.9923447110500285e-06,
      "loss": 0.6535216,
      "num_input_tokens_seen": 63402390,
      "step": 2950,
      "time_per_iteration": 2.693772077560425
    },
    {
      "auxiliary_loss_clip": 0.01196945,
      "auxiliary_loss_mlp": 0.01029671,
      "balance_loss_clip": 1.06002927,
      "balance_loss_mlp": 1.02059364,
      "epoch": 0.35483677027595745,
      "flos": 27341881787520.0,
      "grad_norm": 1.686422969453801,
      "language_loss": 0.75245929,
      "learning_rate": 2.9916683138830295e-06,
      "loss": 0.77472544,
      "num_input_tokens_seen": 63423055,
      "step": 2951,
      "time_per_iteration": 2.6610658168792725
    },
    {
      "auxiliary_loss_clip": 0.01193964,
      "auxiliary_loss_mlp": 0.01034069,
      "balance_loss_clip": 1.05714965,
      "balance_loss_mlp": 1.0245676,
      "epoch": 0.3549570131665965,
      "flos": 13516166678400.0,
      "grad_norm": 2.0470364899816493,
      "language_loss": 0.80847001,
      "learning_rate": 2.9909917662807353e-06,
      "loss": 0.83075035,
      "num_input_tokens_seen": 63440855,
      "step": 2952,
      "time_per_iteration": 2.6624672412872314
    },
    {
      "auxiliary_loss_clip": 0.01197135,
      "auxiliary_loss_mlp": 0.01033402,
      "balance_loss_clip": 1.05813479,
      "balance_loss_mlp": 1.02312589,
      "epoch": 0.3550772560572356,
      "flos": 20887549810560.0,
      "grad_norm": 2.2573908646659637,
      "language_loss": 0.69383341,
      "learning_rate": 2.9903150683457783e-06,
      "loss": 0.71613878,
      "num_input_tokens_seen": 63459400,
      "step": 2953,
      "time_per_iteration": 2.6733646392822266
    },
    {
      "auxiliary_loss_clip": 0.01192265,
      "auxiliary_loss_mlp": 0.01028294,
      "balance_loss_clip": 1.05506372,
      "balance_loss_mlp": 1.01835251,
      "epoch": 0.3551974989478747,
      "flos": 20194042947840.0,
      "grad_norm": 1.990035244109508,
      "language_loss": 0.64169472,
      "learning_rate": 2.9896382201808126e-06,
      "loss": 0.66390026,
      "num_input_tokens_seen": 63476800,
      "step": 2954,
      "time_per_iteration": 2.639744520187378
    },
    {
      "auxiliary_loss_clip": 0.01196757,
      "auxiliary_loss_mlp": 0.01038002,
      "balance_loss_clip": 1.05776775,
      "balance_loss_mlp": 1.02753544,
      "epoch": 0.3553177418385138,
      "flos": 19828831415040.0,
      "grad_norm": 4.957359860302159,
      "language_loss": 0.81231982,
      "learning_rate": 2.988961221888516e-06,
      "loss": 0.83466738,
      "num_input_tokens_seen": 63493475,
      "step": 2955,
      "time_per_iteration": 2.598558187484741
    },
    {
      "auxiliary_loss_clip": 0.01195848,
      "auxiliary_loss_mlp": 0.01034917,
      "balance_loss_clip": 1.05850029,
      "balance_loss_mlp": 1.02577412,
      "epoch": 0.3554379847291529,
      "flos": 14829132516480.0,
      "grad_norm": 2.3728082474145267,
      "language_loss": 0.7886008,
      "learning_rate": 2.988284073571589e-06,
      "loss": 0.81090844,
      "num_input_tokens_seen": 63509560,
      "step": 2956,
      "time_per_iteration": 2.5903263092041016
    },
    {
      "auxiliary_loss_clip": 0.01196056,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.05781031,
      "balance_loss_mlp": 1.01756084,
      "epoch": 0.355558227619792,
      "flos": 20485350247680.0,
      "grad_norm": 2.485680310666278,
      "language_loss": 0.72990823,
      "learning_rate": 2.9876067753327528e-06,
      "loss": 0.75214362,
      "num_input_tokens_seen": 63527290,
      "step": 2957,
      "time_per_iteration": 2.6898484230041504
    },
    {
      "auxiliary_loss_clip": 0.01195221,
      "auxiliary_loss_mlp": 0.01034263,
      "balance_loss_clip": 1.05692446,
      "balance_loss_mlp": 1.0240531,
      "epoch": 0.35567847051043106,
      "flos": 37663613256960.0,
      "grad_norm": 2.0135342684374606,
      "language_loss": 0.80418712,
      "learning_rate": 2.986929327274754e-06,
      "loss": 0.82648194,
      "num_input_tokens_seen": 63547870,
      "step": 2958,
      "time_per_iteration": 2.7220306396484375
    },
    {
      "auxiliary_loss_clip": 0.01198999,
      "auxiliary_loss_mlp": 0.01038503,
      "balance_loss_clip": 1.06120229,
      "balance_loss_mlp": 1.02857304,
      "epoch": 0.35579871340107017,
      "flos": 26943058103040.0,
      "grad_norm": 1.648007915670238,
      "language_loss": 0.78601807,
      "learning_rate": 2.9862517295003617e-06,
      "loss": 0.80839312,
      "num_input_tokens_seen": 63568285,
      "step": 2959,
      "time_per_iteration": 2.66412353515625
    },
    {
      "auxiliary_loss_clip": 0.01197845,
      "auxiliary_loss_mlp": 0.01029285,
      "balance_loss_clip": 1.05889249,
      "balance_loss_mlp": 1.02005267,
      "epoch": 0.3559189562917093,
      "flos": 28293335193600.0,
      "grad_norm": 1.555250890733678,
      "language_loss": 0.72421587,
      "learning_rate": 2.9855739821123654e-06,
      "loss": 0.7464872,
      "num_input_tokens_seen": 63589865,
      "step": 2960,
      "time_per_iteration": 2.696458339691162
    },
    {
      "auxiliary_loss_clip": 0.0119671,
      "auxiliary_loss_mlp": 0.0103088,
      "balance_loss_clip": 1.05831981,
      "balance_loss_mlp": 1.02171302,
      "epoch": 0.35603919918234833,
      "flos": 25664063552640.0,
      "grad_norm": 1.6932840521472483,
      "language_loss": 0.81902975,
      "learning_rate": 2.98489608521358e-06,
      "loss": 0.84130561,
      "num_input_tokens_seen": 63609805,
      "step": 2961,
      "time_per_iteration": 3.588700532913208
    },
    {
      "auxiliary_loss_clip": 0.01198285,
      "auxiliary_loss_mlp": 0.01034259,
      "balance_loss_clip": 1.05967903,
      "balance_loss_mlp": 1.02460265,
      "epoch": 0.35615944207298744,
      "flos": 23000856537600.0,
      "grad_norm": 2.3598030508732557,
      "language_loss": 0.79622757,
      "learning_rate": 2.9842180389068425e-06,
      "loss": 0.81855297,
      "num_input_tokens_seen": 63627115,
      "step": 2962,
      "time_per_iteration": 2.6783905029296875
    },
    {
      "auxiliary_loss_clip": 0.01091588,
      "auxiliary_loss_mlp": 0.01011528,
      "balance_loss_clip": 1.02210677,
      "balance_loss_mlp": 1.00993049,
      "epoch": 0.35627968496362655,
      "flos": 68251283723520.0,
      "grad_norm": 0.9397213661509686,
      "language_loss": 0.59164596,
      "learning_rate": 2.98353984329501e-06,
      "loss": 0.6126771,
      "num_input_tokens_seen": 63691460,
      "step": 2963,
      "time_per_iteration": 3.2359089851379395
    },
    {
      "auxiliary_loss_clip": 0.01197993,
      "auxiliary_loss_mlp": 0.01038884,
      "balance_loss_clip": 1.05860353,
      "balance_loss_mlp": 1.02838755,
      "epoch": 0.3563999278542656,
      "flos": 22641714403200.0,
      "grad_norm": 1.6671151361979553,
      "language_loss": 0.70501792,
      "learning_rate": 2.982861498480965e-06,
      "loss": 0.72738671,
      "num_input_tokens_seen": 63713840,
      "step": 2964,
      "time_per_iteration": 3.59110689163208
    },
    {
      "auxiliary_loss_clip": 0.01193734,
      "auxiliary_loss_mlp": 0.01028838,
      "balance_loss_clip": 1.05773938,
      "balance_loss_mlp": 1.01993287,
      "epoch": 0.3565201707449047,
      "flos": 25952533678080.0,
      "grad_norm": 4.453765987213843,
      "language_loss": 0.82395124,
      "learning_rate": 2.9821830045676122e-06,
      "loss": 0.84617698,
      "num_input_tokens_seen": 63733540,
      "step": 2965,
      "time_per_iteration": 3.605952501296997
    },
    {
      "auxiliary_loss_clip": 0.01199312,
      "auxiliary_loss_mlp": 0.01037099,
      "balance_loss_clip": 1.05985808,
      "balance_loss_mlp": 1.02662647,
      "epoch": 0.3566404136355438,
      "flos": 28475725478400.0,
      "grad_norm": 1.6388426978469817,
      "language_loss": 0.72791541,
      "learning_rate": 2.9815043616578793e-06,
      "loss": 0.75027955,
      "num_input_tokens_seen": 63754335,
      "step": 2966,
      "time_per_iteration": 2.7295749187469482
    },
    {
      "auxiliary_loss_clip": 0.01197901,
      "auxiliary_loss_mlp": 0.01033623,
      "balance_loss_clip": 1.06032228,
      "balance_loss_mlp": 1.02345467,
      "epoch": 0.3567606565261829,
      "flos": 38363117690880.0,
      "grad_norm": 2.1009150265285714,
      "language_loss": 0.76705939,
      "learning_rate": 2.9808255698547145e-06,
      "loss": 0.78937459,
      "num_input_tokens_seen": 63777135,
      "step": 2967,
      "time_per_iteration": 2.843233585357666
    },
    {
      "auxiliary_loss_clip": 0.01199149,
      "auxiliary_loss_mlp": 0.01041576,
      "balance_loss_clip": 1.06172872,
      "balance_loss_mlp": 1.03194416,
      "epoch": 0.356880899416822,
      "flos": 21981029592960.0,
      "grad_norm": 3.490023926264328,
      "language_loss": 0.79976356,
      "learning_rate": 2.9801466292610913e-06,
      "loss": 0.82217079,
      "num_input_tokens_seen": 63797020,
      "step": 2968,
      "time_per_iteration": 2.635484457015991
    },
    {
      "auxiliary_loss_clip": 0.01191875,
      "auxiliary_loss_mlp": 0.01032806,
      "balance_loss_clip": 1.05597091,
      "balance_loss_mlp": 1.02408004,
      "epoch": 0.35700114230746105,
      "flos": 18989132198400.0,
      "grad_norm": 4.112737518005869,
      "language_loss": 0.81269014,
      "learning_rate": 2.979467539980003e-06,
      "loss": 0.83493698,
      "num_input_tokens_seen": 63813810,
      "step": 2969,
      "time_per_iteration": 2.545851230621338
    },
    {
      "auxiliary_loss_clip": 0.01197752,
      "auxiliary_loss_mlp": 0.01030604,
      "balance_loss_clip": 1.05915105,
      "balance_loss_mlp": 1.02083445,
      "epoch": 0.35712138519810016,
      "flos": 19756112330880.0,
      "grad_norm": 2.079139956927367,
      "language_loss": 0.76627052,
      "learning_rate": 2.978788302114468e-06,
      "loss": 0.78855407,
      "num_input_tokens_seen": 63830925,
      "step": 2970,
      "time_per_iteration": 2.634765863418579
    },
    {
      "auxiliary_loss_clip": 0.01198315,
      "auxiliary_loss_mlp": 0.01033028,
      "balance_loss_clip": 1.05966258,
      "balance_loss_mlp": 1.02287781,
      "epoch": 0.35724162808873927,
      "flos": 35183012008320.0,
      "grad_norm": 2.1747974779486623,
      "language_loss": 0.8317675,
      "learning_rate": 2.9781089157675255e-06,
      "loss": 0.85408092,
      "num_input_tokens_seen": 63849385,
      "step": 2971,
      "time_per_iteration": 2.737532377243042
    },
    {
      "auxiliary_loss_clip": 0.01196013,
      "auxiliary_loss_mlp": 0.01030009,
      "balance_loss_clip": 1.05839562,
      "balance_loss_mlp": 1.02056193,
      "epoch": 0.3573618709793783,
      "flos": 25556726736000.0,
      "grad_norm": 2.0994324350613565,
      "language_loss": 0.88607079,
      "learning_rate": 2.977429381042238e-06,
      "loss": 0.90833104,
      "num_input_tokens_seen": 63870060,
      "step": 2972,
      "time_per_iteration": 2.6783194541931152
    },
    {
      "auxiliary_loss_clip": 0.0119438,
      "auxiliary_loss_mlp": 0.01038177,
      "balance_loss_clip": 1.05994487,
      "balance_loss_mlp": 1.0293498,
      "epoch": 0.35748211387001744,
      "flos": 29132352051840.0,
      "grad_norm": 2.269178135664165,
      "language_loss": 0.8977105,
      "learning_rate": 2.9767496980416913e-06,
      "loss": 0.92003608,
      "num_input_tokens_seen": 63889355,
      "step": 2973,
      "time_per_iteration": 2.6626617908477783
    },
    {
      "auxiliary_loss_clip": 0.01196736,
      "auxiliary_loss_mlp": 0.01036393,
      "balance_loss_clip": 1.05865526,
      "balance_loss_mlp": 1.02712488,
      "epoch": 0.35760235676065655,
      "flos": 13954169122560.0,
      "grad_norm": 3.162281422386928,
      "language_loss": 0.80828518,
      "learning_rate": 2.9760698668689914e-06,
      "loss": 0.83061647,
      "num_input_tokens_seen": 63905580,
      "step": 2974,
      "time_per_iteration": 2.5913658142089844
    },
    {
      "auxiliary_loss_clip": 0.01194309,
      "auxiliary_loss_mlp": 0.01033096,
      "balance_loss_clip": 1.05788398,
      "balance_loss_mlp": 1.02413189,
      "epoch": 0.3577225996512956,
      "flos": 44018688977280.0,
      "grad_norm": 2.1680268503665485,
      "language_loss": 0.71264589,
      "learning_rate": 2.975389887627269e-06,
      "loss": 0.73491991,
      "num_input_tokens_seen": 63928180,
      "step": 2975,
      "time_per_iteration": 2.858794927597046
    },
    {
      "auxiliary_loss_clip": 0.0119727,
      "auxiliary_loss_mlp": 0.01033648,
      "balance_loss_clip": 1.05852032,
      "balance_loss_mlp": 1.0237956,
      "epoch": 0.3578428425419347,
      "flos": 17055199013760.0,
      "grad_norm": 3.2765751192753894,
      "language_loss": 0.89721924,
      "learning_rate": 2.9747097604196764e-06,
      "loss": 0.91952837,
      "num_input_tokens_seen": 63944825,
      "step": 2976,
      "time_per_iteration": 2.5924718379974365
    },
    {
      "auxiliary_loss_clip": 0.0108957,
      "auxiliary_loss_mlp": 0.01002466,
      "balance_loss_clip": 1.02042651,
      "balance_loss_mlp": 1.00085664,
      "epoch": 0.3579630854325738,
      "flos": 71676550707840.0,
      "grad_norm": 0.6674344113705813,
      "language_loss": 0.56602913,
      "learning_rate": 2.9740294853493875e-06,
      "loss": 0.58694953,
      "num_input_tokens_seen": 64016385,
      "step": 2977,
      "time_per_iteration": 3.4158987998962402
    },
    {
      "auxiliary_loss_clip": 0.01197446,
      "auxiliary_loss_mlp": 0.01033265,
      "balance_loss_clip": 1.05956507,
      "balance_loss_mlp": 1.02327502,
      "epoch": 0.3580833283232129,
      "flos": 25046651652480.0,
      "grad_norm": 2.2636669704862813,
      "language_loss": 0.67372507,
      "learning_rate": 2.9733490625196008e-06,
      "loss": 0.69603217,
      "num_input_tokens_seen": 64036245,
      "step": 2978,
      "time_per_iteration": 2.655906915664673
    },
    {
      "auxiliary_loss_clip": 0.01194725,
      "auxiliary_loss_mlp": 0.01030525,
      "balance_loss_clip": 1.05760884,
      "balance_loss_mlp": 1.02102995,
      "epoch": 0.358203571213852,
      "flos": 13953127628160.0,
      "grad_norm": 3.489704045816555,
      "language_loss": 0.7514599,
      "learning_rate": 2.9726684920335353e-06,
      "loss": 0.7737124,
      "num_input_tokens_seen": 64054110,
      "step": 2979,
      "time_per_iteration": 2.659677743911743
    },
    {
      "auxiliary_loss_clip": 0.01197758,
      "auxiliary_loss_mlp": 0.01033096,
      "balance_loss_clip": 1.0579145,
      "balance_loss_mlp": 1.02262926,
      "epoch": 0.35832381410449105,
      "flos": 20302457172480.0,
      "grad_norm": 2.1793291729300974,
      "language_loss": 0.81967354,
      "learning_rate": 2.971987773994432e-06,
      "loss": 0.84198201,
      "num_input_tokens_seen": 64070295,
      "step": 2980,
      "time_per_iteration": 2.572006940841675
    },
    {
      "auxiliary_loss_clip": 0.01194433,
      "auxiliary_loss_mlp": 0.01033015,
      "balance_loss_clip": 1.05725336,
      "balance_loss_mlp": 1.0235734,
      "epoch": 0.35844405699513016,
      "flos": 16983234115200.0,
      "grad_norm": 1.8374550520513429,
      "language_loss": 0.8310827,
      "learning_rate": 2.9713069085055566e-06,
      "loss": 0.85335726,
      "num_input_tokens_seen": 64088605,
      "step": 2981,
      "time_per_iteration": 2.608250379562378
    },
    {
      "auxiliary_loss_clip": 0.01195976,
      "auxiliary_loss_mlp": 0.01029037,
      "balance_loss_clip": 1.05914962,
      "balance_loss_mlp": 1.01948214,
      "epoch": 0.35856429988576927,
      "flos": 23216858974080.0,
      "grad_norm": 1.705730134508696,
      "language_loss": 0.78521228,
      "learning_rate": 2.9706258956701958e-06,
      "loss": 0.80746245,
      "num_input_tokens_seen": 64108595,
      "step": 2982,
      "time_per_iteration": 2.59956955909729
    },
    {
      "auxiliary_loss_clip": 0.01198233,
      "auxiliary_loss_mlp": 0.01036005,
      "balance_loss_clip": 1.06003797,
      "balance_loss_mlp": 1.02577651,
      "epoch": 0.3586845427764083,
      "flos": 23034576430080.0,
      "grad_norm": 2.40499180911731,
      "language_loss": 0.77567601,
      "learning_rate": 2.9699447355916575e-06,
      "loss": 0.79801846,
      "num_input_tokens_seen": 64127405,
      "step": 2983,
      "time_per_iteration": 2.6474342346191406
    },
    {
      "auxiliary_loss_clip": 0.01195489,
      "auxiliary_loss_mlp": 0.01027078,
      "balance_loss_clip": 1.05783391,
      "balance_loss_mlp": 1.01792264,
      "epoch": 0.35880478566704743,
      "flos": 20010682995840.0,
      "grad_norm": 1.8576847920544166,
      "language_loss": 0.73739254,
      "learning_rate": 2.969263428373275e-06,
      "loss": 0.75961822,
      "num_input_tokens_seen": 64145755,
      "step": 2984,
      "time_per_iteration": 2.612462282180786
    },
    {
      "auxiliary_loss_clip": 0.01196841,
      "auxiliary_loss_mlp": 0.01033633,
      "balance_loss_clip": 1.06023145,
      "balance_loss_mlp": 1.02451336,
      "epoch": 0.35892502855768654,
      "flos": 13699095667200.0,
      "grad_norm": 2.0593617447170613,
      "language_loss": 0.79160154,
      "learning_rate": 2.9685819741184007e-06,
      "loss": 0.81390631,
      "num_input_tokens_seen": 64164195,
      "step": 2985,
      "time_per_iteration": 2.6321051120758057
    },
    {
      "auxiliary_loss_clip": 0.01191275,
      "auxiliary_loss_mlp": 0.01030449,
      "balance_loss_clip": 1.05697227,
      "balance_loss_mlp": 1.02168691,
      "epoch": 0.3590452714483256,
      "flos": 18114096977280.0,
      "grad_norm": 2.459630890213124,
      "language_loss": 0.68223858,
      "learning_rate": 2.967900372930411e-06,
      "loss": 0.70445579,
      "num_input_tokens_seen": 64182705,
      "step": 2986,
      "time_per_iteration": 2.770029306411743
    },
    {
      "auxiliary_loss_clip": 0.011991,
      "auxiliary_loss_mlp": 0.01033495,
      "balance_loss_clip": 1.05903673,
      "balance_loss_mlp": 1.02279592,
      "epoch": 0.3591655143389647,
      "flos": 17749352321280.0,
      "grad_norm": 3.3983784938543233,
      "language_loss": 0.79082316,
      "learning_rate": 2.9672186249127046e-06,
      "loss": 0.81314909,
      "num_input_tokens_seen": 64202170,
      "step": 2987,
      "time_per_iteration": 2.6272706985473633
    },
    {
      "auxiliary_loss_clip": 0.01196659,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.06075311,
      "balance_loss_mlp": 1.02039182,
      "epoch": 0.3592857572296038,
      "flos": 25224409082880.0,
      "grad_norm": 1.8625476638892629,
      "language_loss": 0.78694069,
      "learning_rate": 2.9665367301687014e-06,
      "loss": 0.80920213,
      "num_input_tokens_seen": 64220415,
      "step": 2988,
      "time_per_iteration": 3.595115900039673
    },
    {
      "auxiliary_loss_clip": 0.01194741,
      "auxiliary_loss_mlp": 0.01028486,
      "balance_loss_clip": 1.05770111,
      "balance_loss_mlp": 1.01905656,
      "epoch": 0.3594060001202429,
      "flos": 29384408764800.0,
      "grad_norm": 2.1768296729599306,
      "language_loss": 0.76539981,
      "learning_rate": 2.965854688801845e-06,
      "loss": 0.78763211,
      "num_input_tokens_seen": 64242475,
      "step": 2989,
      "time_per_iteration": 2.7023847103118896
    },
    {
      "auxiliary_loss_clip": 0.01191785,
      "auxiliary_loss_mlp": 0.01026748,
      "balance_loss_clip": 1.05722725,
      "balance_loss_mlp": 1.0175395,
      "epoch": 0.359526243010882,
      "flos": 17052900543360.0,
      "grad_norm": 2.1671805010963583,
      "language_loss": 0.76445574,
      "learning_rate": 2.9651725009156005e-06,
      "loss": 0.78664112,
      "num_input_tokens_seen": 64260220,
      "step": 2990,
      "time_per_iteration": 2.61106014251709
    },
    {
      "auxiliary_loss_clip": 0.01197205,
      "auxiliary_loss_mlp": 0.01032695,
      "balance_loss_clip": 1.05870795,
      "balance_loss_mlp": 1.02315247,
      "epoch": 0.3596464859015211,
      "flos": 22965089569920.0,
      "grad_norm": 2.2960633553609875,
      "language_loss": 0.74516916,
      "learning_rate": 2.964490166613454e-06,
      "loss": 0.76746821,
      "num_input_tokens_seen": 64280145,
      "step": 2991,
      "time_per_iteration": 4.556478261947632
    },
    {
      "auxiliary_loss_clip": 0.01093252,
      "auxiliary_loss_mlp": 0.01005855,
      "balance_loss_clip": 1.02381945,
      "balance_loss_mlp": 1.0042572,
      "epoch": 0.35976672879216015,
      "flos": 54739462590720.0,
      "grad_norm": 0.7659926179185849,
      "language_loss": 0.57706892,
      "learning_rate": 2.963807685998917e-06,
      "loss": 0.59805989,
      "num_input_tokens_seen": 64336010,
      "step": 2992,
      "time_per_iteration": 3.969104528427124
    },
    {
      "auxiliary_loss_clip": 0.01194218,
      "auxiliary_loss_mlp": 0.01032067,
      "balance_loss_clip": 1.05890059,
      "balance_loss_mlp": 1.02285218,
      "epoch": 0.35988697168279926,
      "flos": 43139020901760.0,
      "grad_norm": 2.0254224686274394,
      "language_loss": 0.78025347,
      "learning_rate": 2.9631250591755196e-06,
      "loss": 0.80251628,
      "num_input_tokens_seen": 64358725,
      "step": 2993,
      "time_per_iteration": 2.7891294956207275
    },
    {
      "auxiliary_loss_clip": 0.0119469,
      "auxiliary_loss_mlp": 0.01031064,
      "balance_loss_clip": 1.05872238,
      "balance_loss_mlp": 1.02165258,
      "epoch": 0.36000721457343837,
      "flos": 35845600239360.0,
      "grad_norm": 2.1098646992146532,
      "language_loss": 0.57212126,
      "learning_rate": 2.962442286246817e-06,
      "loss": 0.59437883,
      "num_input_tokens_seen": 64381555,
      "step": 2994,
      "time_per_iteration": 2.731537342071533
    },
    {
      "auxiliary_loss_clip": 0.01194878,
      "auxiliary_loss_mlp": 0.01031227,
      "balance_loss_clip": 1.05756617,
      "balance_loss_mlp": 1.02211308,
      "epoch": 0.3601274574640774,
      "flos": 18291100222080.0,
      "grad_norm": 2.110329764913845,
      "language_loss": 0.69870222,
      "learning_rate": 2.9617593673163853e-06,
      "loss": 0.72096324,
      "num_input_tokens_seen": 64400375,
      "step": 2995,
      "time_per_iteration": 2.647843599319458
    },
    {
      "auxiliary_loss_clip": 0.01192625,
      "auxiliary_loss_mlp": 0.01029583,
      "balance_loss_clip": 1.05731869,
      "balance_loss_mlp": 1.02089226,
      "epoch": 0.36024770035471654,
      "flos": 13333955961600.0,
      "grad_norm": 2.192469032875218,
      "language_loss": 0.7773453,
      "learning_rate": 2.9610763024878216e-06,
      "loss": 0.79956734,
      "num_input_tokens_seen": 64415880,
      "step": 2996,
      "time_per_iteration": 2.60414457321167
    },
    {
      "auxiliary_loss_clip": 0.01195897,
      "auxiliary_loss_mlp": 0.0103194,
      "balance_loss_clip": 1.05925679,
      "balance_loss_mlp": 1.02348197,
      "epoch": 0.3603679432453556,
      "flos": 20267013427200.0,
      "grad_norm": 1.8630817190052749,
      "language_loss": 0.91365236,
      "learning_rate": 2.960393091864747e-06,
      "loss": 0.93593067,
      "num_input_tokens_seen": 64434260,
      "step": 2997,
      "time_per_iteration": 2.6161108016967773
    },
    {
      "auxiliary_loss_clip": 0.01194938,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.06010056,
      "balance_loss_mlp": 1.01864398,
      "epoch": 0.3604881861359947,
      "flos": 22451135817600.0,
      "grad_norm": 1.7505602416322135,
      "language_loss": 0.75382054,
      "learning_rate": 2.959709735550804e-06,
      "loss": 0.77604526,
      "num_input_tokens_seen": 64453855,
      "step": 2998,
      "time_per_iteration": 2.629288673400879
    },
    {
      "auxiliary_loss_clip": 0.01194678,
      "auxiliary_loss_mlp": 0.0103457,
      "balance_loss_clip": 1.05744886,
      "balance_loss_mlp": 1.0248481,
      "epoch": 0.3606084290266338,
      "flos": 22054251467520.0,
      "grad_norm": 2.5070538814823116,
      "language_loss": 0.75698853,
      "learning_rate": 2.9590262336496575e-06,
      "loss": 0.77928102,
      "num_input_tokens_seen": 64473585,
      "step": 2999,
      "time_per_iteration": 2.6617491245269775
    },
    {
      "auxiliary_loss_clip": 0.01196157,
      "auxiliary_loss_mlp": 0.01035327,
      "balance_loss_clip": 1.05979443,
      "balance_loss_mlp": 1.02575445,
      "epoch": 0.36072867191727287,
      "flos": 15632921111040.0,
      "grad_norm": 1.9224671928393529,
      "language_loss": 0.85319507,
      "learning_rate": 2.9583425862649936e-06,
      "loss": 0.87550992,
      "num_input_tokens_seen": 64491720,
      "step": 3000,
      "time_per_iteration": 2.654026746749878
    },
    {
      "auxiliary_loss_clip": 0.01200946,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.06185746,
      "balance_loss_mlp": 1.01875448,
      "epoch": 0.360848914807912,
      "flos": 19677000625920.0,
      "grad_norm": 2.166133154795431,
      "language_loss": 0.73642325,
      "learning_rate": 2.9576587935005215e-06,
      "loss": 0.75871563,
      "num_input_tokens_seen": 64509800,
      "step": 3001,
      "time_per_iteration": 2.6527445316314697
    },
    {
      "auxiliary_loss_clip": 0.01197217,
      "auxiliary_loss_mlp": 0.01026934,
      "balance_loss_clip": 1.05969572,
      "balance_loss_mlp": 1.01684892,
      "epoch": 0.3609691576985511,
      "flos": 18877808972160.0,
      "grad_norm": 2.8102823048678633,
      "language_loss": 0.72083068,
      "learning_rate": 2.9569748554599713e-06,
      "loss": 0.74307215,
      "num_input_tokens_seen": 64525410,
      "step": 3002,
      "time_per_iteration": 2.692150115966797
    },
    {
      "auxiliary_loss_clip": 0.01196846,
      "auxiliary_loss_mlp": 0.01028193,
      "balance_loss_clip": 1.06144273,
      "balance_loss_mlp": 1.01922822,
      "epoch": 0.36108940058919015,
      "flos": 42224088648960.0,
      "grad_norm": 1.9811189161322307,
      "language_loss": 0.73226106,
      "learning_rate": 2.956290772247097e-06,
      "loss": 0.75451148,
      "num_input_tokens_seen": 64544085,
      "step": 3003,
      "time_per_iteration": 2.7968802452087402
    },
    {
      "auxiliary_loss_clip": 0.01197879,
      "auxiliary_loss_mlp": 0.01023914,
      "balance_loss_clip": 1.0611372,
      "balance_loss_mlp": 1.01536679,
      "epoch": 0.36120964347982926,
      "flos": 23185150243200.0,
      "grad_norm": 1.7107810889937516,
      "language_loss": 0.73089981,
      "learning_rate": 2.9556065439656724e-06,
      "loss": 0.7531178,
      "num_input_tokens_seen": 64563135,
      "step": 3004,
      "time_per_iteration": 2.6527466773986816
    },
    {
      "auxiliary_loss_clip": 0.01191184,
      "auxiliary_loss_mlp": 0.01032007,
      "balance_loss_clip": 1.05747581,
      "balance_loss_mlp": 1.02303612,
      "epoch": 0.36132988637046837,
      "flos": 18113055482880.0,
      "grad_norm": 2.0681747187326947,
      "language_loss": 0.81434327,
      "learning_rate": 2.9549221707194952e-06,
      "loss": 0.83657515,
      "num_input_tokens_seen": 64581985,
      "step": 3005,
      "time_per_iteration": 2.5920016765594482
    },
    {
      "auxiliary_loss_clip": 0.0119734,
      "auxiliary_loss_mlp": 0.01031782,
      "balance_loss_clip": 1.06029963,
      "balance_loss_mlp": 1.02307928,
      "epoch": 0.3614501292611074,
      "flos": 27813101333760.0,
      "grad_norm": 2.421776872330343,
      "language_loss": 0.72535402,
      "learning_rate": 2.954237652612384e-06,
      "loss": 0.74764526,
      "num_input_tokens_seen": 64601035,
      "step": 3006,
      "time_per_iteration": 2.731175422668457
    },
    {
      "auxiliary_loss_clip": 0.01192938,
      "auxiliary_loss_mlp": 0.01028729,
      "balance_loss_clip": 1.0576961,
      "balance_loss_mlp": 1.01946616,
      "epoch": 0.36157037215174653,
      "flos": 22634926732800.0,
      "grad_norm": 1.9169757661793707,
      "language_loss": 0.8431828,
      "learning_rate": 2.9535529897481796e-06,
      "loss": 0.86539954,
      "num_input_tokens_seen": 64618580,
      "step": 3007,
      "time_per_iteration": 2.7038707733154297
    },
    {
      "auxiliary_loss_clip": 0.01197926,
      "auxiliary_loss_mlp": 0.01027608,
      "balance_loss_clip": 1.05913901,
      "balance_loss_mlp": 1.01786876,
      "epoch": 0.36169061504238564,
      "flos": 12600839376000.0,
      "grad_norm": 2.106737362518035,
      "language_loss": 0.76835597,
      "learning_rate": 2.9528681822307446e-06,
      "loss": 0.79061127,
      "num_input_tokens_seen": 64635430,
      "step": 3008,
      "time_per_iteration": 2.5759308338165283
    },
    {
      "auxiliary_loss_clip": 0.01197301,
      "auxiliary_loss_mlp": 0.01038788,
      "balance_loss_clip": 1.06026936,
      "balance_loss_mlp": 1.02946568,
      "epoch": 0.3618108579330247,
      "flos": 26684644682880.0,
      "grad_norm": 2.0080918721418897,
      "language_loss": 0.82421327,
      "learning_rate": 2.952183230163964e-06,
      "loss": 0.84657419,
      "num_input_tokens_seen": 64655005,
      "step": 3009,
      "time_per_iteration": 2.6942989826202393
    },
    {
      "auxiliary_loss_clip": 0.01192307,
      "auxiliary_loss_mlp": 0.01029949,
      "balance_loss_clip": 1.05546665,
      "balance_loss_mlp": 1.02075791,
      "epoch": 0.3619311008236638,
      "flos": 22817029708800.0,
      "grad_norm": 2.008377056152682,
      "language_loss": 0.73106277,
      "learning_rate": 2.9514981336517448e-06,
      "loss": 0.75328529,
      "num_input_tokens_seen": 64674775,
      "step": 3010,
      "time_per_iteration": 2.607074022293091
    },
    {
      "auxiliary_loss_clip": 0.01197108,
      "auxiliary_loss_mlp": 0.01032493,
      "balance_loss_clip": 1.05878115,
      "balance_loss_mlp": 1.02354014,
      "epoch": 0.36205134371430286,
      "flos": 25919603884800.0,
      "grad_norm": 1.8624976629755399,
      "language_loss": 0.81357634,
      "learning_rate": 2.950812892798015e-06,
      "loss": 0.83587235,
      "num_input_tokens_seen": 64695670,
      "step": 3011,
      "time_per_iteration": 2.6733715534210205
    },
    {
      "auxiliary_loss_clip": 0.0119361,
      "auxiliary_loss_mlp": 0.01030763,
      "balance_loss_clip": 1.05875516,
      "balance_loss_mlp": 1.02189934,
      "epoch": 0.362171586604942,
      "flos": 26139592730880.0,
      "grad_norm": 1.8199427682564213,
      "language_loss": 0.87344599,
      "learning_rate": 2.9501275077067256e-06,
      "loss": 0.89568973,
      "num_input_tokens_seen": 64716290,
      "step": 3012,
      "time_per_iteration": 2.6677372455596924
    },
    {
      "auxiliary_loss_clip": 0.01192945,
      "auxiliary_loss_mlp": 0.01028138,
      "balance_loss_clip": 1.05916834,
      "balance_loss_mlp": 1.01941752,
      "epoch": 0.3622918294955811,
      "flos": 28074208273920.0,
      "grad_norm": 1.63153703295591,
      "language_loss": 0.88120365,
      "learning_rate": 2.949441978481848e-06,
      "loss": 0.90341449,
      "num_input_tokens_seen": 64737190,
      "step": 3013,
      "time_per_iteration": 2.71266770362854
    },
    {
      "auxiliary_loss_clip": 0.01196008,
      "auxiliary_loss_mlp": 0.01037926,
      "balance_loss_clip": 1.05738997,
      "balance_loss_mlp": 1.02831197,
      "epoch": 0.36241207238622014,
      "flos": 19828005402240.0,
      "grad_norm": 1.9166534027123499,
      "language_loss": 0.8014766,
      "learning_rate": 2.9487563052273778e-06,
      "loss": 0.82381594,
      "num_input_tokens_seen": 64753950,
      "step": 3014,
      "time_per_iteration": 2.635221242904663
    },
    {
      "auxiliary_loss_clip": 0.0119408,
      "auxiliary_loss_mlp": 0.01035305,
      "balance_loss_clip": 1.0587728,
      "balance_loss_mlp": 1.0262928,
      "epoch": 0.36253231527685925,
      "flos": 21397158017280.0,
      "grad_norm": 5.546251678328043,
      "language_loss": 0.85837245,
      "learning_rate": 2.94807048804733e-06,
      "loss": 0.88066632,
      "num_input_tokens_seen": 64773570,
      "step": 3015,
      "time_per_iteration": 3.5890979766845703
    },
    {
      "auxiliary_loss_clip": 0.01194786,
      "auxiliary_loss_mlp": 0.01029863,
      "balance_loss_clip": 1.05840874,
      "balance_loss_mlp": 1.02018356,
      "epoch": 0.36265255816749836,
      "flos": 18362885552640.0,
      "grad_norm": 1.8946710639868432,
      "language_loss": 0.9020952,
      "learning_rate": 2.9473845270457434e-06,
      "loss": 0.92434174,
      "num_input_tokens_seen": 64790385,
      "step": 3016,
      "time_per_iteration": 2.5981085300445557
    },
    {
      "auxiliary_loss_clip": 0.011909,
      "auxiliary_loss_mlp": 0.01028485,
      "balance_loss_clip": 1.05724716,
      "balance_loss_mlp": 1.01972866,
      "epoch": 0.3627728010581374,
      "flos": 18660046769280.0,
      "grad_norm": 2.75036470899253,
      "language_loss": 0.70038027,
      "learning_rate": 2.946698422326677e-06,
      "loss": 0.72257411,
      "num_input_tokens_seen": 64807845,
      "step": 3017,
      "time_per_iteration": 3.55348801612854
    },
    {
      "auxiliary_loss_clip": 0.01194648,
      "auxiliary_loss_mlp": 0.01034729,
      "balance_loss_clip": 1.05615103,
      "balance_loss_mlp": 1.02543104,
      "epoch": 0.36289304394877653,
      "flos": 27524272072320.0,
      "grad_norm": 2.2455853321242225,
      "language_loss": 0.79349637,
      "learning_rate": 2.946012173994213e-06,
      "loss": 0.81579018,
      "num_input_tokens_seen": 64827630,
      "step": 3018,
      "time_per_iteration": 4.546976327896118
    },
    {
      "auxiliary_loss_clip": 0.01193286,
      "auxiliary_loss_mlp": 0.01033568,
      "balance_loss_clip": 1.05977786,
      "balance_loss_mlp": 1.02401924,
      "epoch": 0.36301328683941564,
      "flos": 34533244932480.0,
      "grad_norm": 1.3585968699946993,
      "language_loss": 0.67717481,
      "learning_rate": 2.945325782152454e-06,
      "loss": 0.69944334,
      "num_input_tokens_seen": 64850665,
      "step": 3019,
      "time_per_iteration": 2.7382113933563232
    },
    {
      "auxiliary_loss_clip": 0.01192231,
      "auxiliary_loss_mlp": 0.01027419,
      "balance_loss_clip": 1.05764461,
      "balance_loss_mlp": 1.01875281,
      "epoch": 0.3631335297300547,
      "flos": 19025976574080.0,
      "grad_norm": 9.827964648050262,
      "language_loss": 0.79022735,
      "learning_rate": 2.9446392469055257e-06,
      "loss": 0.81242383,
      "num_input_tokens_seen": 64868700,
      "step": 3020,
      "time_per_iteration": 2.5872385501861572
    },
    {
      "auxiliary_loss_clip": 0.01194463,
      "auxiliary_loss_mlp": 0.01036875,
      "balance_loss_clip": 1.06124818,
      "balance_loss_mlp": 1.02823794,
      "epoch": 0.3632537726206938,
      "flos": 19536769929600.0,
      "grad_norm": 2.210812942970969,
      "language_loss": 0.8004247,
      "learning_rate": 2.9439525683575745e-06,
      "loss": 0.82273805,
      "num_input_tokens_seen": 64887620,
      "step": 3021,
      "time_per_iteration": 2.6174020767211914
    },
    {
      "auxiliary_loss_clip": 0.0120113,
      "auxiliary_loss_mlp": 0.01040449,
      "balance_loss_clip": 1.06118941,
      "balance_loss_mlp": 1.0304172,
      "epoch": 0.3633740155113329,
      "flos": 21068611292160.0,
      "grad_norm": 2.0289055728385974,
      "language_loss": 0.74809515,
      "learning_rate": 2.9432657466127694e-06,
      "loss": 0.77051091,
      "num_input_tokens_seen": 64907190,
      "step": 3022,
      "time_per_iteration": 2.6383872032165527
    },
    {
      "auxiliary_loss_clip": 0.01196205,
      "auxiliary_loss_mlp": 0.01033367,
      "balance_loss_clip": 1.06199503,
      "balance_loss_mlp": 1.0241884,
      "epoch": 0.36349425840197197,
      "flos": 20298722158080.0,
      "grad_norm": 1.8241014836672071,
      "language_loss": 0.76453972,
      "learning_rate": 2.9425787817753007e-06,
      "loss": 0.78683543,
      "num_input_tokens_seen": 64925850,
      "step": 3023,
      "time_per_iteration": 2.678358554840088
    },
    {
      "auxiliary_loss_clip": 0.01197486,
      "auxiliary_loss_mlp": 0.01031916,
      "balance_loss_clip": 1.05832314,
      "balance_loss_mlp": 1.02234948,
      "epoch": 0.3636145012926111,
      "flos": 29716762331520.0,
      "grad_norm": 1.5847575621810506,
      "language_loss": 0.71626592,
      "learning_rate": 2.94189167394938e-06,
      "loss": 0.73855996,
      "num_input_tokens_seen": 64948285,
      "step": 3024,
      "time_per_iteration": 2.6980974674224854
    },
    {
      "auxiliary_loss_clip": 0.01197628,
      "auxiliary_loss_mlp": 0.01029254,
      "balance_loss_clip": 1.06128073,
      "balance_loss_mlp": 1.01968122,
      "epoch": 0.3637347441832502,
      "flos": 21431847576960.0,
      "grad_norm": 1.7323064540584356,
      "language_loss": 0.81168467,
      "learning_rate": 2.941204423239241e-06,
      "loss": 0.8339535,
      "num_input_tokens_seen": 64967160,
      "step": 3025,
      "time_per_iteration": 2.648318290710449
    },
    {
      "auxiliary_loss_clip": 0.01190683,
      "auxiliary_loss_mlp": 0.01032615,
      "balance_loss_clip": 1.055812,
      "balance_loss_mlp": 1.02384102,
      "epoch": 0.36385498707388925,
      "flos": 29533941083520.0,
      "grad_norm": 1.651509692924,
      "language_loss": 0.76018417,
      "learning_rate": 2.9405170297491395e-06,
      "loss": 0.78241718,
      "num_input_tokens_seen": 64987155,
      "step": 3026,
      "time_per_iteration": 2.6574766635894775
    },
    {
      "auxiliary_loss_clip": 0.01196038,
      "auxiliary_loss_mlp": 0.01033784,
      "balance_loss_clip": 1.06021416,
      "balance_loss_mlp": 1.02396131,
      "epoch": 0.36397522996452836,
      "flos": 22236569925120.0,
      "grad_norm": 2.1723748819031954,
      "language_loss": 0.80399418,
      "learning_rate": 2.939829493583353e-06,
      "loss": 0.8262924,
      "num_input_tokens_seen": 65003800,
      "step": 3027,
      "time_per_iteration": 2.622002601623535
    },
    {
      "auxiliary_loss_clip": 0.01191811,
      "auxiliary_loss_mlp": 0.01030515,
      "balance_loss_clip": 1.05609536,
      "balance_loss_mlp": 1.02153254,
      "epoch": 0.3640954728551674,
      "flos": 21506505995520.0,
      "grad_norm": 2.68339788364002,
      "language_loss": 0.83121228,
      "learning_rate": 2.939141814846179e-06,
      "loss": 0.85343552,
      "num_input_tokens_seen": 65021215,
      "step": 3028,
      "time_per_iteration": 2.5756731033325195
    },
    {
      "auxiliary_loss_clip": 0.01194698,
      "auxiliary_loss_mlp": 0.01033615,
      "balance_loss_clip": 1.05799139,
      "balance_loss_mlp": 1.02435875,
      "epoch": 0.3642157157458065,
      "flos": 17712867081600.0,
      "grad_norm": 1.7177487981554644,
      "language_loss": 0.82421392,
      "learning_rate": 2.938453993641938e-06,
      "loss": 0.84649706,
      "num_input_tokens_seen": 65039590,
      "step": 3029,
      "time_per_iteration": 2.653989315032959
    },
    {
      "auxiliary_loss_clip": 0.01196676,
      "auxiliary_loss_mlp": 0.01031872,
      "balance_loss_clip": 1.06082666,
      "balance_loss_mlp": 1.02235889,
      "epoch": 0.36433595863644563,
      "flos": 17639537466240.0,
      "grad_norm": 2.418277535521276,
      "language_loss": 0.70131999,
      "learning_rate": 2.937766030074973e-06,
      "loss": 0.72360551,
      "num_input_tokens_seen": 65056845,
      "step": 3030,
      "time_per_iteration": 2.631835699081421
    },
    {
      "auxiliary_loss_clip": 0.01194724,
      "auxiliary_loss_mlp": 0.01033177,
      "balance_loss_clip": 1.05763602,
      "balance_loss_mlp": 1.02417684,
      "epoch": 0.3644562015270847,
      "flos": 26833279161600.0,
      "grad_norm": 2.8209518867497416,
      "language_loss": 0.82656467,
      "learning_rate": 2.937077924249646e-06,
      "loss": 0.84884357,
      "num_input_tokens_seen": 65079435,
      "step": 3031,
      "time_per_iteration": 2.69389271736145
    },
    {
      "auxiliary_loss_clip": 0.0119688,
      "auxiliary_loss_mlp": 0.01031291,
      "balance_loss_clip": 1.05921161,
      "balance_loss_mlp": 1.02175474,
      "epoch": 0.3645764444177238,
      "flos": 14282715847680.0,
      "grad_norm": 2.1286591439958373,
      "language_loss": 0.75497925,
      "learning_rate": 2.9363896762703443e-06,
      "loss": 0.77726096,
      "num_input_tokens_seen": 65096500,
      "step": 3032,
      "time_per_iteration": 2.6283388137817383
    },
    {
      "auxiliary_loss_clip": 0.01196191,
      "auxiliary_loss_mlp": 0.01033414,
      "balance_loss_clip": 1.05898213,
      "balance_loss_mlp": 1.02333522,
      "epoch": 0.3646966873083629,
      "flos": 20667489137280.0,
      "grad_norm": 1.6768813726982024,
      "language_loss": 0.84441417,
      "learning_rate": 2.9357012862414725e-06,
      "loss": 0.86671025,
      "num_input_tokens_seen": 65115860,
      "step": 3033,
      "time_per_iteration": 2.668241024017334
    },
    {
      "auxiliary_loss_clip": 0.01195526,
      "auxiliary_loss_mlp": 0.01032867,
      "balance_loss_clip": 1.06022263,
      "balance_loss_mlp": 1.02408683,
      "epoch": 0.36481693019900197,
      "flos": 27782613665280.0,
      "grad_norm": 1.877395952136515,
      "language_loss": 0.71270424,
      "learning_rate": 2.9350127542674593e-06,
      "loss": 0.73498821,
      "num_input_tokens_seen": 65138070,
      "step": 3034,
      "time_per_iteration": 2.678299903869629
    },
    {
      "auxiliary_loss_clip": 0.01195645,
      "auxiliary_loss_mlp": 0.01034738,
      "balance_loss_clip": 1.05749893,
      "balance_loss_mlp": 1.02424133,
      "epoch": 0.3649371730896411,
      "flos": 19712588025600.0,
      "grad_norm": 1.8355357567568191,
      "language_loss": 0.76365119,
      "learning_rate": 2.934324080452755e-06,
      "loss": 0.78595507,
      "num_input_tokens_seen": 65155860,
      "step": 3035,
      "time_per_iteration": 2.6031298637390137
    },
    {
      "auxiliary_loss_clip": 0.01197782,
      "auxiliary_loss_mlp": 0.01031065,
      "balance_loss_clip": 1.05864143,
      "balance_loss_mlp": 1.02076507,
      "epoch": 0.3650574159802802,
      "flos": 24750496016640.0,
      "grad_norm": 1.6538990548460144,
      "language_loss": 0.77881402,
      "learning_rate": 2.9336352649018307e-06,
      "loss": 0.80110246,
      "num_input_tokens_seen": 65175930,
      "step": 3036,
      "time_per_iteration": 2.6546945571899414
    },
    {
      "auxiliary_loss_clip": 0.01196815,
      "auxiliary_loss_mlp": 0.01039219,
      "balance_loss_clip": 1.06166458,
      "balance_loss_mlp": 1.02918124,
      "epoch": 0.36517765887091924,
      "flos": 32853487363200.0,
      "grad_norm": 1.7383627148435497,
      "language_loss": 0.70354229,
      "learning_rate": 2.9329463077191783e-06,
      "loss": 0.72590268,
      "num_input_tokens_seen": 65199305,
      "step": 3037,
      "time_per_iteration": 2.7685258388519287
    },
    {
      "auxiliary_loss_clip": 0.01197128,
      "auxiliary_loss_mlp": 0.01028915,
      "balance_loss_clip": 1.05987048,
      "balance_loss_mlp": 1.0187943,
      "epoch": 0.36529790176155835,
      "flos": 20120318282880.0,
      "grad_norm": 2.4668876485177784,
      "language_loss": 0.6372807,
      "learning_rate": 2.9322572090093135e-06,
      "loss": 0.65954113,
      "num_input_tokens_seen": 65218010,
      "step": 3038,
      "time_per_iteration": 2.635016441345215
    },
    {
      "auxiliary_loss_clip": 0.01194538,
      "auxiliary_loss_mlp": 0.01031702,
      "balance_loss_clip": 1.05692506,
      "balance_loss_mlp": 1.0222429,
      "epoch": 0.36541814465219746,
      "flos": 17639573379840.0,
      "grad_norm": 2.885370200761948,
      "language_loss": 0.76093912,
      "learning_rate": 2.9315679688767713e-06,
      "loss": 0.78320152,
      "num_input_tokens_seen": 65236020,
      "step": 3039,
      "time_per_iteration": 2.602647066116333
    },
    {
      "auxiliary_loss_clip": 0.01194385,
      "auxiliary_loss_mlp": 0.0103074,
      "balance_loss_clip": 1.05848145,
      "balance_loss_mlp": 1.02223992,
      "epoch": 0.3655383875428365,
      "flos": 22674356887680.0,
      "grad_norm": 1.5548256865610466,
      "language_loss": 0.6668613,
      "learning_rate": 2.9308785874261085e-06,
      "loss": 0.68911254,
      "num_input_tokens_seen": 65256210,
      "step": 3040,
      "time_per_iteration": 2.648352861404419
    },
    {
      "auxiliary_loss_clip": 0.01196428,
      "auxiliary_loss_mlp": 0.01034381,
      "balance_loss_clip": 1.06011057,
      "balance_loss_mlp": 1.02485621,
      "epoch": 0.36565863043347563,
      "flos": 21981173247360.0,
      "grad_norm": 2.863080809803847,
      "language_loss": 0.81619489,
      "learning_rate": 2.9301890647619045e-06,
      "loss": 0.838503,
      "num_input_tokens_seen": 65275505,
      "step": 3041,
      "time_per_iteration": 3.547750234603882
    },
    {
      "auxiliary_loss_clip": 0.01198497,
      "auxiliary_loss_mlp": 0.01030843,
      "balance_loss_clip": 1.05981243,
      "balance_loss_mlp": 1.02122831,
      "epoch": 0.36577887332411474,
      "flos": 24827632473600.0,
      "grad_norm": 2.175968447887509,
      "language_loss": 0.80133861,
      "learning_rate": 2.929499400988759e-06,
      "loss": 0.82363206,
      "num_input_tokens_seen": 65296665,
      "step": 3042,
      "time_per_iteration": 2.6615169048309326
    },
    {
      "auxiliary_loss_clip": 0.01198012,
      "auxiliary_loss_mlp": 0.01035019,
      "balance_loss_clip": 1.05884576,
      "balance_loss_mlp": 1.02480268,
      "epoch": 0.3658991162147538,
      "flos": 28293191539200.0,
      "grad_norm": 1.9931239248207593,
      "language_loss": 0.64980251,
      "learning_rate": 2.9288095962112927e-06,
      "loss": 0.67213279,
      "num_input_tokens_seen": 65317370,
      "step": 3043,
      "time_per_iteration": 2.7101292610168457
    },
    {
      "auxiliary_loss_clip": 0.01195056,
      "auxiliary_loss_mlp": 0.010316,
      "balance_loss_clip": 1.06006956,
      "balance_loss_mlp": 1.02197373,
      "epoch": 0.3660193591053929,
      "flos": 17785550252160.0,
      "grad_norm": 2.452637063705962,
      "language_loss": 0.85309494,
      "learning_rate": 2.9281196505341503e-06,
      "loss": 0.87536144,
      "num_input_tokens_seen": 65334540,
      "step": 3044,
      "time_per_iteration": 4.522130250930786
    },
    {
      "auxiliary_loss_clip": 0.01193029,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.05919576,
      "balance_loss_mlp": 1.01825976,
      "epoch": 0.36613960199603196,
      "flos": 10342776839040.0,
      "grad_norm": 2.456295922769481,
      "language_loss": 0.78515875,
      "learning_rate": 2.9274295640619946e-06,
      "loss": 0.80736566,
      "num_input_tokens_seen": 65351670,
      "step": 3045,
      "time_per_iteration": 3.5028398036956787
    },
    {
      "auxiliary_loss_clip": 0.01190814,
      "auxiliary_loss_mlp": 0.01029268,
      "balance_loss_clip": 1.05610585,
      "balance_loss_mlp": 1.02079821,
      "epoch": 0.36625984488667107,
      "flos": 19755609540480.0,
      "grad_norm": 1.699786037234054,
      "language_loss": 0.78393531,
      "learning_rate": 2.9267393368995103e-06,
      "loss": 0.80613613,
      "num_input_tokens_seen": 65370900,
      "step": 3046,
      "time_per_iteration": 2.6437649726867676
    },
    {
      "auxiliary_loss_clip": 0.0119888,
      "auxiliary_loss_mlp": 0.01037047,
      "balance_loss_clip": 1.06119633,
      "balance_loss_mlp": 1.02759969,
      "epoch": 0.3663800877773102,
      "flos": 17674262939520.0,
      "grad_norm": 2.207704618084548,
      "language_loss": 0.74419332,
      "learning_rate": 2.926048969151407e-06,
      "loss": 0.76655257,
      "num_input_tokens_seen": 65388185,
      "step": 3047,
      "time_per_iteration": 2.7108030319213867
    },
    {
      "auxiliary_loss_clip": 0.01198037,
      "auxiliary_loss_mlp": 0.0102974,
      "balance_loss_clip": 1.06082273,
      "balance_loss_mlp": 1.02010822,
      "epoch": 0.36650033066794924,
      "flos": 20303606407680.0,
      "grad_norm": 1.9482877516061103,
      "language_loss": 0.68315899,
      "learning_rate": 2.92535846092241e-06,
      "loss": 0.70543671,
      "num_input_tokens_seen": 65407200,
      "step": 3048,
      "time_per_iteration": 2.643282175064087
    },
    {
      "auxiliary_loss_clip": 0.01197278,
      "auxiliary_loss_mlp": 0.01036187,
      "balance_loss_clip": 1.05809546,
      "balance_loss_mlp": 1.02663815,
      "epoch": 0.36662057355858835,
      "flos": 24716237420160.0,
      "grad_norm": 1.6579841611646209,
      "language_loss": 0.82503808,
      "learning_rate": 2.9246678123172704e-06,
      "loss": 0.84737277,
      "num_input_tokens_seen": 65427290,
      "step": 3049,
      "time_per_iteration": 2.6467835903167725
    },
    {
      "auxiliary_loss_clip": 0.01194666,
      "auxiliary_loss_mlp": 0.01033553,
      "balance_loss_clip": 1.05710244,
      "balance_loss_mlp": 1.02357483,
      "epoch": 0.36674081644922746,
      "flos": 12385267902720.0,
      "grad_norm": 2.2997180287342336,
      "language_loss": 0.74487734,
      "learning_rate": 2.9239770234407596e-06,
      "loss": 0.76715958,
      "num_input_tokens_seen": 65445595,
      "step": 3050,
      "time_per_iteration": 2.5836234092712402
    },
    {
      "auxiliary_loss_clip": 0.0119767,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.06038451,
      "balance_loss_mlp": 1.02014327,
      "epoch": 0.3668610593398665,
      "flos": 21105922544640.0,
      "grad_norm": 1.7057419462828929,
      "language_loss": 0.68627584,
      "learning_rate": 2.9232860943976686e-06,
      "loss": 0.70855039,
      "num_input_tokens_seen": 65466330,
      "step": 3051,
      "time_per_iteration": 2.6679911613464355
    },
    {
      "auxiliary_loss_clip": 0.01191752,
      "auxiliary_loss_mlp": 0.01029432,
      "balance_loss_clip": 1.05831432,
      "balance_loss_mlp": 1.02071786,
      "epoch": 0.3669813022305056,
      "flos": 26758082039040.0,
      "grad_norm": 1.6633202844538113,
      "language_loss": 0.84098566,
      "learning_rate": 2.9225950252928115e-06,
      "loss": 0.86319745,
      "num_input_tokens_seen": 65487180,
      "step": 3052,
      "time_per_iteration": 2.6649556159973145
    },
    {
      "auxiliary_loss_clip": 0.01195065,
      "auxiliary_loss_mlp": 0.01032929,
      "balance_loss_clip": 1.05941963,
      "balance_loss_mlp": 1.02348804,
      "epoch": 0.36710154512114473,
      "flos": 19099521671040.0,
      "grad_norm": 1.9274298430509,
      "language_loss": 0.81188738,
      "learning_rate": 2.9219038162310217e-06,
      "loss": 0.8341673,
      "num_input_tokens_seen": 65505380,
      "step": 3053,
      "time_per_iteration": 2.742309331893921
    },
    {
      "auxiliary_loss_clip": 0.01197848,
      "auxiliary_loss_mlp": 0.01032443,
      "balance_loss_clip": 1.06060624,
      "balance_loss_mlp": 1.02304375,
      "epoch": 0.3672217880117838,
      "flos": 20812029465600.0,
      "grad_norm": 2.1436332784426497,
      "language_loss": 0.8256644,
      "learning_rate": 2.921212467317157e-06,
      "loss": 0.84796727,
      "num_input_tokens_seen": 65524825,
      "step": 3054,
      "time_per_iteration": 2.5928430557250977
    },
    {
      "auxiliary_loss_clip": 0.01197468,
      "auxiliary_loss_mlp": 0.01033439,
      "balance_loss_clip": 1.05877149,
      "balance_loss_mlp": 1.0229311,
      "epoch": 0.3673420309024229,
      "flos": 13590394133760.0,
      "grad_norm": 2.117823389625948,
      "language_loss": 0.79836524,
      "learning_rate": 2.920520978656093e-06,
      "loss": 0.82067424,
      "num_input_tokens_seen": 65541790,
      "step": 3055,
      "time_per_iteration": 2.6172633171081543
    },
    {
      "auxiliary_loss_clip": 0.01191528,
      "auxiliary_loss_mlp": 0.01033808,
      "balance_loss_clip": 1.05700922,
      "balance_loss_mlp": 1.02486706,
      "epoch": 0.367462273793062,
      "flos": 28986877969920.0,
      "grad_norm": 1.8672618534804004,
      "language_loss": 0.76683778,
      "learning_rate": 2.919829350352729e-06,
      "loss": 0.78909111,
      "num_input_tokens_seen": 65563395,
      "step": 3056,
      "time_per_iteration": 2.665243148803711
    },
    {
      "auxiliary_loss_clip": 0.01101396,
      "auxiliary_loss_mlp": 0.01011688,
      "balance_loss_clip": 1.03191197,
      "balance_loss_mlp": 1.01011431,
      "epoch": 0.36758251668370107,
      "flos": 62643148346880.0,
      "grad_norm": 0.8000738078681828,
      "language_loss": 0.60041428,
      "learning_rate": 2.919137582511983e-06,
      "loss": 0.62154508,
      "num_input_tokens_seen": 65619835,
      "step": 3057,
      "time_per_iteration": 3.1241326332092285
    },
    {
      "auxiliary_loss_clip": 0.0119752,
      "auxiliary_loss_mlp": 0.01036888,
      "balance_loss_clip": 1.06060028,
      "balance_loss_mlp": 1.0276432,
      "epoch": 0.3677027595743402,
      "flos": 12713886455040.0,
      "grad_norm": 2.0038882578611408,
      "language_loss": 0.6404866,
      "learning_rate": 2.918445675238797e-06,
      "loss": 0.66283071,
      "num_input_tokens_seen": 65636760,
      "step": 3058,
      "time_per_iteration": 2.6496851444244385
    },
    {
      "auxiliary_loss_clip": 0.01191346,
      "auxiliary_loss_mlp": 0.01026784,
      "balance_loss_clip": 1.05461693,
      "balance_loss_mlp": 1.01772988,
      "epoch": 0.36782300246497923,
      "flos": 25046579825280.0,
      "grad_norm": 1.7380094127961754,
      "language_loss": 0.69663441,
      "learning_rate": 2.917753628638132e-06,
      "loss": 0.7188158,
      "num_input_tokens_seen": 65657065,
      "step": 3059,
      "time_per_iteration": 2.705693006515503
    },
    {
      "auxiliary_loss_clip": 0.01196405,
      "auxiliary_loss_mlp": 0.01035472,
      "balance_loss_clip": 1.06101644,
      "balance_loss_mlp": 1.02538133,
      "epoch": 0.36794324535561834,
      "flos": 17419512706560.0,
      "grad_norm": 3.2506754552174337,
      "language_loss": 0.70353818,
      "learning_rate": 2.9170614428149716e-06,
      "loss": 0.7258569,
      "num_input_tokens_seen": 65675400,
      "step": 3060,
      "time_per_iteration": 2.6108596324920654
    },
    {
      "auxiliary_loss_clip": 0.01198443,
      "auxiliary_loss_mlp": 0.01035728,
      "balance_loss_clip": 1.06297183,
      "balance_loss_mlp": 1.02496886,
      "epoch": 0.36806348824625745,
      "flos": 24089128848000.0,
      "grad_norm": 2.4443408045340824,
      "language_loss": 0.86731172,
      "learning_rate": 2.9163691178743195e-06,
      "loss": 0.88965344,
      "num_input_tokens_seen": 65694050,
      "step": 3061,
      "time_per_iteration": 2.681537389755249
    },
    {
      "auxiliary_loss_clip": 0.01196096,
      "auxiliary_loss_mlp": 0.01026889,
      "balance_loss_clip": 1.06019664,
      "balance_loss_mlp": 1.01850891,
      "epoch": 0.3681837311368965,
      "flos": 20521871400960.0,
      "grad_norm": 1.8628408247670665,
      "language_loss": 0.77301049,
      "learning_rate": 2.9156766539212006e-06,
      "loss": 0.79524034,
      "num_input_tokens_seen": 65711695,
      "step": 3062,
      "time_per_iteration": 2.614450454711914
    },
    {
      "auxiliary_loss_clip": 0.01191607,
      "auxiliary_loss_mlp": 0.0103049,
      "balance_loss_clip": 1.05532217,
      "balance_loss_mlp": 1.02153778,
      "epoch": 0.3683039740275356,
      "flos": 21466644877440.0,
      "grad_norm": 1.9340439136703964,
      "language_loss": 0.71786416,
      "learning_rate": 2.9149840510606614e-06,
      "loss": 0.74008512,
      "num_input_tokens_seen": 65730350,
      "step": 3063,
      "time_per_iteration": 2.7144503593444824
    },
    {
      "auxiliary_loss_clip": 0.01099274,
      "auxiliary_loss_mlp": 0.01001642,
      "balance_loss_clip": 1.02997613,
      "balance_loss_mlp": 1.00019348,
      "epoch": 0.36842421691817473,
      "flos": 70380999987840.0,
      "grad_norm": 1.0488886646415294,
      "language_loss": 0.64172012,
      "learning_rate": 2.914291309397769e-06,
      "loss": 0.66272932,
      "num_input_tokens_seen": 65787820,
      "step": 3064,
      "time_per_iteration": 3.3071370124816895
    },
    {
      "auxiliary_loss_clip": 0.01195537,
      "auxiliary_loss_mlp": 0.0103417,
      "balance_loss_clip": 1.05751967,
      "balance_loss_mlp": 1.02472258,
      "epoch": 0.3685444598088138,
      "flos": 23331378510720.0,
      "grad_norm": 2.2190439871320087,
      "language_loss": 0.78527975,
      "learning_rate": 2.9135984290376117e-06,
      "loss": 0.80757689,
      "num_input_tokens_seen": 65806685,
      "step": 3065,
      "time_per_iteration": 2.6785058975219727
    },
    {
      "auxiliary_loss_clip": 0.01198,
      "auxiliary_loss_mlp": 0.0103053,
      "balance_loss_clip": 1.06195724,
      "balance_loss_mlp": 1.02113628,
      "epoch": 0.3686647026994529,
      "flos": 23070271570560.0,
      "grad_norm": 1.6815244689850284,
      "language_loss": 0.8253358,
      "learning_rate": 2.9129054100853e-06,
      "loss": 0.84762108,
      "num_input_tokens_seen": 65825525,
      "step": 3066,
      "time_per_iteration": 2.6454038619995117
    },
    {
      "auxiliary_loss_clip": 0.01194308,
      "auxiliary_loss_mlp": 0.01031274,
      "balance_loss_clip": 1.05547309,
      "balance_loss_mlp": 1.02169502,
      "epoch": 0.368784945590092,
      "flos": 25119909440640.0,
      "grad_norm": 1.8666413199587941,
      "language_loss": 0.76074409,
      "learning_rate": 2.912212252645963e-06,
      "loss": 0.78299987,
      "num_input_tokens_seen": 65848110,
      "step": 3067,
      "time_per_iteration": 2.608041763305664
    },
    {
      "auxiliary_loss_clip": 0.01200485,
      "auxiliary_loss_mlp": 0.01042232,
      "balance_loss_clip": 1.06119895,
      "balance_loss_mlp": 1.03137183,
      "epoch": 0.36890518848073106,
      "flos": 18442284566400.0,
      "grad_norm": 2.4256517779344717,
      "language_loss": 0.76719832,
      "learning_rate": 2.9115189568247523e-06,
      "loss": 0.78962553,
      "num_input_tokens_seen": 65865670,
      "step": 3068,
      "time_per_iteration": 3.5122339725494385
    },
    {
      "auxiliary_loss_clip": 0.01194142,
      "auxiliary_loss_mlp": 0.01033398,
      "balance_loss_clip": 1.059057,
      "balance_loss_mlp": 1.02416539,
      "epoch": 0.36902543137137017,
      "flos": 16362446336640.0,
      "grad_norm": 2.421036579062912,
      "language_loss": 0.92220163,
      "learning_rate": 2.910825522726841e-06,
      "loss": 0.94447708,
      "num_input_tokens_seen": 65883195,
      "step": 3069,
      "time_per_iteration": 2.586845874786377
    },
    {
      "auxiliary_loss_clip": 0.01191889,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.05723858,
      "balance_loss_mlp": 1.01894498,
      "epoch": 0.3691456742620093,
      "flos": 12275596702080.0,
      "grad_norm": 2.181427926883091,
      "language_loss": 0.77082354,
      "learning_rate": 2.9101319504574215e-06,
      "loss": 0.79302752,
      "num_input_tokens_seen": 65899635,
      "step": 3070,
      "time_per_iteration": 4.554910182952881
    },
    {
      "auxiliary_loss_clip": 0.01196059,
      "auxiliary_loss_mlp": 0.01032446,
      "balance_loss_clip": 1.05879247,
      "balance_loss_mlp": 1.02257586,
      "epoch": 0.36926591715264834,
      "flos": 17786412178560.0,
      "grad_norm": 1.741835054376564,
      "language_loss": 0.76225424,
      "learning_rate": 2.909438240121709e-06,
      "loss": 0.78453934,
      "num_input_tokens_seen": 65919910,
      "step": 3071,
      "time_per_iteration": 3.54675030708313
    },
    {
      "auxiliary_loss_clip": 0.01194942,
      "auxiliary_loss_mlp": 0.01034949,
      "balance_loss_clip": 1.05827105,
      "balance_loss_mlp": 1.02530503,
      "epoch": 0.36938616004328745,
      "flos": 28948309741440.0,
      "grad_norm": 3.1648243858203084,
      "language_loss": 0.70403993,
      "learning_rate": 2.908744391824939e-06,
      "loss": 0.72633886,
      "num_input_tokens_seen": 65940930,
      "step": 3072,
      "time_per_iteration": 2.6920220851898193
    },
    {
      "auxiliary_loss_clip": 0.01194988,
      "auxiliary_loss_mlp": 0.01040118,
      "balance_loss_clip": 1.05732989,
      "balance_loss_mlp": 1.03008628,
      "epoch": 0.36950640293392656,
      "flos": 29205394358400.0,
      "grad_norm": 1.8441486126877242,
      "language_loss": 0.7950536,
      "learning_rate": 2.908050405672367e-06,
      "loss": 0.81740475,
      "num_input_tokens_seen": 65960475,
      "step": 3073,
      "time_per_iteration": 2.6791460514068604
    },
    {
      "auxiliary_loss_clip": 0.01194731,
      "auxiliary_loss_mlp": 0.01032599,
      "balance_loss_clip": 1.05705762,
      "balance_loss_mlp": 1.02286577,
      "epoch": 0.3696266458245656,
      "flos": 24827776128000.0,
      "grad_norm": 2.2980039487167505,
      "language_loss": 0.79256922,
      "learning_rate": 2.9073562817692703e-06,
      "loss": 0.81484252,
      "num_input_tokens_seen": 65979160,
      "step": 3074,
      "time_per_iteration": 2.6868886947631836
    },
    {
      "auxiliary_loss_clip": 0.01097622,
      "auxiliary_loss_mlp": 0.01000921,
      "balance_loss_clip": 1.02849364,
      "balance_loss_mlp": 0.99941272,
      "epoch": 0.3697468887152047,
      "flos": 59887257264000.0,
      "grad_norm": 0.7188882807199584,
      "language_loss": 0.56554282,
      "learning_rate": 2.9066620202209468e-06,
      "loss": 0.5865283,
      "num_input_tokens_seen": 66041650,
      "step": 3075,
      "time_per_iteration": 3.1908915042877197
    },
    {
      "auxiliary_loss_clip": 0.01195406,
      "auxiliary_loss_mlp": 0.01029125,
      "balance_loss_clip": 1.05943012,
      "balance_loss_mlp": 1.01968384,
      "epoch": 0.3698671316058438,
      "flos": 26137581569280.0,
      "grad_norm": 2.1882423765453307,
      "language_loss": 0.78165758,
      "learning_rate": 2.905967621132716e-06,
      "loss": 0.80390286,
      "num_input_tokens_seen": 66059260,
      "step": 3076,
      "time_per_iteration": 2.7177860736846924
    },
    {
      "auxiliary_loss_clip": 0.01198105,
      "auxiliary_loss_mlp": 0.01036562,
      "balance_loss_clip": 1.05994511,
      "balance_loss_mlp": 1.02628565,
      "epoch": 0.3699873744964829,
      "flos": 24607464059520.0,
      "grad_norm": 1.9551062720617947,
      "language_loss": 0.75282359,
      "learning_rate": 2.9052730846099172e-06,
      "loss": 0.77517021,
      "num_input_tokens_seen": 66080605,
      "step": 3077,
      "time_per_iteration": 2.6283762454986572
    },
    {
      "auxiliary_loss_clip": 0.01095465,
      "auxiliary_loss_mlp": 0.01001317,
      "balance_loss_clip": 1.02639997,
      "balance_loss_mlp": 0.99985045,
      "epoch": 0.370107617387122,
      "flos": 64885340050560.0,
      "grad_norm": 0.860325727340029,
      "language_loss": 0.60856915,
      "learning_rate": 2.9045784107579123e-06,
      "loss": 0.62953699,
      "num_input_tokens_seen": 66140710,
      "step": 3078,
      "time_per_iteration": 3.201054811477661
    },
    {
      "auxiliary_loss_clip": 0.01196729,
      "auxiliary_loss_mlp": 0.01031289,
      "balance_loss_clip": 1.05987978,
      "balance_loss_mlp": 1.02223492,
      "epoch": 0.37022786027776106,
      "flos": 15961683317760.0,
      "grad_norm": 1.8291034580605254,
      "language_loss": 0.66909301,
      "learning_rate": 2.9038835996820807e-06,
      "loss": 0.69137323,
      "num_input_tokens_seen": 66158320,
      "step": 3079,
      "time_per_iteration": 2.622363567352295
    },
    {
      "auxiliary_loss_clip": 0.0119555,
      "auxiliary_loss_mlp": 0.0103515,
      "balance_loss_clip": 1.06021917,
      "balance_loss_mlp": 1.02603018,
      "epoch": 0.37034810316840017,
      "flos": 18546927863040.0,
      "grad_norm": 1.7596893071756698,
      "language_loss": 0.7949962,
      "learning_rate": 2.903188651487826e-06,
      "loss": 0.81730318,
      "num_input_tokens_seen": 66176875,
      "step": 3080,
      "time_per_iteration": 2.5890400409698486
    },
    {
      "auxiliary_loss_clip": 0.01198126,
      "auxiliary_loss_mlp": 0.01039649,
      "balance_loss_clip": 1.06135631,
      "balance_loss_mlp": 1.03048229,
      "epoch": 0.3704683460590393,
      "flos": 17821927751040.0,
      "grad_norm": 2.0307879001992415,
      "language_loss": 0.86587179,
      "learning_rate": 2.902493566280571e-06,
      "loss": 0.88824958,
      "num_input_tokens_seen": 66194980,
      "step": 3081,
      "time_per_iteration": 2.676372766494751
    },
    {
      "auxiliary_loss_clip": 0.01197396,
      "auxiliary_loss_mlp": 0.01038677,
      "balance_loss_clip": 1.05787826,
      "balance_loss_mlp": 1.02787662,
      "epoch": 0.37058858894967833,
      "flos": 14134081368960.0,
      "grad_norm": 6.535223595217558,
      "language_loss": 0.81171513,
      "learning_rate": 2.9017983441657595e-06,
      "loss": 0.83407587,
      "num_input_tokens_seen": 66212310,
      "step": 3082,
      "time_per_iteration": 2.604828119277954
    },
    {
      "auxiliary_loss_clip": 0.01195038,
      "auxiliary_loss_mlp": 0.0103279,
      "balance_loss_clip": 1.0584259,
      "balance_loss_mlp": 1.02304447,
      "epoch": 0.37070883184031744,
      "flos": 13954492344960.0,
      "grad_norm": 2.3596631441432727,
      "language_loss": 0.7549324,
      "learning_rate": 2.9011029852488564e-06,
      "loss": 0.77721065,
      "num_input_tokens_seen": 66229545,
      "step": 3083,
      "time_per_iteration": 2.654202699661255
    },
    {
      "auxiliary_loss_clip": 0.01093754,
      "auxiliary_loss_mlp": 0.01002315,
      "balance_loss_clip": 1.02457571,
      "balance_loss_mlp": 1.00081849,
      "epoch": 0.37082907473095655,
      "flos": 52315419306240.0,
      "grad_norm": 0.981235615379663,
      "language_loss": 0.62439787,
      "learning_rate": 2.9004074896353465e-06,
      "loss": 0.64535856,
      "num_input_tokens_seen": 66283545,
      "step": 3084,
      "time_per_iteration": 3.087414264678955
    },
    {
      "auxiliary_loss_clip": 0.01197368,
      "auxiliary_loss_mlp": 0.01034385,
      "balance_loss_clip": 1.06370592,
      "balance_loss_mlp": 1.02482462,
      "epoch": 0.3709493176215956,
      "flos": 15998096730240.0,
      "grad_norm": 1.8451963316300986,
      "language_loss": 0.81688547,
      "learning_rate": 2.8997118574307362e-06,
      "loss": 0.839203,
      "num_input_tokens_seen": 66300500,
      "step": 3085,
      "time_per_iteration": 2.615442991256714
    },
    {
      "auxiliary_loss_clip": 0.01197359,
      "auxiliary_loss_mlp": 0.01037304,
      "balance_loss_clip": 1.0606215,
      "balance_loss_mlp": 1.02815437,
      "epoch": 0.3710695605122347,
      "flos": 20959837931520.0,
      "grad_norm": 2.0338270427780847,
      "language_loss": 0.74196261,
      "learning_rate": 2.899016088740553e-06,
      "loss": 0.76430929,
      "num_input_tokens_seen": 66318610,
      "step": 3086,
      "time_per_iteration": 2.6155054569244385
    },
    {
      "auxiliary_loss_clip": 0.01191688,
      "auxiliary_loss_mlp": 0.01034448,
      "balance_loss_clip": 1.05611444,
      "balance_loss_mlp": 1.02577519,
      "epoch": 0.37118980340287383,
      "flos": 14355578586240.0,
      "grad_norm": 2.151561403413462,
      "language_loss": 0.79440612,
      "learning_rate": 2.898320183670344e-06,
      "loss": 0.8166675,
      "num_input_tokens_seen": 66336025,
      "step": 3087,
      "time_per_iteration": 2.8234939575195312
    },
    {
      "auxiliary_loss_clip": 0.01196632,
      "auxiliary_loss_mlp": 0.01044132,
      "balance_loss_clip": 1.05888605,
      "balance_loss_mlp": 1.03387451,
      "epoch": 0.3713100462935129,
      "flos": 25885381201920.0,
      "grad_norm": 1.8168554341162062,
      "language_loss": 0.89117217,
      "learning_rate": 2.8976241423256767e-06,
      "loss": 0.91357982,
      "num_input_tokens_seen": 66356120,
      "step": 3088,
      "time_per_iteration": 2.6288745403289795
    },
    {
      "auxiliary_loss_clip": 0.01194597,
      "auxiliary_loss_mlp": 0.01045521,
      "balance_loss_clip": 1.06144273,
      "balance_loss_mlp": 1.03610325,
      "epoch": 0.371430289184152,
      "flos": 30518934814080.0,
      "grad_norm": 2.0824973862191003,
      "language_loss": 0.68149602,
      "learning_rate": 2.896927964812142e-06,
      "loss": 0.70389712,
      "num_input_tokens_seen": 66376685,
      "step": 3089,
      "time_per_iteration": 2.7149393558502197
    },
    {
      "auxiliary_loss_clip": 0.01196326,
      "auxiliary_loss_mlp": 0.01029919,
      "balance_loss_clip": 1.06059957,
      "balance_loss_mlp": 1.02044797,
      "epoch": 0.37155053207479105,
      "flos": 15742233175680.0,
      "grad_norm": 2.4652547855829194,
      "language_loss": 0.75118846,
      "learning_rate": 2.8962316512353465e-06,
      "loss": 0.77345097,
      "num_input_tokens_seen": 66394230,
      "step": 3090,
      "time_per_iteration": 2.597835063934326
    },
    {
      "auxiliary_loss_clip": 0.01194754,
      "auxiliary_loss_mlp": 0.01029367,
      "balance_loss_clip": 1.05884063,
      "balance_loss_mlp": 1.01983035,
      "epoch": 0.37167077496543016,
      "flos": 23404061681280.0,
      "grad_norm": 1.656483651582557,
      "language_loss": 0.74540138,
      "learning_rate": 2.8955352017009233e-06,
      "loss": 0.76764262,
      "num_input_tokens_seen": 66413475,
      "step": 3091,
      "time_per_iteration": 2.6477880477905273
    },
    {
      "auxiliary_loss_clip": 0.01197634,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 1.06100857,
      "balance_loss_mlp": 1.02259493,
      "epoch": 0.3717910178560693,
      "flos": 22088653718400.0,
      "grad_norm": 2.4781799513719034,
      "language_loss": 0.77654135,
      "learning_rate": 2.8948386163145212e-06,
      "loss": 0.79883516,
      "num_input_tokens_seen": 66432685,
      "step": 3092,
      "time_per_iteration": 2.649484872817993
    },
    {
      "auxiliary_loss_clip": 0.01195572,
      "auxiliary_loss_mlp": 0.01028805,
      "balance_loss_clip": 1.05976343,
      "balance_loss_mlp": 1.01966178,
      "epoch": 0.3719112607467083,
      "flos": 26939969533440.0,
      "grad_norm": 1.9670704812287394,
      "language_loss": 0.78936851,
      "learning_rate": 2.8941418951818135e-06,
      "loss": 0.81161225,
      "num_input_tokens_seen": 66452245,
      "step": 3093,
      "time_per_iteration": 2.679069995880127
    },
    {
      "auxiliary_loss_clip": 0.01194585,
      "auxiliary_loss_mlp": 0.01037061,
      "balance_loss_clip": 1.05858731,
      "balance_loss_mlp": 1.02829957,
      "epoch": 0.37203150363734744,
      "flos": 12166500119040.0,
      "grad_norm": 2.4087284174523207,
      "language_loss": 0.70603561,
      "learning_rate": 2.8934450384084903e-06,
      "loss": 0.72835207,
      "num_input_tokens_seen": 66469760,
      "step": 3094,
      "time_per_iteration": 3.5374600887298584
    },
    {
      "auxiliary_loss_clip": 0.01197775,
      "auxiliary_loss_mlp": 0.01036012,
      "balance_loss_clip": 1.05899107,
      "balance_loss_mlp": 1.02653456,
      "epoch": 0.37215174652798655,
      "flos": 23697595624320.0,
      "grad_norm": 1.9164347505755988,
      "language_loss": 0.69824028,
      "learning_rate": 2.8927480461002653e-06,
      "loss": 0.72057819,
      "num_input_tokens_seen": 66489730,
      "step": 3095,
      "time_per_iteration": 2.5980916023254395
    },
    {
      "auxiliary_loss_clip": 0.01199339,
      "auxiliary_loss_mlp": 0.01032419,
      "balance_loss_clip": 1.06086612,
      "balance_loss_mlp": 1.02307904,
      "epoch": 0.3722719894186256,
      "flos": 17887751424000.0,
      "grad_norm": 2.902676117146065,
      "language_loss": 0.85981131,
      "learning_rate": 2.892050918362872e-06,
      "loss": 0.88212889,
      "num_input_tokens_seen": 66504785,
      "step": 3096,
      "time_per_iteration": 3.5561697483062744
    },
    {
      "auxiliary_loss_clip": 0.01092209,
      "auxiliary_loss_mlp": 0.01001607,
      "balance_loss_clip": 1.0230782,
      "balance_loss_mlp": 1.0000571,
      "epoch": 0.3723922323092647,
      "flos": 62419891363200.0,
      "grad_norm": 0.8804027935367441,
      "language_loss": 0.55886757,
      "learning_rate": 2.8913536553020626e-06,
      "loss": 0.57980573,
      "num_input_tokens_seen": 66558840,
      "step": 3097,
      "time_per_iteration": 4.953164577484131
    },
    {
      "auxiliary_loss_clip": 0.01195449,
      "auxiliary_loss_mlp": 0.01030853,
      "balance_loss_clip": 1.05994654,
      "balance_loss_mlp": 1.02217424,
      "epoch": 0.3725124751999038,
      "flos": 23039747988480.0,
      "grad_norm": 1.9225812603551746,
      "language_loss": 0.8484627,
      "learning_rate": 2.8906562570236137e-06,
      "loss": 0.87072575,
      "num_input_tokens_seen": 66576750,
      "step": 3098,
      "time_per_iteration": 2.6355340480804443
    },
    {
      "auxiliary_loss_clip": 0.01191398,
      "auxiliary_loss_mlp": 0.01034079,
      "balance_loss_clip": 1.05743694,
      "balance_loss_mlp": 1.02491796,
      "epoch": 0.3726327180905429,
      "flos": 20920551431040.0,
      "grad_norm": 1.6751267134108736,
      "language_loss": 0.76259851,
      "learning_rate": 2.889958723633318e-06,
      "loss": 0.78485334,
      "num_input_tokens_seen": 66595690,
      "step": 3099,
      "time_per_iteration": 2.670198917388916
    },
    {
      "auxiliary_loss_clip": 0.01195024,
      "auxiliary_loss_mlp": 0.01028957,
      "balance_loss_clip": 1.05853486,
      "balance_loss_mlp": 1.01999271,
      "epoch": 0.372752960981182,
      "flos": 30592156688640.0,
      "grad_norm": 1.8270926370216947,
      "language_loss": 0.74258369,
      "learning_rate": 2.889261055236992e-06,
      "loss": 0.76482344,
      "num_input_tokens_seen": 66617905,
      "step": 3100,
      "time_per_iteration": 2.731271743774414
    },
    {
      "auxiliary_loss_clip": 0.01195006,
      "auxiliary_loss_mlp": 0.01033626,
      "balance_loss_clip": 1.05896139,
      "balance_loss_mlp": 1.02433348,
      "epoch": 0.3728732038718211,
      "flos": 25116749043840.0,
      "grad_norm": 1.9473235368133823,
      "language_loss": 0.8277114,
      "learning_rate": 2.8885632519404704e-06,
      "loss": 0.84999776,
      "num_input_tokens_seen": 66638175,
      "step": 3101,
      "time_per_iteration": 2.627164125442505
    },
    {
      "auxiliary_loss_clip": 0.01196259,
      "auxiliary_loss_mlp": 0.01031203,
      "balance_loss_clip": 1.05843139,
      "balance_loss_mlp": 1.02258384,
      "epoch": 0.37299344676246016,
      "flos": 25302048330240.0,
      "grad_norm": 1.9782554562934527,
      "language_loss": 0.75935447,
      "learning_rate": 2.8878653138496107e-06,
      "loss": 0.78162909,
      "num_input_tokens_seen": 66658670,
      "step": 3102,
      "time_per_iteration": 2.6581003665924072
    },
    {
      "auxiliary_loss_clip": 0.01195973,
      "auxiliary_loss_mlp": 0.01029617,
      "balance_loss_clip": 1.05963719,
      "balance_loss_mlp": 1.02081931,
      "epoch": 0.37311368965309927,
      "flos": 23842531002240.0,
      "grad_norm": 2.23858271245276,
      "language_loss": 0.76460618,
      "learning_rate": 2.8871672410702878e-06,
      "loss": 0.78686208,
      "num_input_tokens_seen": 66676030,
      "step": 3103,
      "time_per_iteration": 2.6459548473358154
    },
    {
      "auxiliary_loss_clip": 0.01197174,
      "auxiliary_loss_mlp": 0.01030086,
      "balance_loss_clip": 1.05941355,
      "balance_loss_mlp": 1.02000117,
      "epoch": 0.3732339325437384,
      "flos": 25811943845760.0,
      "grad_norm": 1.8965967230556569,
      "language_loss": 0.82210845,
      "learning_rate": 2.8864690337084008e-06,
      "loss": 0.84438097,
      "num_input_tokens_seen": 66695305,
      "step": 3104,
      "time_per_iteration": 2.7193281650543213
    },
    {
      "auxiliary_loss_clip": 0.01196356,
      "auxiliary_loss_mlp": 0.01030457,
      "balance_loss_clip": 1.0597651,
      "balance_loss_mlp": 1.02189767,
      "epoch": 0.37335417543437743,
      "flos": 26208433146240.0,
      "grad_norm": 2.09881074937054,
      "language_loss": 0.7826978,
      "learning_rate": 2.885770691869866e-06,
      "loss": 0.80496597,
      "num_input_tokens_seen": 66716185,
      "step": 3105,
      "time_per_iteration": 2.6688902378082275
    },
    {
      "auxiliary_loss_clip": 0.01193365,
      "auxiliary_loss_mlp": 0.01030312,
      "balance_loss_clip": 1.06062055,
      "balance_loss_mlp": 1.02219343,
      "epoch": 0.37347441832501654,
      "flos": 24023879792640.0,
      "grad_norm": 3.0152652065242327,
      "language_loss": 0.74303555,
      "learning_rate": 2.8850722156606207e-06,
      "loss": 0.76527226,
      "num_input_tokens_seen": 66734575,
      "step": 3106,
      "time_per_iteration": 2.7091445922851562
    },
    {
      "auxiliary_loss_clip": 0.01192597,
      "auxiliary_loss_mlp": 0.01029064,
      "balance_loss_clip": 1.05884564,
      "balance_loss_mlp": 1.02050447,
      "epoch": 0.3735946612156556,
      "flos": 19714922409600.0,
      "grad_norm": 1.651744560126585,
      "language_loss": 0.66766578,
      "learning_rate": 2.8843736051866252e-06,
      "loss": 0.6898824,
      "num_input_tokens_seen": 66753500,
      "step": 3107,
      "time_per_iteration": 2.6159937381744385
    },
    {
      "auxiliary_loss_clip": 0.01194811,
      "auxiliary_loss_mlp": 0.01027497,
      "balance_loss_clip": 1.06027269,
      "balance_loss_mlp": 1.01834726,
      "epoch": 0.3737149041062947,
      "flos": 23039604334080.0,
      "grad_norm": 1.7374463765150387,
      "language_loss": 0.68966115,
      "learning_rate": 2.8836748605538557e-06,
      "loss": 0.7118842,
      "num_input_tokens_seen": 66775140,
      "step": 3108,
      "time_per_iteration": 2.707636833190918
    },
    {
      "auxiliary_loss_clip": 0.01199053,
      "auxiliary_loss_mlp": 0.01027596,
      "balance_loss_clip": 1.0589813,
      "balance_loss_mlp": 1.01784515,
      "epoch": 0.3738351469969338,
      "flos": 34678108483200.0,
      "grad_norm": 2.873209208346813,
      "language_loss": 0.6321938,
      "learning_rate": 2.882975981868313e-06,
      "loss": 0.65446031,
      "num_input_tokens_seen": 66795525,
      "step": 3109,
      "time_per_iteration": 2.7164478302001953
    },
    {
      "auxiliary_loss_clip": 0.01196535,
      "auxiliary_loss_mlp": 0.01032941,
      "balance_loss_clip": 1.06054342,
      "balance_loss_mlp": 1.02307081,
      "epoch": 0.3739553898875729,
      "flos": 43507967448960.0,
      "grad_norm": 3.1414824833874313,
      "language_loss": 0.68901378,
      "learning_rate": 2.882276969236016e-06,
      "loss": 0.71130854,
      "num_input_tokens_seen": 66816885,
      "step": 3110,
      "time_per_iteration": 2.7964890003204346
    },
    {
      "auxiliary_loss_clip": 0.01195151,
      "auxiliary_loss_mlp": 0.01035569,
      "balance_loss_clip": 1.06022739,
      "balance_loss_mlp": 1.02607441,
      "epoch": 0.374075632778212,
      "flos": 12856487448960.0,
      "grad_norm": 2.0965678326747406,
      "language_loss": 0.76638174,
      "learning_rate": 2.881577822763005e-06,
      "loss": 0.78868902,
      "num_input_tokens_seen": 66834835,
      "step": 3111,
      "time_per_iteration": 2.5818283557891846
    },
    {
      "auxiliary_loss_clip": 0.01191128,
      "auxiliary_loss_mlp": 0.01031539,
      "balance_loss_clip": 1.05593836,
      "balance_loss_mlp": 1.02273512,
      "epoch": 0.3741958756688511,
      "flos": 26024031699840.0,
      "grad_norm": 2.1143315194803285,
      "language_loss": 0.87231302,
      "learning_rate": 2.880878542555338e-06,
      "loss": 0.89453971,
      "num_input_tokens_seen": 66852600,
      "step": 3112,
      "time_per_iteration": 2.668459177017212
    },
    {
      "auxiliary_loss_clip": 0.01193047,
      "auxiliary_loss_mlp": 0.01030411,
      "balance_loss_clip": 1.05762196,
      "balance_loss_mlp": 1.02172709,
      "epoch": 0.37431611855949015,
      "flos": 21433894652160.0,
      "grad_norm": 2.2401052736591467,
      "language_loss": 0.80812365,
      "learning_rate": 2.8801791287190976e-06,
      "loss": 0.83035827,
      "num_input_tokens_seen": 66870595,
      "step": 3113,
      "time_per_iteration": 2.6086814403533936
    },
    {
      "auxiliary_loss_clip": 0.01195448,
      "auxiliary_loss_mlp": 0.01028818,
      "balance_loss_clip": 1.05899119,
      "balance_loss_mlp": 1.01980567,
      "epoch": 0.37443636145012926,
      "flos": 24207096090240.0,
      "grad_norm": 2.8611078479702705,
      "language_loss": 0.85482311,
      "learning_rate": 2.8794795813603817e-06,
      "loss": 0.87706578,
      "num_input_tokens_seen": 66886060,
      "step": 3114,
      "time_per_iteration": 2.6339097023010254
    },
    {
      "auxiliary_loss_clip": 0.01195339,
      "auxiliary_loss_mlp": 0.01028347,
      "balance_loss_clip": 1.05926728,
      "balance_loss_mlp": 1.01871443,
      "epoch": 0.3745566043407684,
      "flos": 15378601841280.0,
      "grad_norm": 1.796119414312257,
      "language_loss": 0.81507176,
      "learning_rate": 2.878779900585314e-06,
      "loss": 0.83730865,
      "num_input_tokens_seen": 66903900,
      "step": 3115,
      "time_per_iteration": 2.691524028778076
    },
    {
      "auxiliary_loss_clip": 0.01192405,
      "auxiliary_loss_mlp": 0.0103593,
      "balance_loss_clip": 1.05802381,
      "balance_loss_mlp": 1.02766263,
      "epoch": 0.37467684723140743,
      "flos": 24608218245120.0,
      "grad_norm": 1.795187137216496,
      "language_loss": 0.75450122,
      "learning_rate": 2.8780800865000336e-06,
      "loss": 0.77678466,
      "num_input_tokens_seen": 66925210,
      "step": 3116,
      "time_per_iteration": 2.621598958969116
    },
    {
      "auxiliary_loss_clip": 0.01089547,
      "auxiliary_loss_mlp": 0.01002603,
      "balance_loss_clip": 1.02065194,
      "balance_loss_mlp": 1.00110674,
      "epoch": 0.37479709012204654,
      "flos": 64377491610240.0,
      "grad_norm": 0.9831934146523207,
      "language_loss": 0.59134477,
      "learning_rate": 2.877380139210702e-06,
      "loss": 0.6122663,
      "num_input_tokens_seen": 66983880,
      "step": 3117,
      "time_per_iteration": 3.1818652153015137
    },
    {
      "auxiliary_loss_clip": 0.01197589,
      "auxiliary_loss_mlp": 0.01028484,
      "balance_loss_clip": 1.0596602,
      "balance_loss_mlp": 1.0189054,
      "epoch": 0.37491733301268565,
      "flos": 23803962773760.0,
      "grad_norm": 1.8572372945383742,
      "language_loss": 0.76583612,
      "learning_rate": 2.876680058823501e-06,
      "loss": 0.78809685,
      "num_input_tokens_seen": 67004280,
      "step": 3118,
      "time_per_iteration": 2.648886203765869
    },
    {
      "auxiliary_loss_clip": 0.01194466,
      "auxiliary_loss_mlp": 0.01027431,
      "balance_loss_clip": 1.05743074,
      "balance_loss_mlp": 1.01797175,
      "epoch": 0.3750375759033247,
      "flos": 32160950167680.0,
      "grad_norm": 2.2826937386225645,
      "language_loss": 0.65949488,
      "learning_rate": 2.8759798454446314e-06,
      "loss": 0.68171388,
      "num_input_tokens_seen": 67027445,
      "step": 3119,
      "time_per_iteration": 2.726520538330078
    },
    {
      "auxiliary_loss_clip": 0.01193387,
      "auxiliary_loss_mlp": 0.0103121,
      "balance_loss_clip": 1.05964112,
      "balance_loss_mlp": 1.02261496,
      "epoch": 0.3751578187939638,
      "flos": 23367791923200.0,
      "grad_norm": 1.8426963581048041,
      "language_loss": 0.81516343,
      "learning_rate": 2.8752794991803173e-06,
      "loss": 0.83740944,
      "num_input_tokens_seen": 67045130,
      "step": 3120,
      "time_per_iteration": 2.653709650039673
    },
    {
      "auxiliary_loss_clip": 0.01195188,
      "auxiliary_loss_mlp": 0.01032374,
      "balance_loss_clip": 1.06224489,
      "balance_loss_mlp": 1.02393425,
      "epoch": 0.37527806168460287,
      "flos": 14605731878400.0,
      "grad_norm": 2.017006628969718,
      "language_loss": 0.75218689,
      "learning_rate": 2.8745790201367976e-06,
      "loss": 0.77446258,
      "num_input_tokens_seen": 67060885,
      "step": 3121,
      "time_per_iteration": 3.5891056060791016
    },
    {
      "auxiliary_loss_clip": 0.01195582,
      "auxiliary_loss_mlp": 0.01035952,
      "balance_loss_clip": 1.05867505,
      "balance_loss_mlp": 1.02683854,
      "epoch": 0.375398304575242,
      "flos": 26390823431040.0,
      "grad_norm": 1.8882031192747901,
      "language_loss": 0.84580117,
      "learning_rate": 2.8738784084203373e-06,
      "loss": 0.8681165,
      "num_input_tokens_seen": 67080960,
      "step": 3122,
      "time_per_iteration": 2.7030029296875
    },
    {
      "auxiliary_loss_clip": 0.01190524,
      "auxiliary_loss_mlp": 0.01026742,
      "balance_loss_clip": 1.05756247,
      "balance_loss_mlp": 1.0177958,
      "epoch": 0.3755185474658811,
      "flos": 22236605838720.0,
      "grad_norm": 1.6599225053387676,
      "language_loss": 0.78533673,
      "learning_rate": 2.873177664137216e-06,
      "loss": 0.80750936,
      "num_input_tokens_seen": 67101890,
      "step": 3123,
      "time_per_iteration": 4.42920446395874
    },
    {
      "auxiliary_loss_clip": 0.01193844,
      "auxiliary_loss_mlp": 0.01029119,
      "balance_loss_clip": 1.05859518,
      "balance_loss_mlp": 1.01999402,
      "epoch": 0.37563879035652015,
      "flos": 30812935633920.0,
      "grad_norm": 1.9822462087194566,
      "language_loss": 0.69302821,
      "learning_rate": 2.8724767873937384e-06,
      "loss": 0.71525782,
      "num_input_tokens_seen": 67126010,
      "step": 3124,
      "time_per_iteration": 2.709127902984619
    },
    {
      "auxiliary_loss_clip": 0.01191606,
      "auxiliary_loss_mlp": 0.01032804,
      "balance_loss_clip": 1.05771768,
      "balance_loss_mlp": 1.02408361,
      "epoch": 0.37575903324715926,
      "flos": 20773533064320.0,
      "grad_norm": 1.991030209721961,
      "language_loss": 0.87619913,
      "learning_rate": 2.871775778296225e-06,
      "loss": 0.89844328,
      "num_input_tokens_seen": 67143100,
      "step": 3125,
      "time_per_iteration": 2.6041979789733887
    },
    {
      "auxiliary_loss_clip": 0.01197414,
      "auxiliary_loss_mlp": 0.01027506,
      "balance_loss_clip": 1.06021202,
      "balance_loss_mlp": 1.01782036,
      "epoch": 0.37587927613779837,
      "flos": 18697681244160.0,
      "grad_norm": 2.789187713651255,
      "language_loss": 0.78141892,
      "learning_rate": 2.8710746369510196e-06,
      "loss": 0.80366808,
      "num_input_tokens_seen": 67161085,
      "step": 3126,
      "time_per_iteration": 2.6125869750976562
    },
    {
      "auxiliary_loss_clip": 0.01196658,
      "auxiliary_loss_mlp": 0.01036262,
      "balance_loss_clip": 1.06041479,
      "balance_loss_mlp": 1.0265646,
      "epoch": 0.3759995190284374,
      "flos": 13624796384640.0,
      "grad_norm": 2.4678796674616126,
      "language_loss": 0.83468723,
      "learning_rate": 2.8703733634644846e-06,
      "loss": 0.85701644,
      "num_input_tokens_seen": 67175840,
      "step": 3127,
      "time_per_iteration": 2.581810474395752
    },
    {
      "auxiliary_loss_clip": 0.01192489,
      "auxiliary_loss_mlp": 0.01028271,
      "balance_loss_clip": 1.05942929,
      "balance_loss_mlp": 1.02004528,
      "epoch": 0.37611976191907653,
      "flos": 20484847457280.0,
      "grad_norm": 1.6615223857108696,
      "language_loss": 0.7895782,
      "learning_rate": 2.869671957943002e-06,
      "loss": 0.81178582,
      "num_input_tokens_seen": 67194995,
      "step": 3128,
      "time_per_iteration": 2.6033637523651123
    },
    {
      "auxiliary_loss_clip": 0.01194097,
      "auxiliary_loss_mlp": 0.01033023,
      "balance_loss_clip": 1.06081831,
      "balance_loss_mlp": 1.02470827,
      "epoch": 0.37624000480971564,
      "flos": 21141797253120.0,
      "grad_norm": 1.7782092038541906,
      "language_loss": 0.74596256,
      "learning_rate": 2.8689704204929747e-06,
      "loss": 0.76823378,
      "num_input_tokens_seen": 67214175,
      "step": 3129,
      "time_per_iteration": 2.622840166091919
    },
    {
      "auxiliary_loss_clip": 0.01193091,
      "auxiliary_loss_mlp": 0.01026857,
      "balance_loss_clip": 1.05862641,
      "balance_loss_mlp": 1.01823258,
      "epoch": 0.3763602477003547,
      "flos": 22564470205440.0,
      "grad_norm": 1.8624918294605317,
      "language_loss": 0.81076908,
      "learning_rate": 2.8682687512208253e-06,
      "loss": 0.83296847,
      "num_input_tokens_seen": 67233185,
      "step": 3130,
      "time_per_iteration": 2.592019557952881
    },
    {
      "auxiliary_loss_clip": 0.01195902,
      "auxiliary_loss_mlp": 0.01033872,
      "balance_loss_clip": 1.05879426,
      "balance_loss_mlp": 1.02421045,
      "epoch": 0.3764804905909938,
      "flos": 27526857851520.0,
      "grad_norm": 1.7879792306951863,
      "language_loss": 0.80494267,
      "learning_rate": 2.8675669502329972e-06,
      "loss": 0.82724041,
      "num_input_tokens_seen": 67254715,
      "step": 3131,
      "time_per_iteration": 2.6624631881713867
    },
    {
      "auxiliary_loss_clip": 0.01193053,
      "auxiliary_loss_mlp": 0.01031324,
      "balance_loss_clip": 1.05797768,
      "balance_loss_mlp": 1.02181745,
      "epoch": 0.3766007334816329,
      "flos": 22528092706560.0,
      "grad_norm": 2.6007553789150326,
      "language_loss": 0.85642201,
      "learning_rate": 2.866865017635952e-06,
      "loss": 0.8786658,
      "num_input_tokens_seen": 67272535,
      "step": 3132,
      "time_per_iteration": 2.5797111988067627
    },
    {
      "auxiliary_loss_clip": 0.01193907,
      "auxiliary_loss_mlp": 0.01029802,
      "balance_loss_clip": 1.05912423,
      "balance_loss_mlp": 1.02085507,
      "epoch": 0.376720976372272,
      "flos": 25957166532480.0,
      "grad_norm": 1.7100969381480338,
      "language_loss": 0.79400009,
      "learning_rate": 2.866162953536174e-06,
      "loss": 0.81623721,
      "num_input_tokens_seen": 67293505,
      "step": 3133,
      "time_per_iteration": 2.6404340267181396
    },
    {
      "auxiliary_loss_clip": 0.01191564,
      "auxiliary_loss_mlp": 0.01029361,
      "balance_loss_clip": 1.05870628,
      "balance_loss_mlp": 1.0209682,
      "epoch": 0.3768412192629111,
      "flos": 18041162411520.0,
      "grad_norm": 1.5493453857202635,
      "language_loss": 0.75026906,
      "learning_rate": 2.8654607580401634e-06,
      "loss": 0.77247834,
      "num_input_tokens_seen": 67313240,
      "step": 3134,
      "time_per_iteration": 2.593663454055786
    },
    {
      "auxiliary_loss_clip": 0.01089887,
      "auxiliary_loss_mlp": 0.01000861,
      "balance_loss_clip": 1.02125454,
      "balance_loss_mlp": 0.99931109,
      "epoch": 0.3769614621535502,
      "flos": 62989472304000.0,
      "grad_norm": 0.8834403405520561,
      "language_loss": 0.65232015,
      "learning_rate": 2.8647584312544446e-06,
      "loss": 0.67322767,
      "num_input_tokens_seen": 67378445,
      "step": 3135,
      "time_per_iteration": 3.253873348236084
    },
    {
      "auxiliary_loss_clip": 0.01193167,
      "auxiliary_loss_mlp": 0.01034482,
      "balance_loss_clip": 1.05954921,
      "balance_loss_mlp": 1.02579188,
      "epoch": 0.37708170504418925,
      "flos": 23661685002240.0,
      "grad_norm": 1.454692265804009,
      "language_loss": 0.85200542,
      "learning_rate": 2.864055973285559e-06,
      "loss": 0.87428188,
      "num_input_tokens_seen": 67400445,
      "step": 3136,
      "time_per_iteration": 2.6462900638580322
    },
    {
      "auxiliary_loss_clip": 0.01189529,
      "auxiliary_loss_mlp": 0.01028322,
      "balance_loss_clip": 1.05839097,
      "balance_loss_mlp": 1.02019823,
      "epoch": 0.37720194793482836,
      "flos": 24423170353920.0,
      "grad_norm": 1.8602344767209165,
      "language_loss": 0.86202586,
      "learning_rate": 2.8633533842400698e-06,
      "loss": 0.88420439,
      "num_input_tokens_seen": 67420645,
      "step": 3137,
      "time_per_iteration": 2.6643550395965576
    },
    {
      "auxiliary_loss_clip": 0.01201121,
      "auxiliary_loss_mlp": 0.01031421,
      "balance_loss_clip": 1.06373692,
      "balance_loss_mlp": 1.02066886,
      "epoch": 0.3773221908254674,
      "flos": 20996502739200.0,
      "grad_norm": 1.7477093669765815,
      "language_loss": 0.77277339,
      "learning_rate": 2.862650664224558e-06,
      "loss": 0.79509878,
      "num_input_tokens_seen": 67439495,
      "step": 3138,
      "time_per_iteration": 2.608403205871582
    },
    {
      "auxiliary_loss_clip": 0.01192071,
      "auxiliary_loss_mlp": 0.0102879,
      "balance_loss_clip": 1.05896187,
      "balance_loss_mlp": 1.02036142,
      "epoch": 0.37744243371610653,
      "flos": 37631724958080.0,
      "grad_norm": 1.3722419589975907,
      "language_loss": 0.6966908,
      "learning_rate": 2.861947813345627e-06,
      "loss": 0.71889943,
      "num_input_tokens_seen": 67462195,
      "step": 3139,
      "time_per_iteration": 2.771491765975952
    },
    {
      "auxiliary_loss_clip": 0.01194161,
      "auxiliary_loss_mlp": 0.01032135,
      "balance_loss_clip": 1.05797911,
      "balance_loss_mlp": 1.02257466,
      "epoch": 0.37756267660674564,
      "flos": 26140526484480.0,
      "grad_norm": 1.756785054027371,
      "language_loss": 0.71916103,
      "learning_rate": 2.8612448317098974e-06,
      "loss": 0.74142408,
      "num_input_tokens_seen": 67482530,
      "step": 3140,
      "time_per_iteration": 2.656214952468872
    },
    {
      "auxiliary_loss_clip": 0.01194528,
      "auxiliary_loss_mlp": 0.01031716,
      "balance_loss_clip": 1.05819988,
      "balance_loss_mlp": 1.02221537,
      "epoch": 0.3776829194973847,
      "flos": 19427888828160.0,
      "grad_norm": 2.120465946953091,
      "language_loss": 0.82987195,
      "learning_rate": 2.8605417194240114e-06,
      "loss": 0.85213441,
      "num_input_tokens_seen": 67500890,
      "step": 3141,
      "time_per_iteration": 2.5913329124450684
    },
    {
      "auxiliary_loss_clip": 0.01187512,
      "auxiliary_loss_mlp": 0.01031643,
      "balance_loss_clip": 1.054901,
      "balance_loss_mlp": 1.0230422,
      "epoch": 0.3778031623880238,
      "flos": 17382309194880.0,
      "grad_norm": 1.9749071958005306,
      "language_loss": 0.78892201,
      "learning_rate": 2.8598384765946315e-06,
      "loss": 0.81111354,
      "num_input_tokens_seen": 67519545,
      "step": 3142,
      "time_per_iteration": 2.583634853363037
    },
    {
      "auxiliary_loss_clip": 0.01189398,
      "auxiliary_loss_mlp": 0.01029539,
      "balance_loss_clip": 1.05482161,
      "balance_loss_mlp": 1.02060401,
      "epoch": 0.3779234052786629,
      "flos": 27125843437440.0,
      "grad_norm": 1.8979249993469327,
      "language_loss": 0.71716344,
      "learning_rate": 2.8591351033284377e-06,
      "loss": 0.73935282,
      "num_input_tokens_seen": 67539275,
      "step": 3143,
      "time_per_iteration": 2.614328384399414
    },
    {
      "auxiliary_loss_clip": 0.01188326,
      "auxiliary_loss_mlp": 0.01026823,
      "balance_loss_clip": 1.05476856,
      "balance_loss_mlp": 1.0180732,
      "epoch": 0.37804364816930197,
      "flos": 19682639061120.0,
      "grad_norm": 2.5932007447820014,
      "language_loss": 0.84065187,
      "learning_rate": 2.8584315997321325e-06,
      "loss": 0.86280334,
      "num_input_tokens_seen": 67558280,
      "step": 3144,
      "time_per_iteration": 2.620128631591797
    },
    {
      "auxiliary_loss_clip": 0.01191114,
      "auxiliary_loss_mlp": 0.01030668,
      "balance_loss_clip": 1.05480123,
      "balance_loss_mlp": 1.02173376,
      "epoch": 0.3781638910599411,
      "flos": 22702905221760.0,
      "grad_norm": 2.5323791262312083,
      "language_loss": 0.78307915,
      "learning_rate": 2.8577279659124356e-06,
      "loss": 0.80529696,
      "num_input_tokens_seen": 67575955,
      "step": 3145,
      "time_per_iteration": 2.5604782104492188
    },
    {
      "auxiliary_loss_clip": 0.011884,
      "auxiliary_loss_mlp": 0.0102566,
      "balance_loss_clip": 1.05794597,
      "balance_loss_mlp": 1.01725554,
      "epoch": 0.3782841339505802,
      "flos": 14647604158080.0,
      "grad_norm": 2.1327548497194035,
      "language_loss": 0.83775902,
      "learning_rate": 2.857024201976089e-06,
      "loss": 0.85989964,
      "num_input_tokens_seen": 67593515,
      "step": 3146,
      "time_per_iteration": 2.6322827339172363
    },
    {
      "auxiliary_loss_clip": 0.01191827,
      "auxiliary_loss_mlp": 0.01029604,
      "balance_loss_clip": 1.05634427,
      "balance_loss_mlp": 1.01931,
      "epoch": 0.37840437684121925,
      "flos": 32818223185920.0,
      "grad_norm": 1.9401311684422444,
      "language_loss": 0.72676122,
      "learning_rate": 2.8563203080298516e-06,
      "loss": 0.74897552,
      "num_input_tokens_seen": 67614290,
      "step": 3147,
      "time_per_iteration": 3.5355148315429688
    },
    {
      "auxiliary_loss_clip": 0.01193933,
      "auxiliary_loss_mlp": 0.01031112,
      "balance_loss_clip": 1.05772424,
      "balance_loss_mlp": 1.02195632,
      "epoch": 0.37852461973185836,
      "flos": 18369206346240.0,
      "grad_norm": 2.3850255622122503,
      "language_loss": 0.89673567,
      "learning_rate": 2.855616284180505e-06,
      "loss": 0.91898608,
      "num_input_tokens_seen": 67631340,
      "step": 3148,
      "time_per_iteration": 2.5918071269989014
    },
    {
      "auxiliary_loss_clip": 0.01088306,
      "auxiliary_loss_mlp": 0.010041,
      "balance_loss_clip": 1.01991653,
      "balance_loss_mlp": 1.00252628,
      "epoch": 0.37864486262249747,
      "flos": 59500680117120.0,
      "grad_norm": 0.8822917232739874,
      "language_loss": 0.66185141,
      "learning_rate": 2.8549121305348477e-06,
      "loss": 0.68277538,
      "num_input_tokens_seen": 67691125,
      "step": 3149,
      "time_per_iteration": 4.992602348327637
    },
    {
      "auxiliary_loss_clip": 0.01185305,
      "auxiliary_loss_mlp": 0.01028849,
      "balance_loss_clip": 1.05560958,
      "balance_loss_mlp": 1.02063537,
      "epoch": 0.3787651055131365,
      "flos": 23363015414400.0,
      "grad_norm": 2.3649837829242615,
      "language_loss": 0.83129609,
      "learning_rate": 2.8542078471997006e-06,
      "loss": 0.85343754,
      "num_input_tokens_seen": 67708740,
      "step": 3150,
      "time_per_iteration": 3.5239598751068115
    },
    {
      "auxiliary_loss_clip": 0.0119082,
      "auxiliary_loss_mlp": 0.01029167,
      "balance_loss_clip": 1.05669498,
      "balance_loss_mlp": 1.02068496,
      "epoch": 0.37888534840377563,
      "flos": 24601394661120.0,
      "grad_norm": 1.9089401248143922,
      "language_loss": 0.75638014,
      "learning_rate": 2.8535034342819013e-06,
      "loss": 0.77857995,
      "num_input_tokens_seen": 67726150,
      "step": 3151,
      "time_per_iteration": 2.6178500652313232
    },
    {
      "auxiliary_loss_clip": 0.01186398,
      "auxiliary_loss_mlp": 0.01028914,
      "balance_loss_clip": 1.05462837,
      "balance_loss_mlp": 1.02074862,
      "epoch": 0.37900559129441475,
      "flos": 23986891762560.0,
      "grad_norm": 1.5957090763560278,
      "language_loss": 0.72729462,
      "learning_rate": 2.85279889188831e-06,
      "loss": 0.74944776,
      "num_input_tokens_seen": 67746525,
      "step": 3152,
      "time_per_iteration": 2.6466894149780273
    },
    {
      "auxiliary_loss_clip": 0.01190677,
      "auxiliary_loss_mlp": 0.01031488,
      "balance_loss_clip": 1.05682755,
      "balance_loss_mlp": 1.02220154,
      "epoch": 0.3791258341850538,
      "flos": 24644667571200.0,
      "grad_norm": 2.4605128183264693,
      "language_loss": 0.81011927,
      "learning_rate": 2.852094220125805e-06,
      "loss": 0.83234096,
      "num_input_tokens_seen": 67766035,
      "step": 3153,
      "time_per_iteration": 2.634528398513794
    },
    {
      "auxiliary_loss_clip": 0.01196242,
      "auxiliary_loss_mlp": 0.01034589,
      "balance_loss_clip": 1.06102371,
      "balance_loss_mlp": 1.02477813,
      "epoch": 0.3792460770756929,
      "flos": 17420841509760.0,
      "grad_norm": 2.2314228146372517,
      "language_loss": 0.71405578,
      "learning_rate": 2.8513894191012846e-06,
      "loss": 0.73636401,
      "num_input_tokens_seen": 67785015,
      "step": 3154,
      "time_per_iteration": 2.648347854614258
    },
    {
      "auxiliary_loss_clip": 0.01189588,
      "auxiliary_loss_mlp": 0.01027987,
      "balance_loss_clip": 1.05597425,
      "balance_loss_mlp": 1.01941586,
      "epoch": 0.37936631996633197,
      "flos": 24206557386240.0,
      "grad_norm": 1.6153412300644043,
      "language_loss": 0.78837317,
      "learning_rate": 2.8506844889216664e-06,
      "loss": 0.8105489,
      "num_input_tokens_seen": 67804400,
      "step": 3155,
      "time_per_iteration": 2.6111741065979004
    },
    {
      "auxiliary_loss_clip": 0.01087756,
      "auxiliary_loss_mlp": 0.01003122,
      "balance_loss_clip": 1.01967835,
      "balance_loss_mlp": 1.00156045,
      "epoch": 0.3794865628569711,
      "flos": 70297114752000.0,
      "grad_norm": 0.8599518075991955,
      "language_loss": 0.62865436,
      "learning_rate": 2.849979429693887e-06,
      "loss": 0.64956319,
      "num_input_tokens_seen": 67865385,
      "step": 3156,
      "time_per_iteration": 3.205920457839966
    },
    {
      "auxiliary_loss_clip": 0.01192218,
      "auxiliary_loss_mlp": 0.01024401,
      "balance_loss_clip": 1.057868,
      "balance_loss_mlp": 1.01566279,
      "epoch": 0.3796068057476102,
      "flos": 15779364860160.0,
      "grad_norm": 2.0976265928032842,
      "language_loss": 0.74218965,
      "learning_rate": 2.8492742415249042e-06,
      "loss": 0.76435578,
      "num_input_tokens_seen": 67883030,
      "step": 3157,
      "time_per_iteration": 2.4543187618255615
    },
    {
      "auxiliary_loss_clip": 0.01189082,
      "auxiliary_loss_mlp": 0.01029513,
      "balance_loss_clip": 1.05627561,
      "balance_loss_mlp": 1.02059054,
      "epoch": 0.37972704863824924,
      "flos": 25191694771200.0,
      "grad_norm": 1.6447867548326818,
      "language_loss": 0.76250964,
      "learning_rate": 2.848568924521694e-06,
      "loss": 0.78469563,
      "num_input_tokens_seen": 67903810,
      "step": 3158,
      "time_per_iteration": 2.4994564056396484
    },
    {
      "auxiliary_loss_clip": 0.01190753,
      "auxiliary_loss_mlp": 0.01029689,
      "balance_loss_clip": 1.05686164,
      "balance_loss_mlp": 1.02030778,
      "epoch": 0.37984729152888835,
      "flos": 26210372480640.0,
      "grad_norm": 3.161229140512964,
      "language_loss": 0.73661542,
      "learning_rate": 2.8478634787912526e-06,
      "loss": 0.75881982,
      "num_input_tokens_seen": 67921865,
      "step": 3159,
      "time_per_iteration": 2.498886823654175
    },
    {
      "auxiliary_loss_clip": 0.0119005,
      "auxiliary_loss_mlp": 0.01033413,
      "balance_loss_clip": 1.05756354,
      "balance_loss_mlp": 1.02505088,
      "epoch": 0.37996753441952746,
      "flos": 25629302165760.0,
      "grad_norm": 2.894073025386233,
      "language_loss": 0.76589626,
      "learning_rate": 2.847157904440596e-06,
      "loss": 0.78813088,
      "num_input_tokens_seen": 67941595,
      "step": 3160,
      "time_per_iteration": 2.6175901889801025
    },
    {
      "auxiliary_loss_clip": 0.01189367,
      "auxiliary_loss_mlp": 0.01027876,
      "balance_loss_clip": 1.05569148,
      "balance_loss_mlp": 1.01931667,
      "epoch": 0.3800877773101665,
      "flos": 20118414862080.0,
      "grad_norm": 3.185743832965742,
      "language_loss": 0.73956823,
      "learning_rate": 2.846452201576759e-06,
      "loss": 0.76174068,
      "num_input_tokens_seen": 67960970,
      "step": 3161,
      "time_per_iteration": 2.5294506549835205
    },
    {
      "auxiliary_loss_clip": 0.01088318,
      "auxiliary_loss_mlp": 0.01001257,
      "balance_loss_clip": 1.02013254,
      "balance_loss_mlp": 0.99974883,
      "epoch": 0.38020802020080563,
      "flos": 63053608037760.0,
      "grad_norm": 1.362025351860967,
      "language_loss": 0.62740451,
      "learning_rate": 2.845746370306795e-06,
      "loss": 0.64830029,
      "num_input_tokens_seen": 68026160,
      "step": 3162,
      "time_per_iteration": 3.232226848602295
    },
    {
      "auxiliary_loss_clip": 0.01191776,
      "auxiliary_loss_mlp": 0.01030625,
      "balance_loss_clip": 1.05874741,
      "balance_loss_mlp": 1.02203619,
      "epoch": 0.38032826309144474,
      "flos": 21288420570240.0,
      "grad_norm": 1.889333646113306,
      "language_loss": 0.78686923,
      "learning_rate": 2.84504041073778e-06,
      "loss": 0.80909324,
      "num_input_tokens_seen": 68044575,
      "step": 3163,
      "time_per_iteration": 2.605268716812134
    },
    {
      "auxiliary_loss_clip": 0.01190421,
      "auxiliary_loss_mlp": 0.01034676,
      "balance_loss_clip": 1.05814505,
      "balance_loss_mlp": 1.02556837,
      "epoch": 0.3804485059820838,
      "flos": 18954119416320.0,
      "grad_norm": 1.7851172948575966,
      "language_loss": 0.79433978,
      "learning_rate": 2.844334322976806e-06,
      "loss": 0.81659079,
      "num_input_tokens_seen": 68064790,
      "step": 3164,
      "time_per_iteration": 2.5741889476776123
    },
    {
      "auxiliary_loss_clip": 0.01195396,
      "auxiliary_loss_mlp": 0.01027772,
      "balance_loss_clip": 1.0583483,
      "balance_loss_mlp": 1.01824093,
      "epoch": 0.3805687488727229,
      "flos": 21833759831040.0,
      "grad_norm": 2.0138500924628637,
      "language_loss": 0.83265722,
      "learning_rate": 2.8436281071309866e-06,
      "loss": 0.85488892,
      "num_input_tokens_seen": 68083330,
      "step": 3165,
      "time_per_iteration": 2.616179943084717
    },
    {
      "auxiliary_loss_clip": 0.01087735,
      "auxiliary_loss_mlp": 0.0100234,
      "balance_loss_clip": 1.01983774,
      "balance_loss_mlp": 1.0007304,
      "epoch": 0.380688991763362,
      "flos": 58546209968640.0,
      "grad_norm": 0.7252249343347286,
      "language_loss": 0.52967387,
      "learning_rate": 2.842921763307455e-06,
      "loss": 0.55057466,
      "num_input_tokens_seen": 68146140,
      "step": 3166,
      "time_per_iteration": 3.1091511249542236
    },
    {
      "auxiliary_loss_clip": 0.01192111,
      "auxiliary_loss_mlp": 0.01025053,
      "balance_loss_clip": 1.05876708,
      "balance_loss_mlp": 1.01614237,
      "epoch": 0.38080923465400107,
      "flos": 23799509487360.0,
      "grad_norm": 1.9635326519841358,
      "language_loss": 0.82591403,
      "learning_rate": 2.842215291613361e-06,
      "loss": 0.84808564,
      "num_input_tokens_seen": 68164520,
      "step": 3167,
      "time_per_iteration": 2.58469820022583
    },
    {
      "auxiliary_loss_clip": 0.01088173,
      "auxiliary_loss_mlp": 0.01001653,
      "balance_loss_clip": 1.02017343,
      "balance_loss_mlp": 1.00006759,
      "epoch": 0.3809294775446402,
      "flos": 54969866380800.0,
      "grad_norm": 0.7959097034758694,
      "language_loss": 0.59231484,
      "learning_rate": 2.8415086921558774e-06,
      "loss": 0.61321306,
      "num_input_tokens_seen": 68227945,
      "step": 3168,
      "time_per_iteration": 3.1434504985809326
    },
    {
      "auxiliary_loss_clip": 0.01186489,
      "auxiliary_loss_mlp": 0.01025851,
      "balance_loss_clip": 1.05366492,
      "balance_loss_mlp": 1.01709473,
      "epoch": 0.38104972043527924,
      "flos": 24643697904000.0,
      "grad_norm": 1.7298365002955844,
      "language_loss": 0.78544748,
      "learning_rate": 2.840801965042194e-06,
      "loss": 0.80757082,
      "num_input_tokens_seen": 68247405,
      "step": 3169,
      "time_per_iteration": 2.6057207584381104
    },
    {
      "auxiliary_loss_clip": 0.01188985,
      "auxiliary_loss_mlp": 0.01026954,
      "balance_loss_clip": 1.05651927,
      "balance_loss_mlp": 1.01861477,
      "epoch": 0.38116996332591835,
      "flos": 22856783086080.0,
      "grad_norm": 4.010743418070282,
      "language_loss": 0.8407473,
      "learning_rate": 2.840095110379521e-06,
      "loss": 0.86290669,
      "num_input_tokens_seen": 68266925,
      "step": 3170,
      "time_per_iteration": 2.5409860610961914
    },
    {
      "auxiliary_loss_clip": 0.0108806,
      "auxiliary_loss_mlp": 0.01001022,
      "balance_loss_clip": 1.01979971,
      "balance_loss_mlp": 0.99950773,
      "epoch": 0.38129020621655746,
      "flos": 60836160804480.0,
      "grad_norm": 0.7160551897418141,
      "language_loss": 0.5389356,
      "learning_rate": 2.8393881282750884e-06,
      "loss": 0.55982643,
      "num_input_tokens_seen": 68329755,
      "step": 3171,
      "time_per_iteration": 3.173096179962158
    },
    {
      "auxiliary_loss_clip": 0.01188865,
      "auxiliary_loss_mlp": 0.01032366,
      "balance_loss_clip": 1.05532169,
      "balance_loss_mlp": 1.02323461,
      "epoch": 0.3814104491071965,
      "flos": 21648101408640.0,
      "grad_norm": 2.046192138777426,
      "language_loss": 0.78874004,
      "learning_rate": 2.838681018836144e-06,
      "loss": 0.81095237,
      "num_input_tokens_seen": 68347075,
      "step": 3172,
      "time_per_iteration": 2.5929617881774902
    },
    {
      "auxiliary_loss_clip": 0.01188914,
      "auxiliary_loss_mlp": 0.01027288,
      "balance_loss_clip": 1.05592501,
      "balance_loss_mlp": 1.01885128,
      "epoch": 0.3815306919978356,
      "flos": 19099090707840.0,
      "grad_norm": 2.268055260369511,
      "language_loss": 0.78602767,
      "learning_rate": 2.837973782169955e-06,
      "loss": 0.80818963,
      "num_input_tokens_seen": 68365450,
      "step": 3173,
      "time_per_iteration": 2.621211051940918
    },
    {
      "auxiliary_loss_clip": 0.01086515,
      "auxiliary_loss_mlp": 0.01001563,
      "balance_loss_clip": 1.01853561,
      "balance_loss_mlp": 1.00005484,
      "epoch": 0.38165093488847474,
      "flos": 67067918156160.0,
      "grad_norm": 0.8204886677366203,
      "language_loss": 0.59226817,
      "learning_rate": 2.8372664183838096e-06,
      "loss": 0.61314893,
      "num_input_tokens_seen": 68428470,
      "step": 3174,
      "time_per_iteration": 4.1043455600738525
    },
    {
      "auxiliary_loss_clip": 0.01190167,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.05739677,
      "balance_loss_mlp": 1.02123463,
      "epoch": 0.3817711777791138,
      "flos": 22341105480960.0,
      "grad_norm": 2.5263765892240166,
      "language_loss": 0.68341672,
      "learning_rate": 2.836558927585015e-06,
      "loss": 0.70561898,
      "num_input_tokens_seen": 68445440,
      "step": 3175,
      "time_per_iteration": 4.386636018753052
    },
    {
      "auxiliary_loss_clip": 0.01190443,
      "auxiliary_loss_mlp": 0.01027286,
      "balance_loss_clip": 1.05531538,
      "balance_loss_mlp": 1.018888,
      "epoch": 0.3818914206697529,
      "flos": 22820621068800.0,
      "grad_norm": 1.8475351376748386,
      "language_loss": 0.82083178,
      "learning_rate": 2.8358513098808957e-06,
      "loss": 0.84300911,
      "num_input_tokens_seen": 68465755,
      "step": 3176,
      "time_per_iteration": 2.624833106994629
    },
    {
      "auxiliary_loss_clip": 0.01190466,
      "auxiliary_loss_mlp": 0.01030649,
      "balance_loss_clip": 1.05614614,
      "balance_loss_mlp": 1.02160692,
      "epoch": 0.382011663560392,
      "flos": 24386074583040.0,
      "grad_norm": 1.7340115260193665,
      "language_loss": 0.76910567,
      "learning_rate": 2.835143565378798e-06,
      "loss": 0.79131681,
      "num_input_tokens_seen": 68486220,
      "step": 3177,
      "time_per_iteration": 2.721579074859619
    },
    {
      "auxiliary_loss_clip": 0.01191561,
      "auxiliary_loss_mlp": 0.0103073,
      "balance_loss_clip": 1.05802417,
      "balance_loss_mlp": 1.02118731,
      "epoch": 0.38213190645103107,
      "flos": 21981568296960.0,
      "grad_norm": 2.2386805864589143,
      "language_loss": 0.78393853,
      "learning_rate": 2.8344356941860847e-06,
      "loss": 0.8061614,
      "num_input_tokens_seen": 68505850,
      "step": 3178,
      "time_per_iteration": 2.6358914375305176
    },
    {
      "auxiliary_loss_clip": 0.01191314,
      "auxiliary_loss_mlp": 0.01033212,
      "balance_loss_clip": 1.05926216,
      "balance_loss_mlp": 1.02410436,
      "epoch": 0.3822521493416702,
      "flos": 35516945773440.0,
      "grad_norm": 2.012742553293361,
      "language_loss": 0.6625731,
      "learning_rate": 2.8337276964101403e-06,
      "loss": 0.68481839,
      "num_input_tokens_seen": 68526290,
      "step": 3179,
      "time_per_iteration": 2.733067035675049
    },
    {
      "auxiliary_loss_clip": 0.01190329,
      "auxiliary_loss_mlp": 0.01028198,
      "balance_loss_clip": 1.05608535,
      "balance_loss_mlp": 1.0192157,
      "epoch": 0.3823723922323093,
      "flos": 21069904181760.0,
      "grad_norm": 1.9662126780233782,
      "language_loss": 0.76518393,
      "learning_rate": 2.833019572158367e-06,
      "loss": 0.78736925,
      "num_input_tokens_seen": 68544725,
      "step": 3180,
      "time_per_iteration": 2.586982011795044
    },
    {
      "auxiliary_loss_clip": 0.01189061,
      "auxiliary_loss_mlp": 0.01030269,
      "balance_loss_clip": 1.05547118,
      "balance_loss_mlp": 1.02061915,
      "epoch": 0.38249263512294834,
      "flos": 19789149864960.0,
      "grad_norm": 2.2307039598650578,
      "language_loss": 0.80292767,
      "learning_rate": 2.8323113215381872e-06,
      "loss": 0.82512093,
      "num_input_tokens_seen": 68563070,
      "step": 3181,
      "time_per_iteration": 2.67223858833313
    },
    {
      "auxiliary_loss_clip": 0.01194832,
      "auxiliary_loss_mlp": 0.01026827,
      "balance_loss_clip": 1.05801857,
      "balance_loss_mlp": 1.01714182,
      "epoch": 0.38261287801358745,
      "flos": 21433930565760.0,
      "grad_norm": 2.207440980174268,
      "language_loss": 0.76069343,
      "learning_rate": 2.831602944657042e-06,
      "loss": 0.78290999,
      "num_input_tokens_seen": 68581150,
      "step": 3182,
      "time_per_iteration": 2.634432077407837
    },
    {
      "auxiliary_loss_clip": 0.01188742,
      "auxiliary_loss_mlp": 0.01033802,
      "balance_loss_clip": 1.05509782,
      "balance_loss_mlp": 1.02474177,
      "epoch": 0.38273312090422656,
      "flos": 21981568296960.0,
      "grad_norm": 2.243876166943945,
      "language_loss": 0.74379218,
      "learning_rate": 2.830894441622391e-06,
      "loss": 0.76601756,
      "num_input_tokens_seen": 68597800,
      "step": 3183,
      "time_per_iteration": 2.5751423835754395
    },
    {
      "auxiliary_loss_clip": 0.01191468,
      "auxiliary_loss_mlp": 0.01028025,
      "balance_loss_clip": 1.05660975,
      "balance_loss_mlp": 1.01875079,
      "epoch": 0.3828533637948656,
      "flos": 24790895838720.0,
      "grad_norm": 1.8228377802536682,
      "language_loss": 0.80018818,
      "learning_rate": 2.8301858125417134e-06,
      "loss": 0.82238317,
      "num_input_tokens_seen": 68617640,
      "step": 3184,
      "time_per_iteration": 2.619372606277466
    },
    {
      "auxiliary_loss_clip": 0.01191081,
      "auxiliary_loss_mlp": 0.01028512,
      "balance_loss_clip": 1.05862355,
      "balance_loss_mlp": 1.02041459,
      "epoch": 0.38297360668550473,
      "flos": 22455445449600.0,
      "grad_norm": 1.8110306242273833,
      "language_loss": 0.73942018,
      "learning_rate": 2.8294770575225082e-06,
      "loss": 0.76161611,
      "num_input_tokens_seen": 68637770,
      "step": 3185,
      "time_per_iteration": 2.5666236877441406
    },
    {
      "auxiliary_loss_clip": 0.01194891,
      "auxiliary_loss_mlp": 0.01034029,
      "balance_loss_clip": 1.06027102,
      "balance_loss_mlp": 1.02454591,
      "epoch": 0.3830938495761438,
      "flos": 24896903852160.0,
      "grad_norm": 1.9929255403409556,
      "language_loss": 0.8354668,
      "learning_rate": 2.828768176672293e-06,
      "loss": 0.85775596,
      "num_input_tokens_seen": 68656885,
      "step": 3186,
      "time_per_iteration": 2.6358909606933594
    },
    {
      "auxiliary_loss_clip": 0.01192941,
      "auxiliary_loss_mlp": 0.01031263,
      "balance_loss_clip": 1.05677152,
      "balance_loss_mlp": 1.02196431,
      "epoch": 0.3832140924667829,
      "flos": 33036236784000.0,
      "grad_norm": 1.8487417250686022,
      "language_loss": 0.71568233,
      "learning_rate": 2.8280591700986044e-06,
      "loss": 0.73792434,
      "num_input_tokens_seen": 68678750,
      "step": 3187,
      "time_per_iteration": 2.687640428543091
    },
    {
      "auxiliary_loss_clip": 0.01191836,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 1.05779898,
      "balance_loss_mlp": 1.0191381,
      "epoch": 0.383334335357422,
      "flos": 31903721896320.0,
      "grad_norm": 1.7912848637053769,
      "language_loss": 0.74867821,
      "learning_rate": 2.827350037908999e-06,
      "loss": 0.77087498,
      "num_input_tokens_seen": 68698190,
      "step": 3188,
      "time_per_iteration": 2.722395896911621
    },
    {
      "auxiliary_loss_clip": 0.01195343,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.05903327,
      "balance_loss_mlp": 1.01738167,
      "epoch": 0.38345457824806106,
      "flos": 19791915212160.0,
      "grad_norm": 2.277767895225166,
      "language_loss": 0.79013944,
      "learning_rate": 2.8266407802110496e-06,
      "loss": 0.81236196,
      "num_input_tokens_seen": 68716445,
      "step": 3189,
      "time_per_iteration": 2.539978504180908
    },
    {
      "auxiliary_loss_clip": 0.01196495,
      "auxiliary_loss_mlp": 0.0103323,
      "balance_loss_clip": 1.05910897,
      "balance_loss_mlp": 1.02388442,
      "epoch": 0.3835748211387002,
      "flos": 22419391173120.0,
      "grad_norm": 1.9870613679879767,
      "language_loss": 0.75845164,
      "learning_rate": 2.8259313971123515e-06,
      "loss": 0.78074884,
      "num_input_tokens_seen": 68737565,
      "step": 3190,
      "time_per_iteration": 2.6403050422668457
    },
    {
      "auxiliary_loss_clip": 0.01192213,
      "auxiliary_loss_mlp": 0.01028585,
      "balance_loss_clip": 1.06018293,
      "balance_loss_mlp": 1.02044916,
      "epoch": 0.3836950640293393,
      "flos": 25118436983040.0,
      "grad_norm": 1.519396656386913,
      "language_loss": 0.78397357,
      "learning_rate": 2.8252218887205166e-06,
      "loss": 0.80618161,
      "num_input_tokens_seen": 68758255,
      "step": 3191,
      "time_per_iteration": 2.6027517318725586
    },
    {
      "auxiliary_loss_clip": 0.01193039,
      "auxiliary_loss_mlp": 0.01031694,
      "balance_loss_clip": 1.05939662,
      "balance_loss_mlp": 1.0224669,
      "epoch": 0.38381530691997834,
      "flos": 21799213925760.0,
      "grad_norm": 1.7897019484556518,
      "language_loss": 0.80724239,
      "learning_rate": 2.824512255143178e-06,
      "loss": 0.82948977,
      "num_input_tokens_seen": 68777490,
      "step": 3192,
      "time_per_iteration": 2.636648654937744
    },
    {
      "auxiliary_loss_clip": 0.01195898,
      "auxiliary_loss_mlp": 0.01032902,
      "balance_loss_clip": 1.06121182,
      "balance_loss_mlp": 1.02422905,
      "epoch": 0.38393554981061745,
      "flos": 21252689516160.0,
      "grad_norm": 1.9108977982347612,
      "language_loss": 0.79319352,
      "learning_rate": 2.8238024964879855e-06,
      "loss": 0.81548154,
      "num_input_tokens_seen": 68798385,
      "step": 3193,
      "time_per_iteration": 2.608508348464966
    },
    {
      "auxiliary_loss_clip": 0.01193481,
      "auxiliary_loss_mlp": 0.01032353,
      "balance_loss_clip": 1.05875993,
      "balance_loss_mlp": 1.02269089,
      "epoch": 0.38405579270125656,
      "flos": 17019360218880.0,
      "grad_norm": 3.5554935232102194,
      "language_loss": 0.76501131,
      "learning_rate": 2.8230926128626095e-06,
      "loss": 0.78726971,
      "num_input_tokens_seen": 68816880,
      "step": 3194,
      "time_per_iteration": 2.6624956130981445
    },
    {
      "auxiliary_loss_clip": 0.01189771,
      "auxiliary_loss_mlp": 0.01027751,
      "balance_loss_clip": 1.0545876,
      "balance_loss_mlp": 1.01875651,
      "epoch": 0.3841760355918956,
      "flos": 21835375943040.0,
      "grad_norm": 2.126888241148747,
      "language_loss": 0.7956059,
      "learning_rate": 2.822382604374738e-06,
      "loss": 0.81778121,
      "num_input_tokens_seen": 68835805,
      "step": 3195,
      "time_per_iteration": 2.6149752140045166
    },
    {
      "auxiliary_loss_clip": 0.01192267,
      "auxiliary_loss_mlp": 0.01033808,
      "balance_loss_clip": 1.05944991,
      "balance_loss_mlp": 1.02413368,
      "epoch": 0.3842962784825347,
      "flos": 25915114684800.0,
      "grad_norm": 7.855529091230539,
      "language_loss": 0.66235197,
      "learning_rate": 2.8216724711320793e-06,
      "loss": 0.68461275,
      "num_input_tokens_seen": 68854930,
      "step": 3196,
      "time_per_iteration": 2.6499576568603516
    },
    {
      "auxiliary_loss_clip": 0.01187568,
      "auxiliary_loss_mlp": 0.01029975,
      "balance_loss_clip": 1.05477142,
      "balance_loss_mlp": 1.02164853,
      "epoch": 0.38441652137317384,
      "flos": 25337492075520.0,
      "grad_norm": 1.624583601793344,
      "language_loss": 0.79442132,
      "learning_rate": 2.820962213242361e-06,
      "loss": 0.81659675,
      "num_input_tokens_seen": 68874260,
      "step": 3197,
      "time_per_iteration": 2.641700029373169
    },
    {
      "auxiliary_loss_clip": 0.01187195,
      "auxiliary_loss_mlp": 0.01026496,
      "balance_loss_clip": 1.05506706,
      "balance_loss_mlp": 1.0173645,
      "epoch": 0.3845367642638129,
      "flos": 18113486446080.0,
      "grad_norm": 2.476095434660787,
      "language_loss": 0.84013003,
      "learning_rate": 2.8202518308133264e-06,
      "loss": 0.86226696,
      "num_input_tokens_seen": 68891535,
      "step": 3198,
      "time_per_iteration": 2.6289026737213135
    },
    {
      "auxiliary_loss_clip": 0.01193742,
      "auxiliary_loss_mlp": 0.01030361,
      "balance_loss_clip": 1.05858076,
      "balance_loss_mlp": 1.02062762,
      "epoch": 0.384657007154452,
      "flos": 25228395492480.0,
      "grad_norm": 1.6908110539610486,
      "language_loss": 0.73561645,
      "learning_rate": 2.8195413239527426e-06,
      "loss": 0.75785744,
      "num_input_tokens_seen": 68911275,
      "step": 3199,
      "time_per_iteration": 2.591724395751953
    },
    {
      "auxiliary_loss_clip": 0.01191637,
      "auxiliary_loss_mlp": 0.010296,
      "balance_loss_clip": 1.05662489,
      "balance_loss_mlp": 1.02095687,
      "epoch": 0.38477725004509106,
      "flos": 19865855358720.0,
      "grad_norm": 2.3231683171211985,
      "language_loss": 0.80635762,
      "learning_rate": 2.8188306927683906e-06,
      "loss": 0.82857001,
      "num_input_tokens_seen": 68930745,
      "step": 3200,
      "time_per_iteration": 3.5240960121154785
    },
    {
      "auxiliary_loss_clip": 0.01190713,
      "auxiliary_loss_mlp": 0.01024723,
      "balance_loss_clip": 1.05756736,
      "balance_loss_mlp": 1.01665258,
      "epoch": 0.38489749293573017,
      "flos": 18259391491200.0,
      "grad_norm": 2.140208620984206,
      "language_loss": 0.74690646,
      "learning_rate": 2.818119937368074e-06,
      "loss": 0.76906085,
      "num_input_tokens_seen": 68949380,
      "step": 3201,
      "time_per_iteration": 4.463828802108765
    },
    {
      "auxiliary_loss_clip": 0.01194316,
      "auxiliary_loss_mlp": 0.01030159,
      "balance_loss_clip": 1.05787492,
      "balance_loss_mlp": 1.02062845,
      "epoch": 0.3850177358263693,
      "flos": 24389163152640.0,
      "grad_norm": 1.9523019421707892,
      "language_loss": 0.65299797,
      "learning_rate": 2.817409057859613e-06,
      "loss": 0.67524278,
      "num_input_tokens_seen": 68968370,
      "step": 3202,
      "time_per_iteration": 3.4558329582214355
    },
    {
      "auxiliary_loss_clip": 0.01196703,
      "auxiliary_loss_mlp": 0.01030787,
      "balance_loss_clip": 1.05981588,
      "balance_loss_mlp": 1.02140522,
      "epoch": 0.38513797871700833,
      "flos": 17671533505920.0,
      "grad_norm": 3.0436928605978126,
      "language_loss": 0.79172671,
      "learning_rate": 2.8166980543508482e-06,
      "loss": 0.81400162,
      "num_input_tokens_seen": 68984260,
      "step": 3203,
      "time_per_iteration": 2.6152243614196777
    },
    {
      "auxiliary_loss_clip": 0.01191773,
      "auxiliary_loss_mlp": 0.01036944,
      "balance_loss_clip": 1.05783486,
      "balance_loss_mlp": 1.02749062,
      "epoch": 0.38525822160764744,
      "flos": 25739583897600.0,
      "grad_norm": 2.399983436524174,
      "language_loss": 0.79530632,
      "learning_rate": 2.815986926949638e-06,
      "loss": 0.81759346,
      "num_input_tokens_seen": 69002760,
      "step": 3204,
      "time_per_iteration": 2.6515142917633057
    },
    {
      "auxiliary_loss_clip": 0.01189788,
      "auxiliary_loss_mlp": 0.01031156,
      "balance_loss_clip": 1.05733442,
      "balance_loss_mlp": 1.0224185,
      "epoch": 0.38537846449828655,
      "flos": 20193647898240.0,
      "grad_norm": 2.3239511751002655,
      "language_loss": 0.80174768,
      "learning_rate": 2.8152756757638597e-06,
      "loss": 0.82395715,
      "num_input_tokens_seen": 69021260,
      "step": 3205,
      "time_per_iteration": 2.674189329147339
    },
    {
      "auxiliary_loss_clip": 0.01195753,
      "auxiliary_loss_mlp": 0.01031213,
      "balance_loss_clip": 1.06020045,
      "balance_loss_mlp": 1.02249861,
      "epoch": 0.3854987073889256,
      "flos": 23039352938880.0,
      "grad_norm": 1.9393974528063584,
      "language_loss": 0.84911358,
      "learning_rate": 2.8145643009014093e-06,
      "loss": 0.87138319,
      "num_input_tokens_seen": 69039755,
      "step": 3206,
      "time_per_iteration": 2.653270959854126
    },
    {
      "auxiliary_loss_clip": 0.01189414,
      "auxiliary_loss_mlp": 0.01027221,
      "balance_loss_clip": 1.0576874,
      "balance_loss_mlp": 1.01924551,
      "epoch": 0.3856189502795647,
      "flos": 20190631155840.0,
      "grad_norm": 2.0485828495683647,
      "language_loss": 0.7951473,
      "learning_rate": 2.813852802470202e-06,
      "loss": 0.81731367,
      "num_input_tokens_seen": 69057650,
      "step": 3207,
      "time_per_iteration": 2.5815305709838867
    },
    {
      "auxiliary_loss_clip": 0.01194078,
      "auxiliary_loss_mlp": 0.01032468,
      "balance_loss_clip": 1.05828798,
      "balance_loss_mlp": 1.02321148,
      "epoch": 0.38573919317020383,
      "flos": 25702631781120.0,
      "grad_norm": 1.6613355454682603,
      "language_loss": 0.72145391,
      "learning_rate": 2.8131411805781717e-06,
      "loss": 0.7437194,
      "num_input_tokens_seen": 69077775,
      "step": 3208,
      "time_per_iteration": 2.7097666263580322
    },
    {
      "auxiliary_loss_clip": 0.0119635,
      "auxiliary_loss_mlp": 0.01039717,
      "balance_loss_clip": 1.06167185,
      "balance_loss_mlp": 1.03027606,
      "epoch": 0.3858594360608429,
      "flos": 29821405628160.0,
      "grad_norm": 2.334944316437693,
      "language_loss": 0.64548469,
      "learning_rate": 2.8124294353332707e-06,
      "loss": 0.66784537,
      "num_input_tokens_seen": 69096450,
      "step": 3209,
      "time_per_iteration": 2.70438551902771
    },
    {
      "auxiliary_loss_clip": 0.0119157,
      "auxiliary_loss_mlp": 0.01027428,
      "balance_loss_clip": 1.05783725,
      "balance_loss_mlp": 1.01804614,
      "epoch": 0.385979678951482,
      "flos": 24790428961920.0,
      "grad_norm": 1.8310333843143356,
      "language_loss": 0.77235103,
      "learning_rate": 2.8117175668434713e-06,
      "loss": 0.794541,
      "num_input_tokens_seen": 69116110,
      "step": 3210,
      "time_per_iteration": 2.671754837036133
    },
    {
      "auxiliary_loss_clip": 0.01191616,
      "auxiliary_loss_mlp": 0.01031261,
      "balance_loss_clip": 1.05602264,
      "balance_loss_mlp": 1.02122355,
      "epoch": 0.3860999218421211,
      "flos": 21287881866240.0,
      "grad_norm": 2.432974561021584,
      "language_loss": 0.70025963,
      "learning_rate": 2.811005575216762e-06,
      "loss": 0.7224884,
      "num_input_tokens_seen": 69134825,
      "step": 3211,
      "time_per_iteration": 2.6354570388793945
    },
    {
      "auxiliary_loss_clip": 0.01190394,
      "auxiliary_loss_mlp": 0.01031008,
      "balance_loss_clip": 1.05620146,
      "balance_loss_mlp": 1.02228737,
      "epoch": 0.38622016473276016,
      "flos": 24536720223360.0,
      "grad_norm": 1.555229631036398,
      "language_loss": 0.78645945,
      "learning_rate": 2.8102934605611513e-06,
      "loss": 0.80867344,
      "num_input_tokens_seen": 69156460,
      "step": 3212,
      "time_per_iteration": 2.65108585357666
    },
    {
      "auxiliary_loss_clip": 0.01195875,
      "auxiliary_loss_mlp": 0.0103118,
      "balance_loss_clip": 1.06058455,
      "balance_loss_mlp": 1.02219748,
      "epoch": 0.3863404076233993,
      "flos": 20558212986240.0,
      "grad_norm": 2.3288176567356977,
      "language_loss": 0.67360508,
      "learning_rate": 2.8095812229846665e-06,
      "loss": 0.69587564,
      "num_input_tokens_seen": 69176420,
      "step": 3213,
      "time_per_iteration": 2.6303000450134277
    },
    {
      "auxiliary_loss_clip": 0.01193484,
      "auxiliary_loss_mlp": 0.01031744,
      "balance_loss_clip": 1.0580101,
      "balance_loss_mlp": 1.02231431,
      "epoch": 0.3864606505140384,
      "flos": 22346277039360.0,
      "grad_norm": 3.203625899789541,
      "language_loss": 0.68925166,
      "learning_rate": 2.808868862595355e-06,
      "loss": 0.71150392,
      "num_input_tokens_seen": 69196665,
      "step": 3214,
      "time_per_iteration": 2.5829410552978516
    },
    {
      "auxiliary_loss_clip": 0.01192778,
      "auxiliary_loss_mlp": 0.01030787,
      "balance_loss_clip": 1.05886412,
      "balance_loss_mlp": 1.02231741,
      "epoch": 0.38658089340467744,
      "flos": 25703601448320.0,
      "grad_norm": 3.0924189191913225,
      "language_loss": 0.79324448,
      "learning_rate": 2.8081563795012795e-06,
      "loss": 0.81548011,
      "num_input_tokens_seen": 69216290,
      "step": 3215,
      "time_per_iteration": 2.6562440395355225
    },
    {
      "auxiliary_loss_clip": 0.01191998,
      "auxiliary_loss_mlp": 0.01028157,
      "balance_loss_clip": 1.05710864,
      "balance_loss_mlp": 1.01904333,
      "epoch": 0.38670113629531655,
      "flos": 33802534558080.0,
      "grad_norm": 1.9983629983528302,
      "language_loss": 0.74006164,
      "learning_rate": 2.807443773810524e-06,
      "loss": 0.76226318,
      "num_input_tokens_seen": 69237550,
      "step": 3216,
      "time_per_iteration": 2.7039482593536377
    },
    {
      "auxiliary_loss_clip": 0.01194262,
      "auxiliary_loss_mlp": 0.01033817,
      "balance_loss_clip": 1.0605166,
      "balance_loss_mlp": 1.0249958,
      "epoch": 0.3868213791859556,
      "flos": 23331522165120.0,
      "grad_norm": 2.219528902930719,
      "language_loss": 0.89523864,
      "learning_rate": 2.80673104563119e-06,
      "loss": 0.91751945,
      "num_input_tokens_seen": 69258175,
      "step": 3217,
      "time_per_iteration": 2.7612078189849854
    },
    {
      "auxiliary_loss_clip": 0.01190295,
      "auxiliary_loss_mlp": 0.01030516,
      "balance_loss_clip": 1.05871487,
      "balance_loss_mlp": 1.02201009,
      "epoch": 0.3869416220765947,
      "flos": 18441530380800.0,
      "grad_norm": 1.9155585584752282,
      "language_loss": 0.79043812,
      "learning_rate": 2.8060181950713976e-06,
      "loss": 0.81264621,
      "num_input_tokens_seen": 69274965,
      "step": 3218,
      "time_per_iteration": 2.5600054264068604
    },
    {
      "auxiliary_loss_clip": 0.01195082,
      "auxiliary_loss_mlp": 0.01030664,
      "balance_loss_clip": 1.05996907,
      "balance_loss_mlp": 1.02206874,
      "epoch": 0.3870618649672338,
      "flos": 15632992938240.0,
      "grad_norm": 2.0344755238388834,
      "language_loss": 0.80963504,
      "learning_rate": 2.805305222239286e-06,
      "loss": 0.83189249,
      "num_input_tokens_seen": 69292220,
      "step": 3219,
      "time_per_iteration": 2.6035003662109375
    },
    {
      "auxiliary_loss_clip": 0.0119132,
      "auxiliary_loss_mlp": 0.010249,
      "balance_loss_clip": 1.0584476,
      "balance_loss_mlp": 1.0159893,
      "epoch": 0.3871821078578729,
      "flos": 23513804709120.0,
      "grad_norm": 2.050228555602415,
      "language_loss": 0.73911929,
      "learning_rate": 2.8045921272430118e-06,
      "loss": 0.76128155,
      "num_input_tokens_seen": 69311900,
      "step": 3220,
      "time_per_iteration": 2.6219983100891113
    },
    {
      "auxiliary_loss_clip": 0.01198047,
      "auxiliary_loss_mlp": 0.01034745,
      "balance_loss_clip": 1.06061816,
      "balance_loss_mlp": 1.02439749,
      "epoch": 0.387302350748512,
      "flos": 17778259791360.0,
      "grad_norm": 2.6124481554435013,
      "language_loss": 0.76923823,
      "learning_rate": 2.803878910190753e-06,
      "loss": 0.79156613,
      "num_input_tokens_seen": 69328820,
      "step": 3221,
      "time_per_iteration": 2.588480234146118
    },
    {
      "auxiliary_loss_clip": 0.01193639,
      "auxiliary_loss_mlp": 0.01034304,
      "balance_loss_clip": 1.05918765,
      "balance_loss_mlp": 1.02560759,
      "epoch": 0.3874225936391511,
      "flos": 11503409097600.0,
      "grad_norm": 2.384445493743594,
      "language_loss": 0.81657386,
      "learning_rate": 2.8031655711907017e-06,
      "loss": 0.83885336,
      "num_input_tokens_seen": 69342525,
      "step": 3222,
      "time_per_iteration": 2.560624361038208
    },
    {
      "auxiliary_loss_clip": 0.0119599,
      "auxiliary_loss_mlp": 0.01037222,
      "balance_loss_clip": 1.05902219,
      "balance_loss_mlp": 1.02840054,
      "epoch": 0.38754283652979016,
      "flos": 21945154884480.0,
      "grad_norm": 2.1510354386589654,
      "language_loss": 0.80607271,
      "learning_rate": 2.8024521103510723e-06,
      "loss": 0.82840478,
      "num_input_tokens_seen": 69359295,
      "step": 3223,
      "time_per_iteration": 2.6486029624938965
    },
    {
      "auxiliary_loss_clip": 0.0119304,
      "auxiliary_loss_mlp": 0.01031978,
      "balance_loss_clip": 1.0585804,
      "balance_loss_mlp": 1.02219081,
      "epoch": 0.38766307942042927,
      "flos": 21175984022400.0,
      "grad_norm": 1.7547730941751263,
      "language_loss": 0.75235468,
      "learning_rate": 2.8017385277800952e-06,
      "loss": 0.77460486,
      "num_input_tokens_seen": 69377650,
      "step": 3224,
      "time_per_iteration": 2.663231611251831
    },
    {
      "auxiliary_loss_clip": 0.01196901,
      "auxiliary_loss_mlp": 0.01029167,
      "balance_loss_clip": 1.05979586,
      "balance_loss_mlp": 1.02007115,
      "epoch": 0.3877833223110684,
      "flos": 27417294391680.0,
      "grad_norm": 2.205209298956628,
      "language_loss": 0.74877369,
      "learning_rate": 2.8010248235860213e-06,
      "loss": 0.77103442,
      "num_input_tokens_seen": 69397765,
      "step": 3225,
      "time_per_iteration": 2.7112717628479004
    },
    {
      "auxiliary_loss_clip": 0.01093872,
      "auxiliary_loss_mlp": 0.01002758,
      "balance_loss_clip": 1.0257411,
      "balance_loss_mlp": 1.00146472,
      "epoch": 0.38790356520170743,
      "flos": 64500019879680.0,
      "grad_norm": 0.830983650474119,
      "language_loss": 0.62782979,
      "learning_rate": 2.8003109978771192e-06,
      "loss": 0.64879608,
      "num_input_tokens_seen": 69458930,
      "step": 3226,
      "time_per_iteration": 3.241405963897705
    },
    {
      "auxiliary_loss_clip": 0.01191462,
      "auxiliary_loss_mlp": 0.01029783,
      "balance_loss_clip": 1.05646324,
      "balance_loss_mlp": 1.02081275,
      "epoch": 0.38802380809234654,
      "flos": 22345415112960.0,
      "grad_norm": 2.233610272624625,
      "language_loss": 0.79163599,
      "learning_rate": 2.799597050761674e-06,
      "loss": 0.8138485,
      "num_input_tokens_seen": 69475135,
      "step": 3227,
      "time_per_iteration": 4.448452472686768
    },
    {
      "auxiliary_loss_clip": 0.0119584,
      "auxiliary_loss_mlp": 0.01026786,
      "balance_loss_clip": 1.06001472,
      "balance_loss_mlp": 1.01782107,
      "epoch": 0.38814405098298566,
      "flos": 25261361199360.0,
      "grad_norm": 1.8737488259412778,
      "language_loss": 0.78955805,
      "learning_rate": 2.7988829823479924e-06,
      "loss": 0.81178427,
      "num_input_tokens_seen": 69493525,
      "step": 3228,
      "time_per_iteration": 4.446200370788574
    },
    {
      "auxiliary_loss_clip": 0.01195219,
      "auxiliary_loss_mlp": 0.010299,
      "balance_loss_clip": 1.05892944,
      "balance_loss_mlp": 1.02059579,
      "epoch": 0.3882642938736247,
      "flos": 18841180078080.0,
      "grad_norm": 1.9061825338533755,
      "language_loss": 0.63711017,
      "learning_rate": 2.7981687927443976e-06,
      "loss": 0.65936136,
      "num_input_tokens_seen": 69510325,
      "step": 3229,
      "time_per_iteration": 2.6122095584869385
    },
    {
      "auxiliary_loss_clip": 0.01190977,
      "auxiliary_loss_mlp": 0.01027634,
      "balance_loss_clip": 1.05614889,
      "balance_loss_mlp": 1.01881242,
      "epoch": 0.3883845367642638,
      "flos": 21652806090240.0,
      "grad_norm": 2.095842959735351,
      "language_loss": 0.85719299,
      "learning_rate": 2.797454482059231e-06,
      "loss": 0.87937915,
      "num_input_tokens_seen": 69530480,
      "step": 3230,
      "time_per_iteration": 2.6216824054718018
    },
    {
      "auxiliary_loss_clip": 0.01193974,
      "auxiliary_loss_mlp": 0.01030565,
      "balance_loss_clip": 1.05899549,
      "balance_loss_mlp": 1.02155924,
      "epoch": 0.3885047796549029,
      "flos": 20557530627840.0,
      "grad_norm": 1.8588088663094848,
      "language_loss": 0.84536517,
      "learning_rate": 2.7967400504008537e-06,
      "loss": 0.86761057,
      "num_input_tokens_seen": 69549780,
      "step": 3231,
      "time_per_iteration": 2.6505892276763916
    },
    {
      "auxiliary_loss_clip": 0.01092388,
      "auxiliary_loss_mlp": 0.01002999,
      "balance_loss_clip": 1.023857,
      "balance_loss_mlp": 1.0017947,
      "epoch": 0.388625022545542,
      "flos": 64325491695360.0,
      "grad_norm": 0.7884258696350398,
      "language_loss": 0.57407606,
      "learning_rate": 2.7960254978776456e-06,
      "loss": 0.59502983,
      "num_input_tokens_seen": 69611870,
      "step": 3232,
      "time_per_iteration": 3.231231212615967
    },
    {
      "auxiliary_loss_clip": 0.01194122,
      "auxiliary_loss_mlp": 0.01032301,
      "balance_loss_clip": 1.05912733,
      "balance_loss_mlp": 1.02380192,
      "epoch": 0.3887452654361811,
      "flos": 18113881495680.0,
      "grad_norm": 1.8537460104822083,
      "language_loss": 0.81638199,
      "learning_rate": 2.7953108245980006e-06,
      "loss": 0.83864617,
      "num_input_tokens_seen": 69630385,
      "step": 3233,
      "time_per_iteration": 2.593148708343506
    },
    {
      "auxiliary_loss_clip": 0.01194481,
      "auxiliary_loss_mlp": 0.01032325,
      "balance_loss_clip": 1.05989301,
      "balance_loss_mlp": 1.0235095,
      "epoch": 0.38886550832682015,
      "flos": 24975261371520.0,
      "grad_norm": 1.5451196702316727,
      "language_loss": 0.73836505,
      "learning_rate": 2.7945960306703365e-06,
      "loss": 0.76063311,
      "num_input_tokens_seen": 69653370,
      "step": 3234,
      "time_per_iteration": 2.736543655395508
    },
    {
      "auxiliary_loss_clip": 0.01194595,
      "auxiliary_loss_mlp": 0.01033916,
      "balance_loss_clip": 1.06019831,
      "balance_loss_mlp": 1.02502298,
      "epoch": 0.38898575121745926,
      "flos": 27199496275200.0,
      "grad_norm": 2.307723044175462,
      "language_loss": 0.65533149,
      "learning_rate": 2.7938811162030865e-06,
      "loss": 0.6776166,
      "num_input_tokens_seen": 69673635,
      "step": 3235,
      "time_per_iteration": 2.7048146724700928
    },
    {
      "auxiliary_loss_clip": 0.01192627,
      "auxiliary_loss_mlp": 0.01026777,
      "balance_loss_clip": 1.05948031,
      "balance_loss_mlp": 1.01886201,
      "epoch": 0.3891059941080984,
      "flos": 28763728727040.0,
      "grad_norm": 1.7282165515663086,
      "language_loss": 0.82264805,
      "learning_rate": 2.793166081304702e-06,
      "loss": 0.84484208,
      "num_input_tokens_seen": 69694130,
      "step": 3236,
      "time_per_iteration": 2.7054617404937744
    },
    {
      "auxiliary_loss_clip": 0.01194637,
      "auxiliary_loss_mlp": 0.01032557,
      "balance_loss_clip": 1.05842423,
      "balance_loss_mlp": 1.02365804,
      "epoch": 0.38922623699873743,
      "flos": 22893447893760.0,
      "grad_norm": 2.292057781469452,
      "language_loss": 0.82493806,
      "learning_rate": 2.7924509260836543e-06,
      "loss": 0.84720999,
      "num_input_tokens_seen": 69713255,
      "step": 3237,
      "time_per_iteration": 2.6110990047454834
    },
    {
      "auxiliary_loss_clip": 0.01193169,
      "auxiliary_loss_mlp": 0.01030629,
      "balance_loss_clip": 1.05823302,
      "balance_loss_mlp": 1.0226835,
      "epoch": 0.38934647988937654,
      "flos": 19792418002560.0,
      "grad_norm": 1.8022835319153094,
      "language_loss": 0.68286812,
      "learning_rate": 2.791735650648431e-06,
      "loss": 0.70510602,
      "num_input_tokens_seen": 69732375,
      "step": 3238,
      "time_per_iteration": 2.6523356437683105
    },
    {
      "auxiliary_loss_clip": 0.01189237,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.05564857,
      "balance_loss_mlp": 1.02048373,
      "epoch": 0.38946672278001565,
      "flos": 19202081978880.0,
      "grad_norm": 2.5041782052839783,
      "language_loss": 0.7446692,
      "learning_rate": 2.791020255107538e-06,
      "loss": 0.76686138,
      "num_input_tokens_seen": 69749745,
      "step": 3239,
      "time_per_iteration": 2.700773000717163
    },
    {
      "auxiliary_loss_clip": 0.01194558,
      "auxiliary_loss_mlp": 0.01033523,
      "balance_loss_clip": 1.05895829,
      "balance_loss_mlp": 1.02444518,
      "epoch": 0.3895869656706547,
      "flos": 24936477661440.0,
      "grad_norm": 1.5754106879970484,
      "language_loss": 0.80704045,
      "learning_rate": 2.7903047395695023e-06,
      "loss": 0.82932127,
      "num_input_tokens_seen": 69769645,
      "step": 3240,
      "time_per_iteration": 2.668442487716675
    },
    {
      "auxiliary_loss_clip": 0.01193055,
      "auxiliary_loss_mlp": 0.01030642,
      "balance_loss_clip": 1.05955195,
      "balance_loss_mlp": 1.02127826,
      "epoch": 0.3897072085612938,
      "flos": 24133622820480.0,
      "grad_norm": 2.5041497238411266,
      "language_loss": 0.90068972,
      "learning_rate": 2.789589104142865e-06,
      "loss": 0.92292672,
      "num_input_tokens_seen": 69787270,
      "step": 3241,
      "time_per_iteration": 2.637202739715576
    },
    {
      "auxiliary_loss_clip": 0.0118862,
      "auxiliary_loss_mlp": 0.01038783,
      "balance_loss_clip": 1.0551784,
      "balance_loss_mlp": 1.02982414,
      "epoch": 0.3898274514519329,
      "flos": 17166342672000.0,
      "grad_norm": 1.9429702008858936,
      "language_loss": 0.76348758,
      "learning_rate": 2.7888733489361895e-06,
      "loss": 0.78576159,
      "num_input_tokens_seen": 69805685,
      "step": 3242,
      "time_per_iteration": 2.578773021697998
    },
    {
      "auxiliary_loss_clip": 0.01088757,
      "auxiliary_loss_mlp": 0.01001295,
      "balance_loss_clip": 1.02041864,
      "balance_loss_mlp": 1.00004888,
      "epoch": 0.389947694342572,
      "flos": 66074807952000.0,
      "grad_norm": 0.7319789052183306,
      "language_loss": 0.58670521,
      "learning_rate": 2.788157474058054e-06,
      "loss": 0.6076057,
      "num_input_tokens_seen": 69867960,
      "step": 3243,
      "time_per_iteration": 3.3053555488586426
    },
    {
      "auxiliary_loss_clip": 0.01187337,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.05597091,
      "balance_loss_mlp": 1.02141619,
      "epoch": 0.3900679372332111,
      "flos": 25740912700800.0,
      "grad_norm": 1.5231180622392573,
      "language_loss": 0.70190382,
      "learning_rate": 2.7874414796170555e-06,
      "loss": 0.72407484,
      "num_input_tokens_seen": 69889450,
      "step": 3244,
      "time_per_iteration": 2.646599292755127
    },
    {
      "auxiliary_loss_clip": 0.01195725,
      "auxiliary_loss_mlp": 0.01036968,
      "balance_loss_clip": 1.06043589,
      "balance_loss_mlp": 1.02701354,
      "epoch": 0.3901881801238502,
      "flos": 11801611808640.0,
      "grad_norm": 3.5834781482390903,
      "language_loss": 0.84272891,
      "learning_rate": 2.7867253657218113e-06,
      "loss": 0.8650558,
      "num_input_tokens_seen": 69903340,
      "step": 3245,
      "time_per_iteration": 2.5890324115753174
    },
    {
      "auxiliary_loss_clip": 0.01195341,
      "auxiliary_loss_mlp": 0.01033661,
      "balance_loss_clip": 1.05858874,
      "balance_loss_mlp": 1.02470195,
      "epoch": 0.39030842301448926,
      "flos": 27308951994240.0,
      "grad_norm": 1.6975380561057651,
      "language_loss": 0.72897351,
      "learning_rate": 2.7860091324809544e-06,
      "loss": 0.7512635,
      "num_input_tokens_seen": 69924400,
      "step": 3246,
      "time_per_iteration": 2.693162202835083
    },
    {
      "auxiliary_loss_clip": 0.01190771,
      "auxiliary_loss_mlp": 0.01029614,
      "balance_loss_clip": 1.05642235,
      "balance_loss_mlp": 1.02043462,
      "epoch": 0.39042866590512837,
      "flos": 27163334257920.0,
      "grad_norm": 1.6218606769434714,
      "language_loss": 0.81323677,
      "learning_rate": 2.7852927800031377e-06,
      "loss": 0.83544064,
      "num_input_tokens_seen": 69944565,
      "step": 3247,
      "time_per_iteration": 2.6663172245025635
    },
    {
      "auxiliary_loss_clip": 0.01194384,
      "auxiliary_loss_mlp": 0.01029129,
      "balance_loss_clip": 1.05885136,
      "balance_loss_mlp": 1.02035546,
      "epoch": 0.3905489087957674,
      "flos": 29716115886720.0,
      "grad_norm": 1.9515590022680451,
      "language_loss": 0.82509065,
      "learning_rate": 2.7845763083970298e-06,
      "loss": 0.8473258,
      "num_input_tokens_seen": 69964965,
      "step": 3248,
      "time_per_iteration": 2.630004644393921
    },
    {
      "auxiliary_loss_clip": 0.01192183,
      "auxiliary_loss_mlp": 0.01029231,
      "balance_loss_clip": 1.05744767,
      "balance_loss_mlp": 1.01965833,
      "epoch": 0.39066915168640653,
      "flos": 24498618871680.0,
      "grad_norm": 2.164691762823905,
      "language_loss": 0.81895781,
      "learning_rate": 2.7838597177713205e-06,
      "loss": 0.84117198,
      "num_input_tokens_seen": 69986055,
      "step": 3249,
      "time_per_iteration": 2.6728551387786865
    },
    {
      "auxiliary_loss_clip": 0.011969,
      "auxiliary_loss_mlp": 0.010332,
      "balance_loss_clip": 1.05966306,
      "balance_loss_mlp": 1.02468204,
      "epoch": 0.39078939457704565,
      "flos": 20558572122240.0,
      "grad_norm": 1.7534397654326224,
      "language_loss": 0.73615074,
      "learning_rate": 2.7831430082347143e-06,
      "loss": 0.75845176,
      "num_input_tokens_seen": 70005260,
      "step": 3250,
      "time_per_iteration": 2.640291690826416
    },
    {
      "auxiliary_loss_clip": 0.01192719,
      "auxiliary_loss_mlp": 0.01028989,
      "balance_loss_clip": 1.06084275,
      "balance_loss_mlp": 1.02072787,
      "epoch": 0.3909096374676847,
      "flos": 22783417557120.0,
      "grad_norm": 2.150712790926101,
      "language_loss": 0.82531488,
      "learning_rate": 2.7824261798959373e-06,
      "loss": 0.84753203,
      "num_input_tokens_seen": 70023440,
      "step": 3251,
      "time_per_iteration": 2.667616605758667
    },
    {
      "auxiliary_loss_clip": 0.01190912,
      "auxiliary_loss_mlp": 0.01027552,
      "balance_loss_clip": 1.05647099,
      "balance_loss_mlp": 1.01862919,
      "epoch": 0.3910298803583238,
      "flos": 23003119094400.0,
      "grad_norm": 2.4229605252396063,
      "language_loss": 0.79846513,
      "learning_rate": 2.78170923286373e-06,
      "loss": 0.82064974,
      "num_input_tokens_seen": 70043040,
      "step": 3252,
      "time_per_iteration": 2.5597147941589355
    },
    {
      "auxiliary_loss_clip": 0.01194402,
      "auxiliary_loss_mlp": 0.01033752,
      "balance_loss_clip": 1.06088471,
      "balance_loss_mlp": 1.02429283,
      "epoch": 0.3911501232489629,
      "flos": 24316264500480.0,
      "grad_norm": 3.4666843718143094,
      "language_loss": 0.83977151,
      "learning_rate": 2.780992167246854e-06,
      "loss": 0.86205304,
      "num_input_tokens_seen": 70060565,
      "step": 3253,
      "time_per_iteration": 4.495419502258301
    },
    {
      "auxiliary_loss_clip": 0.01088471,
      "auxiliary_loss_mlp": 0.01002482,
      "balance_loss_clip": 1.02046525,
      "balance_loss_mlp": 1.00136173,
      "epoch": 0.391270366139602,
      "flos": 60869054684160.0,
      "grad_norm": 0.9694077699865883,
      "language_loss": 0.72159481,
      "learning_rate": 2.7802749831540883e-06,
      "loss": 0.74250436,
      "num_input_tokens_seen": 70119465,
      "step": 3254,
      "time_per_iteration": 4.311952590942383
    },
    {
      "auxiliary_loss_clip": 0.01187794,
      "auxiliary_loss_mlp": 0.01031297,
      "balance_loss_clip": 1.05750179,
      "balance_loss_mlp": 1.02355194,
      "epoch": 0.3913906090302411,
      "flos": 21543494025600.0,
      "grad_norm": 1.9460581293824004,
      "language_loss": 0.81578219,
      "learning_rate": 2.7795576806942268e-06,
      "loss": 0.83797312,
      "num_input_tokens_seen": 70138270,
      "step": 3255,
      "time_per_iteration": 3.569317579269409
    },
    {
      "auxiliary_loss_clip": 0.01087888,
      "auxiliary_loss_mlp": 0.01002981,
      "balance_loss_clip": 1.01986003,
      "balance_loss_mlp": 1.00180674,
      "epoch": 0.3915108519208802,
      "flos": 49839953702400.0,
      "grad_norm": 0.7588694609925257,
      "language_loss": 0.54886818,
      "learning_rate": 2.778840259976085e-06,
      "loss": 0.56977683,
      "num_input_tokens_seen": 70193500,
      "step": 3256,
      "time_per_iteration": 3.140345573425293
    },
    {
      "auxiliary_loss_clip": 0.01192134,
      "auxiliary_loss_mlp": 0.01032412,
      "balance_loss_clip": 1.05735338,
      "balance_loss_mlp": 1.02297103,
      "epoch": 0.39163109481151925,
      "flos": 16506447960960.0,
      "grad_norm": 2.2720716599780624,
      "language_loss": 0.77095664,
      "learning_rate": 2.778122721108495e-06,
      "loss": 0.79320204,
      "num_input_tokens_seen": 70211730,
      "step": 3257,
      "time_per_iteration": 2.633363962173462
    },
    {
      "auxiliary_loss_clip": 0.01192481,
      "auxiliary_loss_mlp": 0.01029019,
      "balance_loss_clip": 1.06128716,
      "balance_loss_mlp": 1.02074575,
      "epoch": 0.39175133770215836,
      "flos": 26067484177920.0,
      "grad_norm": 2.01326619598364,
      "language_loss": 0.88672888,
      "learning_rate": 2.7774050642003076e-06,
      "loss": 0.90894389,
      "num_input_tokens_seen": 70232540,
      "step": 3258,
      "time_per_iteration": 2.649155616760254
    },
    {
      "auxiliary_loss_clip": 0.01196308,
      "auxiliary_loss_mlp": 0.01032143,
      "balance_loss_clip": 1.05903363,
      "balance_loss_mlp": 1.0235424,
      "epoch": 0.3918715805927975,
      "flos": 21872076664320.0,
      "grad_norm": 2.246279805201766,
      "language_loss": 0.93650115,
      "learning_rate": 2.7766872893603896e-06,
      "loss": 0.95878571,
      "num_input_tokens_seen": 70252515,
      "step": 3259,
      "time_per_iteration": 2.6947295665740967
    },
    {
      "auxiliary_loss_clip": 0.01191168,
      "auxiliary_loss_mlp": 0.01033223,
      "balance_loss_clip": 1.0596956,
      "balance_loss_mlp": 1.02579927,
      "epoch": 0.39199182348343653,
      "flos": 20376181837440.0,
      "grad_norm": 1.7405792317612245,
      "language_loss": 0.73518836,
      "learning_rate": 2.7759693966976275e-06,
      "loss": 0.75743228,
      "num_input_tokens_seen": 70271020,
      "step": 3260,
      "time_per_iteration": 2.6785433292388916
    },
    {
      "auxiliary_loss_clip": 0.01192758,
      "auxiliary_loss_mlp": 0.01030188,
      "balance_loss_clip": 1.05534768,
      "balance_loss_mlp": 1.02100301,
      "epoch": 0.39211206637407564,
      "flos": 21683545153920.0,
      "grad_norm": 1.8787772803443736,
      "language_loss": 0.85126507,
      "learning_rate": 2.7752513863209242e-06,
      "loss": 0.87349451,
      "num_input_tokens_seen": 70289600,
      "step": 3261,
      "time_per_iteration": 2.6209123134613037
    },
    {
      "auxiliary_loss_clip": 0.01190879,
      "auxiliary_loss_mlp": 0.01029587,
      "balance_loss_clip": 1.05958796,
      "balance_loss_mlp": 1.0212487,
      "epoch": 0.39223230926471475,
      "flos": 21066276908160.0,
      "grad_norm": 1.5986014185650639,
      "language_loss": 0.84540021,
      "learning_rate": 2.774533258339203e-06,
      "loss": 0.86760491,
      "num_input_tokens_seen": 70307060,
      "step": 3262,
      "time_per_iteration": 2.6550071239471436
    },
    {
      "auxiliary_loss_clip": 0.01194068,
      "auxiliary_loss_mlp": 0.01032288,
      "balance_loss_clip": 1.05764699,
      "balance_loss_mlp": 1.02308536,
      "epoch": 0.3923525521553538,
      "flos": 17603016312960.0,
      "grad_norm": 2.108542199073818,
      "language_loss": 0.79911351,
      "learning_rate": 2.7738150128614014e-06,
      "loss": 0.82137704,
      "num_input_tokens_seen": 70324465,
      "step": 3263,
      "time_per_iteration": 2.5640480518341064
    },
    {
      "auxiliary_loss_clip": 0.01191355,
      "auxiliary_loss_mlp": 0.01034276,
      "balance_loss_clip": 1.05798316,
      "balance_loss_mlp": 1.02532935,
      "epoch": 0.3924727950459929,
      "flos": 20558284813440.0,
      "grad_norm": 2.0261438417715962,
      "language_loss": 0.89754909,
      "learning_rate": 2.7730966499964777e-06,
      "loss": 0.91980535,
      "num_input_tokens_seen": 70341415,
      "step": 3264,
      "time_per_iteration": 2.6752047538757324
    },
    {
      "auxiliary_loss_clip": 0.0119456,
      "auxiliary_loss_mlp": 0.01031588,
      "balance_loss_clip": 1.05882549,
      "balance_loss_mlp": 1.02237928,
      "epoch": 0.39259303793663197,
      "flos": 16216110328320.0,
      "grad_norm": 2.4031616069846895,
      "language_loss": 0.80678302,
      "learning_rate": 2.772378169853408e-06,
      "loss": 0.82904452,
      "num_input_tokens_seen": 70358985,
      "step": 3265,
      "time_per_iteration": 2.6115622520446777
    },
    {
      "auxiliary_loss_clip": 0.01192874,
      "auxiliary_loss_mlp": 0.01035356,
      "balance_loss_clip": 1.06116486,
      "balance_loss_mlp": 1.02670765,
      "epoch": 0.3927132808272711,
      "flos": 16797001075200.0,
      "grad_norm": 2.000235390717877,
      "language_loss": 0.74532521,
      "learning_rate": 2.771659572541183e-06,
      "loss": 0.76760757,
      "num_input_tokens_seen": 70376915,
      "step": 3266,
      "time_per_iteration": 2.6237523555755615
    },
    {
      "auxiliary_loss_clip": 0.01194761,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.0618943,
      "balance_loss_mlp": 1.02017951,
      "epoch": 0.3928335237179102,
      "flos": 20267228908800.0,
      "grad_norm": 2.089956734655723,
      "language_loss": 0.87400949,
      "learning_rate": 2.7709408581688143e-06,
      "loss": 0.89624149,
      "num_input_tokens_seen": 70396900,
      "step": 3267,
      "time_per_iteration": 2.6279237270355225
    },
    {
      "auxiliary_loss_clip": 0.01190699,
      "auxiliary_loss_mlp": 0.01029617,
      "balance_loss_clip": 1.0597322,
      "balance_loss_mlp": 1.02060485,
      "epoch": 0.39295376660854925,
      "flos": 24973250209920.0,
      "grad_norm": 1.5989540218343723,
      "language_loss": 0.87770224,
      "learning_rate": 2.7702220268453307e-06,
      "loss": 0.89990544,
      "num_input_tokens_seen": 70417260,
      "step": 3268,
      "time_per_iteration": 2.7000648975372314
    },
    {
      "auxiliary_loss_clip": 0.01192962,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.05732775,
      "balance_loss_mlp": 1.01705194,
      "epoch": 0.39307400949918836,
      "flos": 18697788984960.0,
      "grad_norm": 8.078338470544228,
      "language_loss": 0.85227799,
      "learning_rate": 2.7695030786797785e-06,
      "loss": 0.87447286,
      "num_input_tokens_seen": 70433155,
      "step": 3269,
      "time_per_iteration": 2.521491527557373
    },
    {
      "auxiliary_loss_clip": 0.01194959,
      "auxiliary_loss_mlp": 0.01028973,
      "balance_loss_clip": 1.06102693,
      "balance_loss_mlp": 1.01977646,
      "epoch": 0.39319425238982747,
      "flos": 22415476590720.0,
      "grad_norm": 2.3228770999863824,
      "language_loss": 0.74773109,
      "learning_rate": 2.7687840137812206e-06,
      "loss": 0.76997048,
      "num_input_tokens_seen": 70451240,
      "step": 3270,
      "time_per_iteration": 2.620899200439453
    },
    {
      "auxiliary_loss_clip": 0.01087294,
      "auxiliary_loss_mlp": 0.01001094,
      "balance_loss_clip": 1.01943111,
      "balance_loss_mlp": 0.99996704,
      "epoch": 0.3933144952804665,
      "flos": 66192954762240.0,
      "grad_norm": 0.7955614934483801,
      "language_loss": 0.6202687,
      "learning_rate": 2.7680648322587395e-06,
      "loss": 0.64115262,
      "num_input_tokens_seen": 70516115,
      "step": 3271,
      "time_per_iteration": 3.2391247749328613
    },
    {
      "auxiliary_loss_clip": 0.01192798,
      "auxiliary_loss_mlp": 0.01023658,
      "balance_loss_clip": 1.05871797,
      "balance_loss_mlp": 1.01487875,
      "epoch": 0.39343473817110564,
      "flos": 15487159720320.0,
      "grad_norm": 1.9615709329147732,
      "language_loss": 0.8099609,
      "learning_rate": 2.7673455342214334e-06,
      "loss": 0.83212543,
      "num_input_tokens_seen": 70533105,
      "step": 3272,
      "time_per_iteration": 2.661306381225586
    },
    {
      "auxiliary_loss_clip": 0.01192554,
      "auxiliary_loss_mlp": 0.01042689,
      "balance_loss_clip": 1.06008434,
      "balance_loss_mlp": 1.0341239,
      "epoch": 0.39355498106174475,
      "flos": 21324905809920.0,
      "grad_norm": 1.9367014866011494,
      "language_loss": 0.75913346,
      "learning_rate": 2.7666261197784198e-06,
      "loss": 0.78148592,
      "num_input_tokens_seen": 70551920,
      "step": 3273,
      "time_per_iteration": 2.668227434158325
    },
    {
      "auxiliary_loss_clip": 0.01190797,
      "auxiliary_loss_mlp": 0.01030994,
      "balance_loss_clip": 1.05888677,
      "balance_loss_mlp": 1.02192211,
      "epoch": 0.3936752239523838,
      "flos": 13296357400320.0,
      "grad_norm": 2.0476278940987878,
      "language_loss": 0.76230186,
      "learning_rate": 2.7659065890388336e-06,
      "loss": 0.78451979,
      "num_input_tokens_seen": 70567920,
      "step": 3274,
      "time_per_iteration": 2.5890285968780518
    },
    {
      "auxiliary_loss_clip": 0.01192472,
      "auxiliary_loss_mlp": 0.01029381,
      "balance_loss_clip": 1.05828536,
      "balance_loss_mlp": 1.02087533,
      "epoch": 0.3937954668430229,
      "flos": 16800161472000.0,
      "grad_norm": 1.9683055429915957,
      "language_loss": 0.85371697,
      "learning_rate": 2.7651869421118266e-06,
      "loss": 0.87593549,
      "num_input_tokens_seen": 70584530,
      "step": 3275,
      "time_per_iteration": 2.5888583660125732
    },
    {
      "auxiliary_loss_clip": 0.01193641,
      "auxiliary_loss_mlp": 0.01029592,
      "balance_loss_clip": 1.06159973,
      "balance_loss_mlp": 1.02087164,
      "epoch": 0.393915709733662,
      "flos": 21064229832960.0,
      "grad_norm": 1.7204499005043423,
      "language_loss": 0.83168852,
      "learning_rate": 2.76446717910657e-06,
      "loss": 0.85392082,
      "num_input_tokens_seen": 70605235,
      "step": 3276,
      "time_per_iteration": 2.6439337730407715
    },
    {
      "auxiliary_loss_clip": 0.011889,
      "auxiliary_loss_mlp": 0.01031623,
      "balance_loss_clip": 1.0558449,
      "balance_loss_mlp": 1.0227834,
      "epoch": 0.3940359526243011,
      "flos": 17165265264000.0,
      "grad_norm": 2.306597908392116,
      "language_loss": 0.76721418,
      "learning_rate": 2.763747300132249e-06,
      "loss": 0.78941941,
      "num_input_tokens_seen": 70622675,
      "step": 3277,
      "time_per_iteration": 2.6356008052825928
    },
    {
      "auxiliary_loss_clip": 0.01192285,
      "auxiliary_loss_mlp": 0.01024909,
      "balance_loss_clip": 1.05991817,
      "balance_loss_mlp": 1.01657367,
      "epoch": 0.3941561955149402,
      "flos": 20995856294400.0,
      "grad_norm": 1.7319240337153776,
      "language_loss": 0.86307585,
      "learning_rate": 2.7630273052980704e-06,
      "loss": 0.88524783,
      "num_input_tokens_seen": 70643265,
      "step": 3278,
      "time_per_iteration": 2.631371259689331
    },
    {
      "auxiliary_loss_clip": 0.01190518,
      "auxiliary_loss_mlp": 0.0103196,
      "balance_loss_clip": 1.05833471,
      "balance_loss_mlp": 1.02330565,
      "epoch": 0.39427643840557924,
      "flos": 18843406721280.0,
      "grad_norm": 2.102061657547151,
      "language_loss": 0.67108047,
      "learning_rate": 2.7623071947132554e-06,
      "loss": 0.69330525,
      "num_input_tokens_seen": 70660295,
      "step": 3279,
      "time_per_iteration": 3.6015067100524902
    },
    {
      "auxiliary_loss_clip": 0.01193537,
      "auxiliary_loss_mlp": 0.0102746,
      "balance_loss_clip": 1.05963814,
      "balance_loss_mlp": 1.01820946,
      "epoch": 0.39439668129621835,
      "flos": 23258659426560.0,
      "grad_norm": 2.051059021906272,
      "language_loss": 0.78619552,
      "learning_rate": 2.7615869684870458e-06,
      "loss": 0.80840546,
      "num_input_tokens_seen": 70679605,
      "step": 3280,
      "time_per_iteration": 3.610377311706543
    },
    {
      "auxiliary_loss_clip": 0.01193443,
      "auxiliary_loss_mlp": 0.01029336,
      "balance_loss_clip": 1.06002939,
      "balance_loss_mlp": 1.02072275,
      "epoch": 0.39451692418685746,
      "flos": 26652289507200.0,
      "grad_norm": 1.6865705915529667,
      "language_loss": 0.84749639,
      "learning_rate": 2.7608666267286986e-06,
      "loss": 0.86972415,
      "num_input_tokens_seen": 70699835,
      "step": 3281,
      "time_per_iteration": 3.550365447998047
    },
    {
      "auxiliary_loss_clip": 0.01192804,
      "auxiliary_loss_mlp": 0.01025929,
      "balance_loss_clip": 1.05683851,
      "balance_loss_mlp": 1.01621938,
      "epoch": 0.3946371670774965,
      "flos": 18258709132800.0,
      "grad_norm": 2.194072908159809,
      "language_loss": 0.86177921,
      "learning_rate": 2.760146169547489e-06,
      "loss": 0.88396657,
      "num_input_tokens_seen": 70716600,
      "step": 3282,
      "time_per_iteration": 2.5837109088897705
    },
    {
      "auxiliary_loss_clip": 0.01195312,
      "auxiliary_loss_mlp": 0.01030316,
      "balance_loss_clip": 1.06285906,
      "balance_loss_mlp": 1.02151275,
      "epoch": 0.39475740996813563,
      "flos": 24206126423040.0,
      "grad_norm": 1.4900089212992214,
      "language_loss": 0.76367497,
      "learning_rate": 2.75942559705271e-06,
      "loss": 0.78593135,
      "num_input_tokens_seen": 70736335,
      "step": 3283,
      "time_per_iteration": 2.6123292446136475
    },
    {
      "auxiliary_loss_clip": 0.01194124,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.0579654,
      "balance_loss_mlp": 1.02052736,
      "epoch": 0.39487765285877474,
      "flos": 19317858491520.0,
      "grad_norm": 1.934135818455664,
      "language_loss": 0.89057446,
      "learning_rate": 2.7587049093536713e-06,
      "loss": 0.91281521,
      "num_input_tokens_seen": 70752665,
      "step": 3284,
      "time_per_iteration": 2.6280338764190674
    },
    {
      "auxiliary_loss_clip": 0.01189315,
      "auxiliary_loss_mlp": 0.01023637,
      "balance_loss_clip": 1.05676615,
      "balance_loss_mlp": 1.01591825,
      "epoch": 0.3949978957494138,
      "flos": 17311744926720.0,
      "grad_norm": 2.356990675571194,
      "language_loss": 0.80700469,
      "learning_rate": 2.757984106559701e-06,
      "loss": 0.82913417,
      "num_input_tokens_seen": 70771650,
      "step": 3285,
      "time_per_iteration": 2.6004605293273926
    },
    {
      "auxiliary_loss_clip": 0.01191992,
      "auxiliary_loss_mlp": 0.01029726,
      "balance_loss_clip": 1.0606091,
      "balance_loss_mlp": 1.02137542,
      "epoch": 0.3951181386400529,
      "flos": 36317861280000.0,
      "grad_norm": 2.8953751694465475,
      "language_loss": 0.71504426,
      "learning_rate": 2.7572631887801446e-06,
      "loss": 0.73726141,
      "num_input_tokens_seen": 70793275,
      "step": 3286,
      "time_per_iteration": 2.719367504119873
    },
    {
      "auxiliary_loss_clip": 0.0119631,
      "auxiliary_loss_mlp": 0.01027427,
      "balance_loss_clip": 1.0600996,
      "balance_loss_mlp": 1.01905894,
      "epoch": 0.395238381530692,
      "flos": 23110348170240.0,
      "grad_norm": 2.636363061074751,
      "language_loss": 0.76794964,
      "learning_rate": 2.7565421561243654e-06,
      "loss": 0.790187,
      "num_input_tokens_seen": 70811440,
      "step": 3287,
      "time_per_iteration": 2.6232035160064697
    },
    {
      "auxiliary_loss_clip": 0.01189437,
      "auxiliary_loss_mlp": 0.01025284,
      "balance_loss_clip": 1.05711055,
      "balance_loss_mlp": 1.01712418,
      "epoch": 0.3953586244213311,
      "flos": 24347614095360.0,
      "grad_norm": 2.0867519596248227,
      "language_loss": 0.82731712,
      "learning_rate": 2.7558210087017413e-06,
      "loss": 0.84946436,
      "num_input_tokens_seen": 70831375,
      "step": 3288,
      "time_per_iteration": 2.6528255939483643
    },
    {
      "auxiliary_loss_clip": 0.01192401,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.05821586,
      "balance_loss_mlp": 1.0199846,
      "epoch": 0.3954788673119702,
      "flos": 23440080044160.0,
      "grad_norm": 2.0480693898450784,
      "language_loss": 0.73325139,
      "learning_rate": 2.7550997466216724e-06,
      "loss": 0.75547004,
      "num_input_tokens_seen": 70849170,
      "step": 3289,
      "time_per_iteration": 2.643242120742798
    },
    {
      "auxiliary_loss_clip": 0.01193175,
      "auxiliary_loss_mlp": 0.01034894,
      "balance_loss_clip": 1.05967772,
      "balance_loss_mlp": 1.02618575,
      "epoch": 0.3955991102026093,
      "flos": 17494063384320.0,
      "grad_norm": 1.9888061440564095,
      "language_loss": 0.81324977,
      "learning_rate": 2.7543783699935714e-06,
      "loss": 0.8355304,
      "num_input_tokens_seen": 70867200,
      "step": 3290,
      "time_per_iteration": 2.5721209049224854
    },
    {
      "auxiliary_loss_clip": 0.0119457,
      "auxiliary_loss_mlp": 0.01028179,
      "balance_loss_clip": 1.06085324,
      "balance_loss_mlp": 1.01922643,
      "epoch": 0.39571935309324835,
      "flos": 18221326053120.0,
      "grad_norm": 2.5493689301788103,
      "language_loss": 0.86162686,
      "learning_rate": 2.753656878926872e-06,
      "loss": 0.88385439,
      "num_input_tokens_seen": 70883080,
      "step": 3291,
      "time_per_iteration": 2.580909490585327
    },
    {
      "auxiliary_loss_clip": 0.01189834,
      "auxiliary_loss_mlp": 0.01024712,
      "balance_loss_clip": 1.05575478,
      "balance_loss_mlp": 1.0164448,
      "epoch": 0.39583959598388746,
      "flos": 17748813617280.0,
      "grad_norm": 2.0986650925204864,
      "language_loss": 0.74078822,
      "learning_rate": 2.752935273531023e-06,
      "loss": 0.76293373,
      "num_input_tokens_seen": 70901230,
      "step": 3292,
      "time_per_iteration": 2.6363449096679688
    },
    {
      "auxiliary_loss_clip": 0.01196244,
      "auxiliary_loss_mlp": 0.01033697,
      "balance_loss_clip": 1.06031609,
      "balance_loss_mlp": 1.02395749,
      "epoch": 0.39595983887452657,
      "flos": 19352368483200.0,
      "grad_norm": 1.7988855289722814,
      "language_loss": 0.78532255,
      "learning_rate": 2.752213553915492e-06,
      "loss": 0.80762196,
      "num_input_tokens_seen": 70919585,
      "step": 3293,
      "time_per_iteration": 2.6241679191589355
    },
    {
      "auxiliary_loss_clip": 0.01086646,
      "auxiliary_loss_mlp": 0.01004467,
      "balance_loss_clip": 1.01896143,
      "balance_loss_mlp": 1.00317311,
      "epoch": 0.3960800817651656,
      "flos": 60682282940160.0,
      "grad_norm": 0.8129510585385032,
      "language_loss": 0.66027951,
      "learning_rate": 2.751491720189762e-06,
      "loss": 0.68119073,
      "num_input_tokens_seen": 70977695,
      "step": 3294,
      "time_per_iteration": 3.1524181365966797
    },
    {
      "auxiliary_loss_clip": 0.01190927,
      "auxiliary_loss_mlp": 0.0103175,
      "balance_loss_clip": 1.05851865,
      "balance_loss_mlp": 1.02321434,
      "epoch": 0.39620032465580474,
      "flos": 16836718538880.0,
      "grad_norm": 2.4731749671533887,
      "language_loss": 0.91661704,
      "learning_rate": 2.7507697724633364e-06,
      "loss": 0.93884385,
      "num_input_tokens_seen": 70994455,
      "step": 3295,
      "time_per_iteration": 2.705322027206421
    },
    {
      "auxiliary_loss_clip": 0.0108652,
      "auxiliary_loss_mlp": 0.01002301,
      "balance_loss_clip": 1.01887846,
      "balance_loss_mlp": 1.00106716,
      "epoch": 0.3963205675464438,
      "flos": 69071445941760.0,
      "grad_norm": 0.8601529981176829,
      "language_loss": 0.54615438,
      "learning_rate": 2.7500477108457327e-06,
      "loss": 0.56704259,
      "num_input_tokens_seen": 71046465,
      "step": 3296,
      "time_per_iteration": 3.005817413330078
    },
    {
      "auxiliary_loss_clip": 0.01193322,
      "auxiliary_loss_mlp": 0.01027377,
      "balance_loss_clip": 1.05862892,
      "balance_loss_mlp": 1.01904392,
      "epoch": 0.3964408104370829,
      "flos": 25667439431040.0,
      "grad_norm": 2.593915629844651,
      "language_loss": 0.80747283,
      "learning_rate": 2.7493255354464877e-06,
      "loss": 0.82967985,
      "num_input_tokens_seen": 71064275,
      "step": 3297,
      "time_per_iteration": 2.650820732116699
    },
    {
      "auxiliary_loss_clip": 0.01192872,
      "auxiliary_loss_mlp": 0.0103423,
      "balance_loss_clip": 1.05870891,
      "balance_loss_mlp": 1.02558708,
      "epoch": 0.396561053327722,
      "flos": 24277480790400.0,
      "grad_norm": 2.187031925002843,
      "language_loss": 0.7611264,
      "learning_rate": 2.748603246375156e-06,
      "loss": 0.78339744,
      "num_input_tokens_seen": 71082290,
      "step": 3298,
      "time_per_iteration": 2.651885747909546
    },
    {
      "auxiliary_loss_clip": 0.01195604,
      "auxiliary_loss_mlp": 0.01034847,
      "balance_loss_clip": 1.06112444,
      "balance_loss_mlp": 1.02556705,
      "epoch": 0.39668129621836107,
      "flos": 20522302364160.0,
      "grad_norm": 2.2313848063434314,
      "language_loss": 0.69805348,
      "learning_rate": 2.7478808437413055e-06,
      "loss": 0.72035801,
      "num_input_tokens_seen": 71101700,
      "step": 3299,
      "time_per_iteration": 2.5875566005706787
    },
    {
      "auxiliary_loss_clip": 0.01193598,
      "auxiliary_loss_mlp": 0.01036009,
      "balance_loss_clip": 1.05855119,
      "balance_loss_mlp": 1.02717543,
      "epoch": 0.3968015391090002,
      "flos": 27052585649280.0,
      "grad_norm": 2.2044077051467004,
      "language_loss": 0.65901172,
      "learning_rate": 2.7471583276545263e-06,
      "loss": 0.68130779,
      "num_input_tokens_seen": 71122360,
      "step": 3300,
      "time_per_iteration": 2.655041456222534
    },
    {
      "auxiliary_loss_clip": 0.01190845,
      "auxiliary_loss_mlp": 0.01033579,
      "balance_loss_clip": 1.05577648,
      "balance_loss_mlp": 1.02485907,
      "epoch": 0.3969217819996393,
      "flos": 12531819392640.0,
      "grad_norm": 2.534580110968834,
      "language_loss": 0.70559764,
      "learning_rate": 2.7464356982244224e-06,
      "loss": 0.72784185,
      "num_input_tokens_seen": 71140360,
      "step": 3301,
      "time_per_iteration": 2.5839223861694336
    },
    {
      "auxiliary_loss_clip": 0.01084837,
      "auxiliary_loss_mlp": 0.01001394,
      "balance_loss_clip": 1.01747501,
      "balance_loss_mlp": 1.00014865,
      "epoch": 0.39704202489027834,
      "flos": 66241399230720.0,
      "grad_norm": 0.7955913519568305,
      "language_loss": 0.61794531,
      "learning_rate": 2.745712955560617e-06,
      "loss": 0.63880765,
      "num_input_tokens_seen": 71196565,
      "step": 3302,
      "time_per_iteration": 3.1448652744293213
    },
    {
      "auxiliary_loss_clip": 0.01196293,
      "auxiliary_loss_mlp": 0.01035634,
      "balance_loss_clip": 1.06195319,
      "balance_loss_mlp": 1.02650833,
      "epoch": 0.39716226778091746,
      "flos": 16982982720000.0,
      "grad_norm": 2.426315393761732,
      "language_loss": 0.76908338,
      "learning_rate": 2.7449900997727496e-06,
      "loss": 0.79140258,
      "num_input_tokens_seen": 71214675,
      "step": 3303,
      "time_per_iteration": 2.5860722064971924
    },
    {
      "auxiliary_loss_clip": 0.01192394,
      "auxiliary_loss_mlp": 0.01032534,
      "balance_loss_clip": 1.0603776,
      "balance_loss_mlp": 1.02438569,
      "epoch": 0.39728251067155657,
      "flos": 23477139901440.0,
      "grad_norm": 1.8013439340772717,
      "language_loss": 0.83865243,
      "learning_rate": 2.744267130970476e-06,
      "loss": 0.86090171,
      "num_input_tokens_seen": 71234400,
      "step": 3304,
      "time_per_iteration": 2.618253707885742
    },
    {
      "auxiliary_loss_clip": 0.0119552,
      "auxiliary_loss_mlp": 0.01035438,
      "balance_loss_clip": 1.05959177,
      "balance_loss_mlp": 1.02624083,
      "epoch": 0.3974027535621956,
      "flos": 20704441253760.0,
      "grad_norm": 1.7983182832126356,
      "language_loss": 0.76842964,
      "learning_rate": 2.7435440492634697e-06,
      "loss": 0.79073924,
      "num_input_tokens_seen": 71253725,
      "step": 3305,
      "time_per_iteration": 2.548621416091919
    },
    {
      "auxiliary_loss_clip": 0.01196778,
      "auxiliary_loss_mlp": 0.01031275,
      "balance_loss_clip": 1.05725682,
      "balance_loss_mlp": 1.02034926,
      "epoch": 0.39752299645283473,
      "flos": 21543278544000.0,
      "grad_norm": 2.041262551889157,
      "language_loss": 0.67459357,
      "learning_rate": 2.7428208547614228e-06,
      "loss": 0.69687414,
      "num_input_tokens_seen": 71273220,
      "step": 3306,
      "time_per_iteration": 4.513635873794556
    },
    {
      "auxiliary_loss_clip": 0.01196703,
      "auxiliary_loss_mlp": 0.01036963,
      "balance_loss_clip": 1.06150424,
      "balance_loss_mlp": 1.02833247,
      "epoch": 0.39764323934347384,
      "flos": 19208295031680.0,
      "grad_norm": 2.591305651680353,
      "language_loss": 0.77079272,
      "learning_rate": 2.742097547574043e-06,
      "loss": 0.79312938,
      "num_input_tokens_seen": 71291445,
      "step": 3307,
      "time_per_iteration": 3.5965471267700195
    },
    {
      "auxiliary_loss_clip": 0.01192824,
      "auxiliary_loss_mlp": 0.0102839,
      "balance_loss_clip": 1.05755472,
      "balance_loss_mlp": 1.01882327,
      "epoch": 0.3977634822341129,
      "flos": 20850202644480.0,
      "grad_norm": 1.9099137973847296,
      "language_loss": 0.78055364,
      "learning_rate": 2.7413741278110544e-06,
      "loss": 0.80276579,
      "num_input_tokens_seen": 71310135,
      "step": 3308,
      "time_per_iteration": 3.5449209213256836
    },
    {
      "auxiliary_loss_clip": 0.01196447,
      "auxiliary_loss_mlp": 0.01038614,
      "balance_loss_clip": 1.06083596,
      "balance_loss_mlp": 1.02910709,
      "epoch": 0.397883725124752,
      "flos": 39786042038400.0,
      "grad_norm": 2.279738126112645,
      "language_loss": 0.68844318,
      "learning_rate": 2.7406505955822016e-06,
      "loss": 0.71079385,
      "num_input_tokens_seen": 71331160,
      "step": 3309,
      "time_per_iteration": 2.7221877574920654
    },
    {
      "auxiliary_loss_clip": 0.01194905,
      "auxiliary_loss_mlp": 0.0103117,
      "balance_loss_clip": 1.05974185,
      "balance_loss_mlp": 1.02180648,
      "epoch": 0.39800396801539106,
      "flos": 17379507934080.0,
      "grad_norm": 2.8995842650949757,
      "language_loss": 0.66258419,
      "learning_rate": 2.7399269509972415e-06,
      "loss": 0.68484503,
      "num_input_tokens_seen": 71345315,
      "step": 3310,
      "time_per_iteration": 2.583604574203491
    },
    {
      "auxiliary_loss_clip": 0.01198314,
      "auxiliary_loss_mlp": 0.01032944,
      "balance_loss_clip": 1.05911183,
      "balance_loss_mlp": 1.02265644,
      "epoch": 0.3981242109060302,
      "flos": 19202764337280.0,
      "grad_norm": 3.07643285531222,
      "language_loss": 0.84849393,
      "learning_rate": 2.7392031941659514e-06,
      "loss": 0.87080646,
      "num_input_tokens_seen": 71363160,
      "step": 3311,
      "time_per_iteration": 2.577305316925049
    },
    {
      "auxiliary_loss_clip": 0.011974,
      "auxiliary_loss_mlp": 0.01030188,
      "balance_loss_clip": 1.06184185,
      "balance_loss_mlp": 1.0213306,
      "epoch": 0.3982444537966693,
      "flos": 24565124903040.0,
      "grad_norm": 2.2987741980553773,
      "language_loss": 0.86299193,
      "learning_rate": 2.7384793251981244e-06,
      "loss": 0.88526785,
      "num_input_tokens_seen": 71382145,
      "step": 3312,
      "time_per_iteration": 2.6545557975769043
    },
    {
      "auxiliary_loss_clip": 0.0119623,
      "auxiliary_loss_mlp": 0.01032687,
      "balance_loss_clip": 1.06016684,
      "balance_loss_mlp": 1.02402008,
      "epoch": 0.39836469668730834,
      "flos": 26213856099840.0,
      "grad_norm": 2.10524606709844,
      "language_loss": 0.80993813,
      "learning_rate": 2.737755344203571e-06,
      "loss": 0.83222729,
      "num_input_tokens_seen": 71402095,
      "step": 3313,
      "time_per_iteration": 2.6354317665100098
    },
    {
      "auxiliary_loss_clip": 0.01193945,
      "auxiliary_loss_mlp": 0.01030725,
      "balance_loss_clip": 1.06047678,
      "balance_loss_mlp": 1.02204108,
      "epoch": 0.39848493957794745,
      "flos": 27636134002560.0,
      "grad_norm": 2.112515350201231,
      "language_loss": 0.79949522,
      "learning_rate": 2.7370312512921186e-06,
      "loss": 0.82174194,
      "num_input_tokens_seen": 71423875,
      "step": 3314,
      "time_per_iteration": 2.669482946395874
    },
    {
      "auxiliary_loss_clip": 0.01196215,
      "auxiliary_loss_mlp": 0.01035745,
      "balance_loss_clip": 1.05935597,
      "balance_loss_mlp": 1.02616692,
      "epoch": 0.39860518246858656,
      "flos": 12239326944000.0,
      "grad_norm": 3.7412785431765263,
      "language_loss": 0.76647443,
      "learning_rate": 2.736307046573611e-06,
      "loss": 0.7887941,
      "num_input_tokens_seen": 71439745,
      "step": 3315,
      "time_per_iteration": 2.575464963912964
    },
    {
      "auxiliary_loss_clip": 0.01190577,
      "auxiliary_loss_mlp": 0.01027426,
      "balance_loss_clip": 1.0570972,
      "balance_loss_mlp": 1.01899767,
      "epoch": 0.3987254253592256,
      "flos": 22379135005440.0,
      "grad_norm": 1.9238852763106655,
      "language_loss": 0.81809545,
      "learning_rate": 2.73558273015791e-06,
      "loss": 0.84027541,
      "num_input_tokens_seen": 71459575,
      "step": 3316,
      "time_per_iteration": 2.6394519805908203
    },
    {
      "auxiliary_loss_clip": 0.01200087,
      "auxiliary_loss_mlp": 0.01033964,
      "balance_loss_clip": 1.06290913,
      "balance_loss_mlp": 1.02415335,
      "epoch": 0.3988456682498647,
      "flos": 23514020190720.0,
      "grad_norm": 2.269696712980742,
      "language_loss": 0.70258248,
      "learning_rate": 2.734858302154894e-06,
      "loss": 0.72492296,
      "num_input_tokens_seen": 71481075,
      "step": 3317,
      "time_per_iteration": 2.628382921218872
    },
    {
      "auxiliary_loss_clip": 0.01192124,
      "auxiliary_loss_mlp": 0.01032381,
      "balance_loss_clip": 1.05687594,
      "balance_loss_mlp": 1.02354193,
      "epoch": 0.39896591114050384,
      "flos": 19208761908480.0,
      "grad_norm": 1.9468601284551936,
      "language_loss": 0.76156086,
      "learning_rate": 2.734133762674457e-06,
      "loss": 0.78380591,
      "num_input_tokens_seen": 71500665,
      "step": 3318,
      "time_per_iteration": 2.619288206100464
    },
    {
      "auxiliary_loss_clip": 0.01196937,
      "auxiliary_loss_mlp": 0.01027149,
      "balance_loss_clip": 1.06014681,
      "balance_loss_mlp": 1.01746285,
      "epoch": 0.3990861540311429,
      "flos": 28401031146240.0,
      "grad_norm": 2.439915338594767,
      "language_loss": 0.70861208,
      "learning_rate": 2.7334091118265124e-06,
      "loss": 0.73085296,
      "num_input_tokens_seen": 71522560,
      "step": 3319,
      "time_per_iteration": 2.6574199199676514
    },
    {
      "auxiliary_loss_clip": 0.01082898,
      "auxiliary_loss_mlp": 0.01001875,
      "balance_loss_clip": 1.01581693,
      "balance_loss_mlp": 1.00052214,
      "epoch": 0.399206396921782,
      "flos": 61758563086080.0,
      "grad_norm": 0.6765745134811585,
      "language_loss": 0.57804912,
      "learning_rate": 2.732684349720989e-06,
      "loss": 0.59889686,
      "num_input_tokens_seen": 71590520,
      "step": 3320,
      "time_per_iteration": 3.1702678203582764
    },
    {
      "auxiliary_loss_clip": 0.0119864,
      "auxiliary_loss_mlp": 0.01034287,
      "balance_loss_clip": 1.06160998,
      "balance_loss_mlp": 1.02522659,
      "epoch": 0.3993266398124211,
      "flos": 28074567409920.0,
      "grad_norm": 2.050693335722656,
      "language_loss": 0.75581199,
      "learning_rate": 2.7319594764678318e-06,
      "loss": 0.77814126,
      "num_input_tokens_seen": 71612620,
      "step": 3321,
      "time_per_iteration": 2.6608800888061523
    },
    {
      "auxiliary_loss_clip": 0.01194692,
      "auxiliary_loss_mlp": 0.01027646,
      "balance_loss_clip": 1.05803239,
      "balance_loss_mlp": 1.01798391,
      "epoch": 0.39944688270306017,
      "flos": 23225083188480.0,
      "grad_norm": 1.6698342414258631,
      "language_loss": 0.83166754,
      "learning_rate": 2.7312344921770044e-06,
      "loss": 0.8538909,
      "num_input_tokens_seen": 71634320,
      "step": 3322,
      "time_per_iteration": 2.6890335083007812
    },
    {
      "auxiliary_loss_clip": 0.01192453,
      "auxiliary_loss_mlp": 0.01031797,
      "balance_loss_clip": 1.05841398,
      "balance_loss_mlp": 1.02286196,
      "epoch": 0.3995671255936993,
      "flos": 19390433921280.0,
      "grad_norm": 1.9033229440390171,
      "language_loss": 0.78376132,
      "learning_rate": 2.7305093969584857e-06,
      "loss": 0.80600381,
      "num_input_tokens_seen": 71653145,
      "step": 3323,
      "time_per_iteration": 2.627779483795166
    },
    {
      "auxiliary_loss_clip": 0.01193274,
      "auxiliary_loss_mlp": 0.01033927,
      "balance_loss_clip": 1.05869853,
      "balance_loss_mlp": 1.02496219,
      "epoch": 0.3996873684843384,
      "flos": 23842638743040.0,
      "grad_norm": 2.7475995842224887,
      "language_loss": 0.7995019,
      "learning_rate": 2.729784190922272e-06,
      "loss": 0.82177383,
      "num_input_tokens_seen": 71674580,
      "step": 3324,
      "time_per_iteration": 2.630448341369629
    },
    {
      "auxiliary_loss_clip": 0.01082028,
      "auxiliary_loss_mlp": 0.01001163,
      "balance_loss_clip": 1.01510549,
      "balance_loss_mlp": 0.99982178,
      "epoch": 0.39980761137497745,
      "flos": 66576877280640.0,
      "grad_norm": 0.9434495808545748,
      "language_loss": 0.57259029,
      "learning_rate": 2.729058874178378e-06,
      "loss": 0.59342217,
      "num_input_tokens_seen": 71745260,
      "step": 3325,
      "time_per_iteration": 3.2394442558288574
    },
    {
      "auxiliary_loss_clip": 0.01196909,
      "auxiliary_loss_mlp": 0.01029288,
      "balance_loss_clip": 1.06031585,
      "balance_loss_mlp": 1.01889944,
      "epoch": 0.39992785426561656,
      "flos": 28549162834560.0,
      "grad_norm": 2.413033119946212,
      "language_loss": 0.69206637,
      "learning_rate": 2.7283334468368315e-06,
      "loss": 0.71432835,
      "num_input_tokens_seen": 71766540,
      "step": 3326,
      "time_per_iteration": 2.6712357997894287
    },
    {
      "auxiliary_loss_clip": 0.01195908,
      "auxiliary_loss_mlp": 0.01030965,
      "balance_loss_clip": 1.05906987,
      "balance_loss_mlp": 1.02202439,
      "epoch": 0.4000480971562556,
      "flos": 15049408671360.0,
      "grad_norm": 2.0461161775587895,
      "language_loss": 0.73101854,
      "learning_rate": 2.72760790900768e-06,
      "loss": 0.75328732,
      "num_input_tokens_seen": 71783125,
      "step": 3327,
      "time_per_iteration": 2.6058168411254883
    },
    {
      "auxiliary_loss_clip": 0.0119472,
      "auxiliary_loss_mlp": 0.01034463,
      "balance_loss_clip": 1.05944502,
      "balance_loss_mlp": 1.02518272,
      "epoch": 0.4001683400468947,
      "flos": 23915609222400.0,
      "grad_norm": 1.7105525021957415,
      "language_loss": 0.78399992,
      "learning_rate": 2.7268822608009875e-06,
      "loss": 0.8062917,
      "num_input_tokens_seen": 71802500,
      "step": 3328,
      "time_per_iteration": 2.5957555770874023
    },
    {
      "auxiliary_loss_clip": 0.01192827,
      "auxiliary_loss_mlp": 0.01027402,
      "balance_loss_clip": 1.0572896,
      "balance_loss_mlp": 1.01774001,
      "epoch": 0.40028858293753383,
      "flos": 24352677912960.0,
      "grad_norm": 2.177707850195025,
      "language_loss": 0.78510523,
      "learning_rate": 2.726156502326834e-06,
      "loss": 0.80730754,
      "num_input_tokens_seen": 71823800,
      "step": 3329,
      "time_per_iteration": 2.7252001762390137
    },
    {
      "auxiliary_loss_clip": 0.01080467,
      "auxiliary_loss_mlp": 0.0100126,
      "balance_loss_clip": 1.01364613,
      "balance_loss_mlp": 0.99997276,
      "epoch": 0.4004088258281729,
      "flos": 66787025800320.0,
      "grad_norm": 0.6974435937130143,
      "language_loss": 0.60231209,
      "learning_rate": 2.725430633695316e-06,
      "loss": 0.62312937,
      "num_input_tokens_seen": 71886880,
      "step": 3330,
      "time_per_iteration": 3.1818301677703857
    },
    {
      "auxiliary_loss_clip": 0.01079416,
      "auxiliary_loss_mlp": 0.01000965,
      "balance_loss_clip": 1.01255226,
      "balance_loss_mlp": 0.99970704,
      "epoch": 0.400529068718812,
      "flos": 58598386473600.0,
      "grad_norm": 0.8952839163169917,
      "language_loss": 0.5795601,
      "learning_rate": 2.7247046550165485e-06,
      "loss": 0.60036391,
      "num_input_tokens_seen": 71939005,
      "step": 3331,
      "time_per_iteration": 4.062800884246826
    },
    {
      "auxiliary_loss_clip": 0.0119675,
      "auxiliary_loss_mlp": 0.01036379,
      "balance_loss_clip": 1.06123161,
      "balance_loss_mlp": 1.02753413,
      "epoch": 0.4006493116094511,
      "flos": 25377460934400.0,
      "grad_norm": 1.56684174664597,
      "language_loss": 0.76085359,
      "learning_rate": 2.7239785664006606e-06,
      "loss": 0.78318489,
      "num_input_tokens_seen": 71962545,
      "step": 3332,
      "time_per_iteration": 3.6561543941497803
    },
    {
      "auxiliary_loss_clip": 0.01079035,
      "auxiliary_loss_mlp": 0.01002708,
      "balance_loss_clip": 1.01249599,
      "balance_loss_mlp": 1.00142086,
      "epoch": 0.40076955450009016,
      "flos": 60280729822080.0,
      "grad_norm": 0.7744566908521522,
      "language_loss": 0.61744225,
      "learning_rate": 2.7232523679578002e-06,
      "loss": 0.63825977,
      "num_input_tokens_seen": 72025625,
      "step": 3333,
      "time_per_iteration": 4.118993043899536
    },
    {
      "auxiliary_loss_clip": 0.01192299,
      "auxiliary_loss_mlp": 0.01033863,
      "balance_loss_clip": 1.05723357,
      "balance_loss_mlp": 1.02458286,
      "epoch": 0.4008897973907293,
      "flos": 16617268396800.0,
      "grad_norm": 2.2580836707730976,
      "language_loss": 0.79646111,
      "learning_rate": 2.7225260597981295e-06,
      "loss": 0.81872272,
      "num_input_tokens_seen": 72043330,
      "step": 3334,
      "time_per_iteration": 2.6030495166778564
    },
    {
      "auxiliary_loss_clip": 0.01193262,
      "auxiliary_loss_mlp": 0.01030232,
      "balance_loss_clip": 1.05548072,
      "balance_loss_mlp": 1.02061796,
      "epoch": 0.4010100402813684,
      "flos": 15377344865280.0,
      "grad_norm": 2.6527135724619995,
      "language_loss": 0.78555548,
      "learning_rate": 2.721799642031831e-06,
      "loss": 0.8077904,
      "num_input_tokens_seen": 72059500,
      "step": 3335,
      "time_per_iteration": 2.5817365646362305
    },
    {
      "auxiliary_loss_clip": 0.01195974,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.05934823,
      "balance_loss_mlp": 1.01918197,
      "epoch": 0.40113028317200744,
      "flos": 13298835438720.0,
      "grad_norm": 4.008126317302368,
      "language_loss": 0.776371,
      "learning_rate": 2.721073114769101e-06,
      "loss": 0.79861951,
      "num_input_tokens_seen": 72077175,
      "step": 3336,
      "time_per_iteration": 2.63877272605896
    },
    {
      "auxiliary_loss_clip": 0.01189773,
      "auxiliary_loss_mlp": 0.01027353,
      "balance_loss_clip": 1.05582762,
      "balance_loss_mlp": 1.01812601,
      "epoch": 0.40125052606264655,
      "flos": 20668027841280.0,
      "grad_norm": 2.1813297318486082,
      "language_loss": 0.75365037,
      "learning_rate": 2.7203464781201523e-06,
      "loss": 0.77582163,
      "num_input_tokens_seen": 72096490,
      "step": 3337,
      "time_per_iteration": 2.624777317047119
    },
    {
      "auxiliary_loss_clip": 0.01195328,
      "auxiliary_loss_mlp": 0.01033051,
      "balance_loss_clip": 1.06019592,
      "balance_loss_mlp": 1.02335954,
      "epoch": 0.40137076895328566,
      "flos": 24607679541120.0,
      "grad_norm": 1.839602169934323,
      "language_loss": 0.77946621,
      "learning_rate": 2.719619732195215e-06,
      "loss": 0.80174994,
      "num_input_tokens_seen": 72118130,
      "step": 3338,
      "time_per_iteration": 2.6827571392059326
    },
    {
      "auxiliary_loss_clip": 0.01192254,
      "auxiliary_loss_mlp": 0.01030606,
      "balance_loss_clip": 1.05631208,
      "balance_loss_mlp": 1.02127206,
      "epoch": 0.4014910118439247,
      "flos": 24206593299840.0,
      "grad_norm": 1.6367666802862064,
      "language_loss": 0.73049486,
      "learning_rate": 2.7188928771045377e-06,
      "loss": 0.75272346,
      "num_input_tokens_seen": 72139450,
      "step": 3339,
      "time_per_iteration": 2.6265454292297363
    },
    {
      "auxiliary_loss_clip": 0.01192888,
      "auxiliary_loss_mlp": 0.01031252,
      "balance_loss_clip": 1.05773008,
      "balance_loss_mlp": 1.02201879,
      "epoch": 0.4016112547345638,
      "flos": 26725080418560.0,
      "grad_norm": 1.9142162541961971,
      "language_loss": 0.8031255,
      "learning_rate": 2.7181659129583815e-06,
      "loss": 0.82536685,
      "num_input_tokens_seen": 72159040,
      "step": 3340,
      "time_per_iteration": 2.6812849044799805
    },
    {
      "auxiliary_loss_clip": 0.01190012,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.05496597,
      "balance_loss_mlp": 1.01963341,
      "epoch": 0.4017314976252029,
      "flos": 21288025520640.0,
      "grad_norm": 1.7729448037865074,
      "language_loss": 0.7601189,
      "learning_rate": 2.7174388398670276e-06,
      "loss": 0.78230691,
      "num_input_tokens_seen": 72178220,
      "step": 3341,
      "time_per_iteration": 2.5692198276519775
    },
    {
      "auxiliary_loss_clip": 0.01191628,
      "auxiliary_loss_mlp": 0.01034417,
      "balance_loss_clip": 1.05404353,
      "balance_loss_mlp": 1.02474356,
      "epoch": 0.401851740515842,
      "flos": 25484690010240.0,
      "grad_norm": 5.045562129390404,
      "language_loss": 0.92143416,
      "learning_rate": 2.716711657940773e-06,
      "loss": 0.94369459,
      "num_input_tokens_seen": 72199230,
      "step": 3342,
      "time_per_iteration": 2.680769205093384
    },
    {
      "auxiliary_loss_clip": 0.01077025,
      "auxiliary_loss_mlp": 0.01002031,
      "balance_loss_clip": 1.01071954,
      "balance_loss_mlp": 1.00076187,
      "epoch": 0.4019719834064811,
      "flos": 55395334978560.0,
      "grad_norm": 0.8114497965573392,
      "language_loss": 0.56490695,
      "learning_rate": 2.7159843672899284e-06,
      "loss": 0.58569753,
      "num_input_tokens_seen": 72263430,
      "step": 3343,
      "time_per_iteration": 3.293067455291748
    },
    {
      "auxiliary_loss_clip": 0.01193354,
      "auxiliary_loss_mlp": 0.01032595,
      "balance_loss_clip": 1.05824304,
      "balance_loss_mlp": 1.02326632,
      "epoch": 0.40209222629712016,
      "flos": 18180100218240.0,
      "grad_norm": 1.809565710482191,
      "language_loss": 0.81532967,
      "learning_rate": 2.715256968024825e-06,
      "loss": 0.83758914,
      "num_input_tokens_seen": 72280505,
      "step": 3344,
      "time_per_iteration": 2.6118078231811523
    },
    {
      "auxiliary_loss_clip": 0.01193505,
      "auxiliary_loss_mlp": 0.01033771,
      "balance_loss_clip": 1.05811572,
      "balance_loss_mlp": 1.02423465,
      "epoch": 0.40221246918775927,
      "flos": 25961009287680.0,
      "grad_norm": 1.5290874560617853,
      "language_loss": 0.82422471,
      "learning_rate": 2.7145294602558083e-06,
      "loss": 0.84649754,
      "num_input_tokens_seen": 72301215,
      "step": 3345,
      "time_per_iteration": 2.6993870735168457
    },
    {
      "auxiliary_loss_clip": 0.01194864,
      "auxiliary_loss_mlp": 0.01030327,
      "balance_loss_clip": 1.05868411,
      "balance_loss_mlp": 1.0207665,
      "epoch": 0.4023327120783984,
      "flos": 33838912056960.0,
      "grad_norm": 1.8772186916167806,
      "language_loss": 0.70917165,
      "learning_rate": 2.713801844093241e-06,
      "loss": 0.73142356,
      "num_input_tokens_seen": 72322365,
      "step": 3346,
      "time_per_iteration": 2.7147464752197266
    },
    {
      "auxiliary_loss_clip": 0.01192408,
      "auxiliary_loss_mlp": 0.01033115,
      "balance_loss_clip": 1.05780411,
      "balance_loss_mlp": 1.02362037,
      "epoch": 0.40245295496903744,
      "flos": 26900252069760.0,
      "grad_norm": 1.872473536142306,
      "language_loss": 0.88379133,
      "learning_rate": 2.7130741196475014e-06,
      "loss": 0.90604657,
      "num_input_tokens_seen": 72340495,
      "step": 3347,
      "time_per_iteration": 2.6888065338134766
    },
    {
      "auxiliary_loss_clip": 0.01200782,
      "auxiliary_loss_mlp": 0.01031828,
      "balance_loss_clip": 1.06044888,
      "balance_loss_mlp": 1.02192807,
      "epoch": 0.40257319785967655,
      "flos": 36902738436480.0,
      "grad_norm": 1.9656159686941077,
      "language_loss": 0.79678822,
      "learning_rate": 2.7123462870289848e-06,
      "loss": 0.81911433,
      "num_input_tokens_seen": 72360545,
      "step": 3348,
      "time_per_iteration": 2.741168260574341
    },
    {
      "auxiliary_loss_clip": 0.01192947,
      "auxiliary_loss_mlp": 0.01028791,
      "balance_loss_clip": 1.05600643,
      "balance_loss_mlp": 1.01984382,
      "epoch": 0.40269344075031566,
      "flos": 24353180703360.0,
      "grad_norm": 2.0328815746207063,
      "language_loss": 0.81413376,
      "learning_rate": 2.711618346348102e-06,
      "loss": 0.8363511,
      "num_input_tokens_seen": 72381070,
      "step": 3349,
      "time_per_iteration": 2.719801425933838
    },
    {
      "auxiliary_loss_clip": 0.01192774,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.05900717,
      "balance_loss_mlp": 1.01808071,
      "epoch": 0.4028136836409547,
      "flos": 14389657614720.0,
      "grad_norm": 2.0133653496023687,
      "language_loss": 0.63657719,
      "learning_rate": 2.7108902977152825e-06,
      "loss": 0.65877885,
      "num_input_tokens_seen": 72398970,
      "step": 3350,
      "time_per_iteration": 2.5871834754943848
    },
    {
      "auxiliary_loss_clip": 0.01193068,
      "auxiliary_loss_mlp": 0.01036895,
      "balance_loss_clip": 1.05828905,
      "balance_loss_mlp": 1.02772164,
      "epoch": 0.4029339265315938,
      "flos": 26136037284480.0,
      "grad_norm": 2.159600462471045,
      "language_loss": 0.75221491,
      "learning_rate": 2.7101621412409704e-06,
      "loss": 0.77451456,
      "num_input_tokens_seen": 72418455,
      "step": 3351,
      "time_per_iteration": 2.685938835144043
    },
    {
      "auxiliary_loss_clip": 0.01194366,
      "auxiliary_loss_mlp": 0.01036427,
      "balance_loss_clip": 1.05771899,
      "balance_loss_mlp": 1.02694392,
      "epoch": 0.40305416942223293,
      "flos": 23256325042560.0,
      "grad_norm": 2.028307088732297,
      "language_loss": 0.86036998,
      "learning_rate": 2.7094338770356256e-06,
      "loss": 0.88267791,
      "num_input_tokens_seen": 72437540,
      "step": 3352,
      "time_per_iteration": 2.6307120323181152
    },
    {
      "auxiliary_loss_clip": 0.01194343,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.05840945,
      "balance_loss_mlp": 1.01691639,
      "epoch": 0.403174412312872,
      "flos": 27089645506560.0,
      "grad_norm": 1.9792265292353237,
      "language_loss": 0.6421988,
      "learning_rate": 2.708705505209726e-06,
      "loss": 0.66441041,
      "num_input_tokens_seen": 72458315,
      "step": 3353,
      "time_per_iteration": 2.752859354019165
    },
    {
      "auxiliary_loss_clip": 0.01193437,
      "auxiliary_loss_mlp": 0.01032055,
      "balance_loss_clip": 1.059394,
      "balance_loss_mlp": 1.02282786,
      "epoch": 0.4032946552035111,
      "flos": 21756336065280.0,
      "grad_norm": 2.6920142523791264,
      "language_loss": 0.91779953,
      "learning_rate": 2.7079770258737646e-06,
      "loss": 0.94005442,
      "num_input_tokens_seen": 72476225,
      "step": 3354,
      "time_per_iteration": 2.6928958892822266
    },
    {
      "auxiliary_loss_clip": 0.01196814,
      "auxiliary_loss_mlp": 0.01030656,
      "balance_loss_clip": 1.05893993,
      "balance_loss_mlp": 1.02029681,
      "epoch": 0.4034148980941502,
      "flos": 17343956448000.0,
      "grad_norm": 2.885522899716653,
      "language_loss": 0.75348735,
      "learning_rate": 2.707248439138251e-06,
      "loss": 0.77576208,
      "num_input_tokens_seen": 72492460,
      "step": 3355,
      "time_per_iteration": 2.6246464252471924
    },
    {
      "auxiliary_loss_clip": 0.01195374,
      "auxiliary_loss_mlp": 0.01032725,
      "balance_loss_clip": 1.0611701,
      "balance_loss_mlp": 1.02308083,
      "epoch": 0.40353514098478926,
      "flos": 22017838055040.0,
      "grad_norm": 2.8708404453939265,
      "language_loss": 0.65207601,
      "learning_rate": 2.7065197451137114e-06,
      "loss": 0.674357,
      "num_input_tokens_seen": 72513840,
      "step": 3356,
      "time_per_iteration": 2.603532552719116
    },
    {
      "auxiliary_loss_clip": 0.01193009,
      "auxiliary_loss_mlp": 0.01032997,
      "balance_loss_clip": 1.05685973,
      "balance_loss_mlp": 1.02374589,
      "epoch": 0.4036553838754284,
      "flos": 14246446089600.0,
      "grad_norm": 2.4657586403383447,
      "language_loss": 0.67426413,
      "learning_rate": 2.7057909439106894e-06,
      "loss": 0.6965242,
      "num_input_tokens_seen": 72531695,
      "step": 3357,
      "time_per_iteration": 3.653101921081543
    },
    {
      "auxiliary_loss_clip": 0.01194694,
      "auxiliary_loss_mlp": 0.01035284,
      "balance_loss_clip": 1.05846167,
      "balance_loss_mlp": 1.02579463,
      "epoch": 0.40377562676606743,
      "flos": 24790644443520.0,
      "grad_norm": 2.1123562682002865,
      "language_loss": 0.78759021,
      "learning_rate": 2.7050620356397417e-06,
      "loss": 0.80989003,
      "num_input_tokens_seen": 72550645,
      "step": 3358,
      "time_per_iteration": 3.518630266189575
    },
    {
      "auxiliary_loss_clip": 0.01190666,
      "auxiliary_loss_mlp": 0.01036664,
      "balance_loss_clip": 1.05917764,
      "balance_loss_mlp": 1.02794349,
      "epoch": 0.40389586965670654,
      "flos": 24061226958720.0,
      "grad_norm": 1.9570765653917905,
      "language_loss": 0.72664696,
      "learning_rate": 2.7043330204114437e-06,
      "loss": 0.74892026,
      "num_input_tokens_seen": 72569355,
      "step": 3359,
      "time_per_iteration": 2.735957384109497
    },
    {
      "auxiliary_loss_clip": 0.01186486,
      "auxiliary_loss_mlp": 0.01029533,
      "balance_loss_clip": 1.05491734,
      "balance_loss_mlp": 1.02064574,
      "epoch": 0.40401611254734565,
      "flos": 16399613934720.0,
      "grad_norm": 1.9497186764365566,
      "language_loss": 0.85305798,
      "learning_rate": 2.7036038983363862e-06,
      "loss": 0.87521815,
      "num_input_tokens_seen": 72585960,
      "step": 3360,
      "time_per_iteration": 3.593388557434082
    },
    {
      "auxiliary_loss_clip": 0.01188677,
      "auxiliary_loss_mlp": 0.01029464,
      "balance_loss_clip": 1.05707049,
      "balance_loss_mlp": 1.02156687,
      "epoch": 0.4041363554379847,
      "flos": 23988220565760.0,
      "grad_norm": 1.7395918157036205,
      "language_loss": 0.84318441,
      "learning_rate": 2.702874669525177e-06,
      "loss": 0.86536586,
      "num_input_tokens_seen": 72604440,
      "step": 3361,
      "time_per_iteration": 2.6813886165618896
    },
    {
      "auxiliary_loss_clip": 0.01195086,
      "auxiliary_loss_mlp": 0.01034798,
      "balance_loss_clip": 1.06135893,
      "balance_loss_mlp": 1.02616096,
      "epoch": 0.4042565983286238,
      "flos": 28401964899840.0,
      "grad_norm": 2.12407713650367,
      "language_loss": 0.69767392,
      "learning_rate": 2.7021453340884394e-06,
      "loss": 0.71997273,
      "num_input_tokens_seen": 72622165,
      "step": 3362,
      "time_per_iteration": 2.6458513736724854
    },
    {
      "auxiliary_loss_clip": 0.01192883,
      "auxiliary_loss_mlp": 0.01030953,
      "balance_loss_clip": 1.05951321,
      "balance_loss_mlp": 1.02178037,
      "epoch": 0.40437684121926293,
      "flos": 17710963660800.0,
      "grad_norm": 6.4196214049916405,
      "language_loss": 0.73003805,
      "learning_rate": 2.7014158921368125e-06,
      "loss": 0.75227642,
      "num_input_tokens_seen": 72640490,
      "step": 3363,
      "time_per_iteration": 2.6459643840789795
    },
    {
      "auxiliary_loss_clip": 0.01194008,
      "auxiliary_loss_mlp": 0.01029426,
      "balance_loss_clip": 1.05946183,
      "balance_loss_mlp": 1.02001476,
      "epoch": 0.404497084109902,
      "flos": 24018959629440.0,
      "grad_norm": 2.0358947453362783,
      "language_loss": 0.85949618,
      "learning_rate": 2.700686343780953e-06,
      "loss": 0.88173056,
      "num_input_tokens_seen": 72660360,
      "step": 3364,
      "time_per_iteration": 2.614319086074829
    },
    {
      "auxiliary_loss_clip": 0.01190449,
      "auxiliary_loss_mlp": 0.01036554,
      "balance_loss_clip": 1.05667591,
      "balance_loss_mlp": 1.02790582,
      "epoch": 0.4046173270005411,
      "flos": 22929861306240.0,
      "grad_norm": 1.7188347232904573,
      "language_loss": 0.88504338,
      "learning_rate": 2.699956689131532e-06,
      "loss": 0.90731341,
      "num_input_tokens_seen": 72680345,
      "step": 3365,
      "time_per_iteration": 2.639678478240967
    },
    {
      "auxiliary_loss_clip": 0.01191874,
      "auxiliary_loss_mlp": 0.01030964,
      "balance_loss_clip": 1.0568223,
      "balance_loss_mlp": 1.02221453,
      "epoch": 0.4047375698911802,
      "flos": 20668135582080.0,
      "grad_norm": 2.396924638302634,
      "language_loss": 0.84986562,
      "learning_rate": 2.699226928299238e-06,
      "loss": 0.87209404,
      "num_input_tokens_seen": 72698365,
      "step": 3366,
      "time_per_iteration": 2.597240686416626
    },
    {
      "auxiliary_loss_clip": 0.01188377,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.05682492,
      "balance_loss_mlp": 1.02090621,
      "epoch": 0.40485781278181926,
      "flos": 28912865996160.0,
      "grad_norm": 2.4546080958712255,
      "language_loss": 0.79088104,
      "learning_rate": 2.698497061394774e-06,
      "loss": 0.81306493,
      "num_input_tokens_seen": 72716850,
      "step": 3367,
      "time_per_iteration": 2.706160545349121
    },
    {
      "auxiliary_loss_clip": 0.01197992,
      "auxiliary_loss_mlp": 0.01028074,
      "balance_loss_clip": 1.06223285,
      "balance_loss_mlp": 1.01849008,
      "epoch": 0.40497805567245837,
      "flos": 23148377694720.0,
      "grad_norm": 2.536748903574192,
      "language_loss": 0.80553603,
      "learning_rate": 2.6977670885288627e-06,
      "loss": 0.82779682,
      "num_input_tokens_seen": 72738250,
      "step": 3368,
      "time_per_iteration": 2.6566028594970703
    },
    {
      "auxiliary_loss_clip": 0.01192967,
      "auxiliary_loss_mlp": 0.01032601,
      "balance_loss_clip": 1.05693126,
      "balance_loss_mlp": 1.02333283,
      "epoch": 0.4050982985630975,
      "flos": 16289404030080.0,
      "grad_norm": 2.334685992060761,
      "language_loss": 0.75045878,
      "learning_rate": 2.6970370098122378e-06,
      "loss": 0.7727145,
      "num_input_tokens_seen": 72755235,
      "step": 3369,
      "time_per_iteration": 2.6335699558258057
    },
    {
      "auxiliary_loss_clip": 0.01192823,
      "auxiliary_loss_mlp": 0.01030424,
      "balance_loss_clip": 1.05848408,
      "balance_loss_mlp": 1.02086306,
      "epoch": 0.40521854145373654,
      "flos": 34459484353920.0,
      "grad_norm": 1.598131078794537,
      "language_loss": 0.86899394,
      "learning_rate": 2.6963068253556535e-06,
      "loss": 0.89122641,
      "num_input_tokens_seen": 72776620,
      "step": 3370,
      "time_per_iteration": 2.675605535507202
    },
    {
      "auxiliary_loss_clip": 0.01195433,
      "auxiliary_loss_mlp": 0.01032879,
      "balance_loss_clip": 1.05779529,
      "balance_loss_mlp": 1.02283525,
      "epoch": 0.40533878434437565,
      "flos": 25331099454720.0,
      "grad_norm": 1.8843452633207252,
      "language_loss": 0.85717541,
      "learning_rate": 2.6955765352698763e-06,
      "loss": 0.87945855,
      "num_input_tokens_seen": 72796765,
      "step": 3371,
      "time_per_iteration": 2.680635690689087
    },
    {
      "auxiliary_loss_clip": 0.01193599,
      "auxiliary_loss_mlp": 0.01029764,
      "balance_loss_clip": 1.05662942,
      "balance_loss_mlp": 1.01965535,
      "epoch": 0.40545902723501476,
      "flos": 15012061505280.0,
      "grad_norm": 1.8676158438365507,
      "language_loss": 0.73145223,
      "learning_rate": 2.6948461396656923e-06,
      "loss": 0.75368583,
      "num_input_tokens_seen": 72814175,
      "step": 3372,
      "time_per_iteration": 2.6163768768310547
    },
    {
      "auxiliary_loss_clip": 0.01195763,
      "auxiliary_loss_mlp": 0.01029952,
      "balance_loss_clip": 1.0582093,
      "balance_loss_mlp": 1.02057016,
      "epoch": 0.4055792701256538,
      "flos": 25521103422720.0,
      "grad_norm": 2.515150277271155,
      "language_loss": 0.73880827,
      "learning_rate": 2.6941156386539013e-06,
      "loss": 0.76106536,
      "num_input_tokens_seen": 72834125,
      "step": 3373,
      "time_per_iteration": 2.6715712547302246
    },
    {
      "auxiliary_loss_clip": 0.01189693,
      "auxiliary_loss_mlp": 0.01031709,
      "balance_loss_clip": 1.05729342,
      "balance_loss_mlp": 1.02213049,
      "epoch": 0.4056995130162929,
      "flos": 19574583972480.0,
      "grad_norm": 2.43889351923012,
      "language_loss": 0.81128299,
      "learning_rate": 2.6933850323453203e-06,
      "loss": 0.83349693,
      "num_input_tokens_seen": 72852570,
      "step": 3374,
      "time_per_iteration": 2.6446266174316406
    },
    {
      "auxiliary_loss_clip": 0.01191801,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 1.05971146,
      "balance_loss_mlp": 1.0242424,
      "epoch": 0.405819755906932,
      "flos": 15413794191360.0,
      "grad_norm": 1.8399710967963512,
      "language_loss": 0.74725795,
      "learning_rate": 2.6926543208507806e-06,
      "loss": 0.76951069,
      "num_input_tokens_seen": 72871250,
      "step": 3375,
      "time_per_iteration": 2.5444557666778564
    },
    {
      "auxiliary_loss_clip": 0.01190645,
      "auxiliary_loss_mlp": 0.01029915,
      "balance_loss_clip": 1.05753446,
      "balance_loss_mlp": 1.01972866,
      "epoch": 0.4059399987975711,
      "flos": 21433930565760.0,
      "grad_norm": 2.175492262175169,
      "language_loss": 0.79913205,
      "learning_rate": 2.6919235042811316e-06,
      "loss": 0.82133764,
      "num_input_tokens_seen": 72890035,
      "step": 3376,
      "time_per_iteration": 2.6133229732513428
    },
    {
      "auxiliary_loss_clip": 0.01194769,
      "auxiliary_loss_mlp": 0.01031597,
      "balance_loss_clip": 1.05972767,
      "balance_loss_mlp": 1.02179837,
      "epoch": 0.4060602416882102,
      "flos": 25556942217600.0,
      "grad_norm": 1.9602081280045578,
      "language_loss": 0.76431876,
      "learning_rate": 2.691192582747237e-06,
      "loss": 0.78658247,
      "num_input_tokens_seen": 72909665,
      "step": 3377,
      "time_per_iteration": 2.6209466457366943
    },
    {
      "auxiliary_loss_clip": 0.01191567,
      "auxiliary_loss_mlp": 0.01031318,
      "balance_loss_clip": 1.05791771,
      "balance_loss_mlp": 1.02161992,
      "epoch": 0.40618048457884925,
      "flos": 23766759262080.0,
      "grad_norm": 6.683576403421113,
      "language_loss": 0.73736721,
      "learning_rate": 2.6904615563599765e-06,
      "loss": 0.75959611,
      "num_input_tokens_seen": 72929465,
      "step": 3378,
      "time_per_iteration": 2.648223638534546
    },
    {
      "auxiliary_loss_clip": 0.01190851,
      "auxiliary_loss_mlp": 0.01030696,
      "balance_loss_clip": 1.05702662,
      "balance_loss_mlp": 1.02154624,
      "epoch": 0.40630072746948837,
      "flos": 17639681120640.0,
      "grad_norm": 2.2972593382375317,
      "language_loss": 0.83371758,
      "learning_rate": 2.6897304252302477e-06,
      "loss": 0.85593307,
      "num_input_tokens_seen": 72946785,
      "step": 3379,
      "time_per_iteration": 2.5569934844970703
    },
    {
      "auxiliary_loss_clip": 0.01081755,
      "auxiliary_loss_mlp": 0.01002122,
      "balance_loss_clip": 1.01526773,
      "balance_loss_mlp": 1.00093603,
      "epoch": 0.4064209703601275,
      "flos": 60836053063680.0,
      "grad_norm": 0.7885371961680956,
      "language_loss": 0.54815161,
      "learning_rate": 2.688999189468962e-06,
      "loss": 0.56899035,
      "num_input_tokens_seen": 73003215,
      "step": 3380,
      "time_per_iteration": 3.0907492637634277
    },
    {
      "auxiliary_loss_clip": 0.01192668,
      "auxiliary_loss_mlp": 0.01034127,
      "balance_loss_clip": 1.05761504,
      "balance_loss_mlp": 1.02489424,
      "epoch": 0.40654121325076653,
      "flos": 24024346669440.0,
      "grad_norm": 2.591322418718579,
      "language_loss": 0.76269919,
      "learning_rate": 2.6882678491870464e-06,
      "loss": 0.78496718,
      "num_input_tokens_seen": 73023650,
      "step": 3381,
      "time_per_iteration": 2.651655673980713
    },
    {
      "auxiliary_loss_clip": 0.01194426,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.05841827,
      "balance_loss_mlp": 1.01931942,
      "epoch": 0.40666145614140564,
      "flos": 27344252085120.0,
      "grad_norm": 2.0889226501619644,
      "language_loss": 0.71471918,
      "learning_rate": 2.6875364044954453e-06,
      "loss": 0.73694694,
      "num_input_tokens_seen": 73043880,
      "step": 3382,
      "time_per_iteration": 2.714158773422241
    },
    {
      "auxiliary_loss_clip": 0.01189937,
      "auxiliary_loss_mlp": 0.0102942,
      "balance_loss_clip": 1.05728889,
      "balance_loss_mlp": 1.02056301,
      "epoch": 0.40678169903204475,
      "flos": 26176724415360.0,
      "grad_norm": 2.6637571940521587,
      "language_loss": 0.82196337,
      "learning_rate": 2.6868048555051185e-06,
      "loss": 0.84415692,
      "num_input_tokens_seen": 73065410,
      "step": 3383,
      "time_per_iteration": 3.6804912090301514
    },
    {
      "auxiliary_loss_clip": 0.01195572,
      "auxiliary_loss_mlp": 0.01033436,
      "balance_loss_clip": 1.06005073,
      "balance_loss_mlp": 1.02363694,
      "epoch": 0.4069019419226838,
      "flos": 28622420622720.0,
      "grad_norm": 2.4016855158625576,
      "language_loss": 0.85516214,
      "learning_rate": 2.686073202327041e-06,
      "loss": 0.87745225,
      "num_input_tokens_seen": 73084410,
      "step": 3384,
      "time_per_iteration": 2.724564790725708
    },
    {
      "auxiliary_loss_clip": 0.01190465,
      "auxiliary_loss_mlp": 0.01033493,
      "balance_loss_clip": 1.05723536,
      "balance_loss_mlp": 1.0246594,
      "epoch": 0.4070221848133229,
      "flos": 25229006023680.0,
      "grad_norm": 1.589385855327594,
      "language_loss": 0.73309797,
      "learning_rate": 2.6853414450722043e-06,
      "loss": 0.7553376,
      "num_input_tokens_seen": 73104075,
      "step": 3385,
      "time_per_iteration": 2.709934711456299
    },
    {
      "auxiliary_loss_clip": 0.01190627,
      "auxiliary_loss_mlp": 0.01031207,
      "balance_loss_clip": 1.05796671,
      "balance_loss_mlp": 1.02273703,
      "epoch": 0.40714242770396203,
      "flos": 18405224709120.0,
      "grad_norm": 2.6891141374404595,
      "language_loss": 0.85307288,
      "learning_rate": 2.684609583851616e-06,
      "loss": 0.87529117,
      "num_input_tokens_seen": 73122250,
      "step": 3386,
      "time_per_iteration": 5.335178375244141
    },
    {
      "auxiliary_loss_clip": 0.011916,
      "auxiliary_loss_mlp": 0.01031009,
      "balance_loss_clip": 1.0571059,
      "balance_loss_mlp": 1.02150762,
      "epoch": 0.4072626705946011,
      "flos": 30228920403840.0,
      "grad_norm": 1.6058215717807756,
      "language_loss": 0.80688679,
      "learning_rate": 2.683877618776297e-06,
      "loss": 0.82911289,
      "num_input_tokens_seen": 73144505,
      "step": 3387,
      "time_per_iteration": 2.6845948696136475
    },
    {
      "auxiliary_loss_clip": 0.01192577,
      "auxiliary_loss_mlp": 0.01033987,
      "balance_loss_clip": 1.05683148,
      "balance_loss_mlp": 1.02387261,
      "epoch": 0.4073829134852402,
      "flos": 21834549930240.0,
      "grad_norm": 3.458922553053462,
      "language_loss": 0.74246657,
      "learning_rate": 2.6831455499572876e-06,
      "loss": 0.76473224,
      "num_input_tokens_seen": 73162440,
      "step": 3388,
      "time_per_iteration": 2.6540236473083496
    },
    {
      "auxiliary_loss_clip": 0.01190863,
      "auxiliary_loss_mlp": 0.01032263,
      "balance_loss_clip": 1.05630445,
      "balance_loss_mlp": 1.02293456,
      "epoch": 0.40750315637587925,
      "flos": 25260211964160.0,
      "grad_norm": 2.2797176665247947,
      "language_loss": 0.77823246,
      "learning_rate": 2.682413377505641e-06,
      "loss": 0.80046374,
      "num_input_tokens_seen": 73181245,
      "step": 3389,
      "time_per_iteration": 2.6509597301483154
    },
    {
      "auxiliary_loss_clip": 0.01190666,
      "auxiliary_loss_mlp": 0.01033753,
      "balance_loss_clip": 1.05555725,
      "balance_loss_mlp": 1.02336955,
      "epoch": 0.40762339926651836,
      "flos": 19712767593600.0,
      "grad_norm": 1.8669028951607387,
      "language_loss": 0.76570958,
      "learning_rate": 2.6816811015324284e-06,
      "loss": 0.78795379,
      "num_input_tokens_seen": 73199295,
      "step": 3390,
      "time_per_iteration": 2.606863498687744
    },
    {
      "auxiliary_loss_clip": 0.01079489,
      "auxiliary_loss_mlp": 0.01001101,
      "balance_loss_clip": 1.01322246,
      "balance_loss_mlp": 0.99999803,
      "epoch": 0.40774364215715747,
      "flos": 71449307314560.0,
      "grad_norm": 0.7301936073587937,
      "language_loss": 0.56678712,
      "learning_rate": 2.6809487221487343e-06,
      "loss": 0.58759296,
      "num_input_tokens_seen": 73258780,
      "step": 3391,
      "time_per_iteration": 3.0706064701080322
    },
    {
      "auxiliary_loss_clip": 0.01190057,
      "auxiliary_loss_mlp": 0.01037485,
      "balance_loss_clip": 1.05539429,
      "balance_loss_mlp": 1.02783513,
      "epoch": 0.4078638850477965,
      "flos": 15084134144640.0,
      "grad_norm": 2.314483233783014,
      "language_loss": 0.81670797,
      "learning_rate": 2.6802162394656605e-06,
      "loss": 0.83898342,
      "num_input_tokens_seen": 73275490,
      "step": 3392,
      "time_per_iteration": 2.6009361743927
    },
    {
      "auxiliary_loss_clip": 0.0118588,
      "auxiliary_loss_mlp": 0.01027106,
      "balance_loss_clip": 1.05434895,
      "balance_loss_mlp": 1.01819468,
      "epoch": 0.40798412793843564,
      "flos": 23842890138240.0,
      "grad_norm": 2.9201903746021736,
      "language_loss": 0.71877837,
      "learning_rate": 2.679483653594324e-06,
      "loss": 0.74090821,
      "num_input_tokens_seen": 73297260,
      "step": 3393,
      "time_per_iteration": 2.617309093475342
    },
    {
      "auxiliary_loss_clip": 0.01191908,
      "auxiliary_loss_mlp": 0.01027842,
      "balance_loss_clip": 1.05603671,
      "balance_loss_mlp": 1.01722622,
      "epoch": 0.40810437082907475,
      "flos": 21065774117760.0,
      "grad_norm": 2.1808547873734705,
      "language_loss": 0.76479435,
      "learning_rate": 2.678750964645857e-06,
      "loss": 0.78699183,
      "num_input_tokens_seen": 73316340,
      "step": 3394,
      "time_per_iteration": 2.594665050506592
    },
    {
      "auxiliary_loss_clip": 0.01195084,
      "auxiliary_loss_mlp": 0.01043426,
      "balance_loss_clip": 1.05906916,
      "balance_loss_mlp": 1.03301287,
      "epoch": 0.4082246137197138,
      "flos": 11321377948800.0,
      "grad_norm": 2.447216585973976,
      "language_loss": 0.83657783,
      "learning_rate": 2.6780181727314094e-06,
      "loss": 0.85896289,
      "num_input_tokens_seen": 73331245,
      "step": 3395,
      "time_per_iteration": 2.6199111938476562
    },
    {
      "auxiliary_loss_clip": 0.0119357,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.05871463,
      "balance_loss_mlp": 1.02213085,
      "epoch": 0.4083448566103529,
      "flos": 19062569554560.0,
      "grad_norm": 1.852428571901943,
      "language_loss": 0.77568078,
      "learning_rate": 2.6772852779621435e-06,
      "loss": 0.79793239,
      "num_input_tokens_seen": 73349105,
      "step": 3396,
      "time_per_iteration": 2.590456485748291
    },
    {
      "auxiliary_loss_clip": 0.01188896,
      "auxiliary_loss_mlp": 0.0102923,
      "balance_loss_clip": 1.05736637,
      "balance_loss_mlp": 1.02042603,
      "epoch": 0.408465099500992,
      "flos": 23550254035200.0,
      "grad_norm": 1.859281607901097,
      "language_loss": 0.87166226,
      "learning_rate": 2.676552280449239e-06,
      "loss": 0.89384353,
      "num_input_tokens_seen": 73368990,
      "step": 3397,
      "time_per_iteration": 2.692159652709961
    },
    {
      "auxiliary_loss_clip": 0.0119149,
      "auxiliary_loss_mlp": 0.01032549,
      "balance_loss_clip": 1.05761373,
      "balance_loss_mlp": 1.02328086,
      "epoch": 0.4085853423916311,
      "flos": 12750012558720.0,
      "grad_norm": 2.1655031750080243,
      "language_loss": 0.75936925,
      "learning_rate": 2.6758191803038917e-06,
      "loss": 0.78160965,
      "num_input_tokens_seen": 73387485,
      "step": 3398,
      "time_per_iteration": 2.565335512161255
    },
    {
      "auxiliary_loss_clip": 0.01197094,
      "auxiliary_loss_mlp": 0.01031091,
      "balance_loss_clip": 1.05986381,
      "balance_loss_mlp": 1.02130413,
      "epoch": 0.4087055852822702,
      "flos": 24353072962560.0,
      "grad_norm": 1.9041641180231113,
      "language_loss": 0.83181298,
      "learning_rate": 2.6750859776373125e-06,
      "loss": 0.85409486,
      "num_input_tokens_seen": 73406940,
      "step": 3399,
      "time_per_iteration": 2.648921012878418
    },
    {
      "auxiliary_loss_clip": 0.01078664,
      "auxiliary_loss_mlp": 0.01002036,
      "balance_loss_clip": 1.01278138,
      "balance_loss_mlp": 1.00092781,
      "epoch": 0.4088258281729093,
      "flos": 66387950720640.0,
      "grad_norm": 0.7708536253490833,
      "language_loss": 0.60393274,
      "learning_rate": 2.674352672560727e-06,
      "loss": 0.62473977,
      "num_input_tokens_seen": 73468385,
      "step": 3400,
      "time_per_iteration": 3.168309211730957
    },
    {
      "auxiliary_loss_clip": 0.01193673,
      "auxiliary_loss_mlp": 0.01033201,
      "balance_loss_clip": 1.05946589,
      "balance_loss_mlp": 1.02361703,
      "epoch": 0.40894607106354836,
      "flos": 20449260057600.0,
      "grad_norm": 1.9701500594843617,
      "language_loss": 0.76831973,
      "learning_rate": 2.673619265185377e-06,
      "loss": 0.79058844,
      "num_input_tokens_seen": 73488225,
      "step": 3401,
      "time_per_iteration": 2.636655569076538
    },
    {
      "auxiliary_loss_clip": 0.01194652,
      "auxiliary_loss_mlp": 0.01041585,
      "balance_loss_clip": 1.05910063,
      "balance_loss_mlp": 1.03201294,
      "epoch": 0.40906631395418747,
      "flos": 27053627143680.0,
      "grad_norm": 1.7587372503704115,
      "language_loss": 0.78202075,
      "learning_rate": 2.672885755622521e-06,
      "loss": 0.8043831,
      "num_input_tokens_seen": 73510640,
      "step": 3402,
      "time_per_iteration": 2.5895073413848877
    },
    {
      "auxiliary_loss_clip": 0.01189116,
      "auxiliary_loss_mlp": 0.01030015,
      "balance_loss_clip": 1.05543756,
      "balance_loss_mlp": 1.02125931,
      "epoch": 0.4091865568448266,
      "flos": 25484151306240.0,
      "grad_norm": 2.1206305866910675,
      "language_loss": 0.7012831,
      "learning_rate": 2.67215214398343e-06,
      "loss": 0.72347438,
      "num_input_tokens_seen": 73530655,
      "step": 3403,
      "time_per_iteration": 2.703031063079834
    },
    {
      "auxiliary_loss_clip": 0.0119518,
      "auxiliary_loss_mlp": 0.01034927,
      "balance_loss_clip": 1.05739021,
      "balance_loss_mlp": 1.02518189,
      "epoch": 0.40930679973546563,
      "flos": 28657864368000.0,
      "grad_norm": 2.782420070208454,
      "language_loss": 0.7851916,
      "learning_rate": 2.671418430379393e-06,
      "loss": 0.80749273,
      "num_input_tokens_seen": 73549340,
      "step": 3404,
      "time_per_iteration": 2.650104284286499
    },
    {
      "auxiliary_loss_clip": 0.01189526,
      "auxiliary_loss_mlp": 0.01030761,
      "balance_loss_clip": 1.05626512,
      "balance_loss_mlp": 1.02121878,
      "epoch": 0.40942704262610474,
      "flos": 20886292834560.0,
      "grad_norm": 2.0483743757018655,
      "language_loss": 0.835769,
      "learning_rate": 2.670684614921715e-06,
      "loss": 0.85797191,
      "num_input_tokens_seen": 73568315,
      "step": 3405,
      "time_per_iteration": 2.5822267532348633
    },
    {
      "auxiliary_loss_clip": 0.01190707,
      "auxiliary_loss_mlp": 0.01034196,
      "balance_loss_clip": 1.05718708,
      "balance_loss_mlp": 1.02490962,
      "epoch": 0.4095472855167438,
      "flos": 21618080616960.0,
      "grad_norm": 3.6846216584209692,
      "language_loss": 0.69376653,
      "learning_rate": 2.6699506977217128e-06,
      "loss": 0.71601558,
      "num_input_tokens_seen": 73588490,
      "step": 3406,
      "time_per_iteration": 2.6547441482543945
    },
    {
      "auxiliary_loss_clip": 0.01192135,
      "auxiliary_loss_mlp": 0.0103562,
      "balance_loss_clip": 1.05805564,
      "balance_loss_mlp": 1.02575576,
      "epoch": 0.4096675284073829,
      "flos": 27926112499200.0,
      "grad_norm": 2.000524808575555,
      "language_loss": 0.70163488,
      "learning_rate": 2.6692166788907233e-06,
      "loss": 0.72391242,
      "num_input_tokens_seen": 73608685,
      "step": 3407,
      "time_per_iteration": 2.68566632270813
    },
    {
      "auxiliary_loss_clip": 0.01192153,
      "auxiliary_loss_mlp": 0.01034071,
      "balance_loss_clip": 1.05846047,
      "balance_loss_mlp": 1.02356827,
      "epoch": 0.409787771298022,
      "flos": 19206607092480.0,
      "grad_norm": 1.8005340470519746,
      "language_loss": 0.76703912,
      "learning_rate": 2.6684825585400957e-06,
      "loss": 0.7893014,
      "num_input_tokens_seen": 73627630,
      "step": 3408,
      "time_per_iteration": 2.5714235305786133
    },
    {
      "auxiliary_loss_clip": 0.01078863,
      "auxiliary_loss_mlp": 0.01002473,
      "balance_loss_clip": 1.01289678,
      "balance_loss_mlp": 1.00140595,
      "epoch": 0.4099080141886611,
      "flos": 59269234832640.0,
      "grad_norm": 0.8103697780856193,
      "language_loss": 0.65111727,
      "learning_rate": 2.6677483367811947e-06,
      "loss": 0.67193061,
      "num_input_tokens_seen": 73687670,
      "step": 3409,
      "time_per_iteration": 4.193026065826416
    },
    {
      "auxiliary_loss_clip": 0.01189842,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.05505252,
      "balance_loss_mlp": 1.02144051,
      "epoch": 0.4100282570793002,
      "flos": 21906443001600.0,
      "grad_norm": 2.0115206688221843,
      "language_loss": 0.75846899,
      "learning_rate": 2.6670140137254028e-06,
      "loss": 0.78067571,
      "num_input_tokens_seen": 73707145,
      "step": 3410,
      "time_per_iteration": 2.6068458557128906
    },
    {
      "auxiliary_loss_clip": 0.01188332,
      "auxiliary_loss_mlp": 0.01030622,
      "balance_loss_clip": 1.05499971,
      "balance_loss_mlp": 1.02114511,
      "epoch": 0.4101484999699393,
      "flos": 18551596631040.0,
      "grad_norm": 2.520093310882291,
      "language_loss": 0.90213132,
      "learning_rate": 2.666279589484115e-06,
      "loss": 0.92432088,
      "num_input_tokens_seen": 73725045,
      "step": 3411,
      "time_per_iteration": 2.578481912612915
    },
    {
      "auxiliary_loss_clip": 0.01193428,
      "auxiliary_loss_mlp": 0.01029056,
      "balance_loss_clip": 1.05895376,
      "balance_loss_mlp": 1.01951957,
      "epoch": 0.41026874286057835,
      "flos": 19094529680640.0,
      "grad_norm": 1.8600029357405148,
      "language_loss": 0.81215394,
      "learning_rate": 2.6655450641687435e-06,
      "loss": 0.83437884,
      "num_input_tokens_seen": 73742610,
      "step": 3412,
      "time_per_iteration": 3.542126417160034
    },
    {
      "auxiliary_loss_clip": 0.01190536,
      "auxiliary_loss_mlp": 0.01031659,
      "balance_loss_clip": 1.05867386,
      "balance_loss_mlp": 1.02253318,
      "epoch": 0.41038898575121746,
      "flos": 31209568588800.0,
      "grad_norm": 1.764182718243202,
      "language_loss": 0.69122076,
      "learning_rate": 2.664810437890715e-06,
      "loss": 0.71344274,
      "num_input_tokens_seen": 73764280,
      "step": 3413,
      "time_per_iteration": 3.6048786640167236
    },
    {
      "auxiliary_loss_clip": 0.0118983,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.05913615,
      "balance_loss_mlp": 1.02109075,
      "epoch": 0.41050922864185657,
      "flos": 14355865895040.0,
      "grad_norm": 2.032514093524514,
      "language_loss": 0.79333472,
      "learning_rate": 2.6640757107614714e-06,
      "loss": 0.81553322,
      "num_input_tokens_seen": 73782375,
      "step": 3414,
      "time_per_iteration": 2.6003665924072266
    },
    {
      "auxiliary_loss_clip": 0.01192818,
      "auxiliary_loss_mlp": 0.01028777,
      "balance_loss_clip": 1.05771995,
      "balance_loss_mlp": 1.01907897,
      "epoch": 0.4106294715324956,
      "flos": 30956290813440.0,
      "grad_norm": 2.0562052099136596,
      "language_loss": 0.69005692,
      "learning_rate": 2.6633408828924697e-06,
      "loss": 0.71227288,
      "num_input_tokens_seen": 73801240,
      "step": 3415,
      "time_per_iteration": 2.675771951675415
    },
    {
      "auxiliary_loss_clip": 0.01192436,
      "auxiliary_loss_mlp": 0.01032319,
      "balance_loss_clip": 1.05837834,
      "balance_loss_mlp": 1.02286601,
      "epoch": 0.41074971442313474,
      "flos": 24457321209600.0,
      "grad_norm": 2.0862423446024527,
      "language_loss": 0.69764984,
      "learning_rate": 2.662605954395185e-06,
      "loss": 0.71989739,
      "num_input_tokens_seen": 73821200,
      "step": 3416,
      "time_per_iteration": 2.617403507232666
    },
    {
      "auxiliary_loss_clip": 0.0119132,
      "auxiliary_loss_mlp": 0.01032987,
      "balance_loss_clip": 1.0571878,
      "balance_loss_mlp": 1.02455318,
      "epoch": 0.41086995731377385,
      "flos": 21542991235200.0,
      "grad_norm": 1.6837084733046195,
      "language_loss": 0.83313835,
      "learning_rate": 2.6618709253811027e-06,
      "loss": 0.85538143,
      "num_input_tokens_seen": 73840655,
      "step": 3417,
      "time_per_iteration": 2.71211838722229
    },
    {
      "auxiliary_loss_clip": 0.01186745,
      "auxiliary_loss_mlp": 0.01022093,
      "balance_loss_clip": 1.05762339,
      "balance_loss_mlp": 1.01427269,
      "epoch": 0.4109902002044129,
      "flos": 20702753314560.0,
      "grad_norm": 1.8327978172810984,
      "language_loss": 0.87597269,
      "learning_rate": 2.6611357959617277e-06,
      "loss": 0.89806104,
      "num_input_tokens_seen": 73860275,
      "step": 3418,
      "time_per_iteration": 2.627385377883911
    },
    {
      "auxiliary_loss_clip": 0.01193277,
      "auxiliary_loss_mlp": 0.01032985,
      "balance_loss_clip": 1.05844712,
      "balance_loss_mlp": 1.02365112,
      "epoch": 0.411110443095052,
      "flos": 18179992477440.0,
      "grad_norm": 2.112123398147564,
      "language_loss": 0.91189992,
      "learning_rate": 2.660400566248578e-06,
      "loss": 0.93416262,
      "num_input_tokens_seen": 73878400,
      "step": 3419,
      "time_per_iteration": 2.59743070602417
    },
    {
      "auxiliary_loss_clip": 0.011919,
      "auxiliary_loss_mlp": 0.01034377,
      "balance_loss_clip": 1.05557013,
      "balance_loss_mlp": 1.02416122,
      "epoch": 0.41123068598569107,
      "flos": 14575244209920.0,
      "grad_norm": 2.5129042333116596,
      "language_loss": 0.6725632,
      "learning_rate": 2.6596652363531876e-06,
      "loss": 0.69482595,
      "num_input_tokens_seen": 73894275,
      "step": 3420,
      "time_per_iteration": 2.5693156719207764
    },
    {
      "auxiliary_loss_clip": 0.01190519,
      "auxiliary_loss_mlp": 0.01026792,
      "balance_loss_clip": 1.05795217,
      "balance_loss_mlp": 1.01802993,
      "epoch": 0.4113509288763302,
      "flos": 21177995184000.0,
      "grad_norm": 1.5576822729986703,
      "language_loss": 0.78232211,
      "learning_rate": 2.6589298063871055e-06,
      "loss": 0.80449522,
      "num_input_tokens_seen": 73914450,
      "step": 3421,
      "time_per_iteration": 2.6068623065948486
    },
    {
      "auxiliary_loss_clip": 0.01188206,
      "auxiliary_loss_mlp": 0.01030478,
      "balance_loss_clip": 1.05699837,
      "balance_loss_mlp": 1.02179325,
      "epoch": 0.4114711717669693,
      "flos": 18442212739200.0,
      "grad_norm": 1.8869926869697606,
      "language_loss": 0.69521844,
      "learning_rate": 2.658194276461895e-06,
      "loss": 0.71740526,
      "num_input_tokens_seen": 73932375,
      "step": 3422,
      "time_per_iteration": 2.572283983230591
    },
    {
      "auxiliary_loss_clip": 0.01193548,
      "auxiliary_loss_mlp": 0.01039497,
      "balance_loss_clip": 1.05536389,
      "balance_loss_mlp": 1.02991843,
      "epoch": 0.41159141465760835,
      "flos": 27233395735680.0,
      "grad_norm": 2.0617178597399985,
      "language_loss": 0.66719502,
      "learning_rate": 2.6574586466891368e-06,
      "loss": 0.68952549,
      "num_input_tokens_seen": 73952850,
      "step": 3423,
      "time_per_iteration": 2.6415064334869385
    },
    {
      "auxiliary_loss_clip": 0.01186491,
      "auxiliary_loss_mlp": 0.01033421,
      "balance_loss_clip": 1.05464268,
      "balance_loss_mlp": 1.0242126,
      "epoch": 0.41171165754824746,
      "flos": 20006876154240.0,
      "grad_norm": 2.0317060748805513,
      "language_loss": 0.65318847,
      "learning_rate": 2.6567229171804247e-06,
      "loss": 0.67538762,
      "num_input_tokens_seen": 73970735,
      "step": 3424,
      "time_per_iteration": 2.606349468231201
    },
    {
      "auxiliary_loss_clip": 0.01191349,
      "auxiliary_loss_mlp": 0.01032183,
      "balance_loss_clip": 1.05428493,
      "balance_loss_mlp": 1.0217818,
      "epoch": 0.41183190043888657,
      "flos": 18004318035840.0,
      "grad_norm": 4.000634428717412,
      "language_loss": 0.87915045,
      "learning_rate": 2.655987088047368e-06,
      "loss": 0.90138578,
      "num_input_tokens_seen": 73989080,
      "step": 3425,
      "time_per_iteration": 2.605288028717041
    },
    {
      "auxiliary_loss_clip": 0.01193789,
      "auxiliary_loss_mlp": 0.01027921,
      "balance_loss_clip": 1.05828381,
      "balance_loss_mlp": 1.0178901,
      "epoch": 0.4119521433295256,
      "flos": 27163370171520.0,
      "grad_norm": 1.9510030363927093,
      "language_loss": 0.79049897,
      "learning_rate": 2.6552511594015912e-06,
      "loss": 0.81271601,
      "num_input_tokens_seen": 74009470,
      "step": 3426,
      "time_per_iteration": 2.774817705154419
    },
    {
      "auxiliary_loss_clip": 0.01191675,
      "auxiliary_loss_mlp": 0.01033806,
      "balance_loss_clip": 1.05628419,
      "balance_loss_mlp": 1.0234468,
      "epoch": 0.41207238622016473,
      "flos": 15122020014720.0,
      "grad_norm": 2.141131362622358,
      "language_loss": 0.85551453,
      "learning_rate": 2.654515131354735e-06,
      "loss": 0.87776935,
      "num_input_tokens_seen": 74027735,
      "step": 3427,
      "time_per_iteration": 2.642432689666748
    },
    {
      "auxiliary_loss_clip": 0.0119174,
      "auxiliary_loss_mlp": 0.01035626,
      "balance_loss_clip": 1.0572896,
      "balance_loss_mlp": 1.02614856,
      "epoch": 0.41219262911080384,
      "flos": 27052872958080.0,
      "grad_norm": 1.98447127996506,
      "language_loss": 0.85250038,
      "learning_rate": 2.653779004018453e-06,
      "loss": 0.87477404,
      "num_input_tokens_seen": 74048300,
      "step": 3428,
      "time_per_iteration": 2.642119884490967
    },
    {
      "auxiliary_loss_clip": 0.01189299,
      "auxiliary_loss_mlp": 0.01031571,
      "balance_loss_clip": 1.05562639,
      "balance_loss_mlp": 1.02245164,
      "epoch": 0.4123128720014429,
      "flos": 24686360282880.0,
      "grad_norm": 1.7711722946953052,
      "language_loss": 0.82482904,
      "learning_rate": 2.653042777504417e-06,
      "loss": 0.84703779,
      "num_input_tokens_seen": 74070890,
      "step": 3429,
      "time_per_iteration": 2.7745110988616943
    },
    {
      "auxiliary_loss_clip": 0.01192412,
      "auxiliary_loss_mlp": 0.01032093,
      "balance_loss_clip": 1.05448222,
      "balance_loss_mlp": 1.02200842,
      "epoch": 0.412433114892082,
      "flos": 26244774731520.0,
      "grad_norm": 2.7956896174645194,
      "language_loss": 0.79741073,
      "learning_rate": 2.6523064519243105e-06,
      "loss": 0.81965578,
      "num_input_tokens_seen": 74090460,
      "step": 3430,
      "time_per_iteration": 2.6887075901031494
    },
    {
      "auxiliary_loss_clip": 0.01191979,
      "auxiliary_loss_mlp": 0.01031568,
      "balance_loss_clip": 1.05728126,
      "balance_loss_mlp": 1.02103567,
      "epoch": 0.4125533577827211,
      "flos": 21361031913600.0,
      "grad_norm": 2.396749994498446,
      "language_loss": 0.79317796,
      "learning_rate": 2.6515700273898333e-06,
      "loss": 0.81541348,
      "num_input_tokens_seen": 74108335,
      "step": 3431,
      "time_per_iteration": 2.6292591094970703
    },
    {
      "auxiliary_loss_clip": 0.01190515,
      "auxiliary_loss_mlp": 0.01030814,
      "balance_loss_clip": 1.05722618,
      "balance_loss_mlp": 1.02177238,
      "epoch": 0.4126736006733602,
      "flos": 26067556005120.0,
      "grad_norm": 2.9433543977587076,
      "language_loss": 0.68668711,
      "learning_rate": 2.6508335040127018e-06,
      "loss": 0.70890033,
      "num_input_tokens_seen": 74128030,
      "step": 3432,
      "time_per_iteration": 2.672684907913208
    },
    {
      "auxiliary_loss_clip": 0.01191977,
      "auxiliary_loss_mlp": 0.01034005,
      "balance_loss_clip": 1.05730224,
      "balance_loss_mlp": 1.02487969,
      "epoch": 0.4127938435639993,
      "flos": 25666146541440.0,
      "grad_norm": 1.4723065571631317,
      "language_loss": 0.76968634,
      "learning_rate": 2.6500968819046446e-06,
      "loss": 0.79194611,
      "num_input_tokens_seen": 74148330,
      "step": 3433,
      "time_per_iteration": 2.6503145694732666
    },
    {
      "auxiliary_loss_clip": 0.01187817,
      "auxiliary_loss_mlp": 0.01028985,
      "balance_loss_clip": 1.05535722,
      "balance_loss_mlp": 1.01981187,
      "epoch": 0.4129140864546384,
      "flos": 17995914253440.0,
      "grad_norm": 2.36492286478678,
      "language_loss": 0.59098494,
      "learning_rate": 2.649360161177408e-06,
      "loss": 0.61315292,
      "num_input_tokens_seen": 74163390,
      "step": 3434,
      "time_per_iteration": 2.6507298946380615
    },
    {
      "auxiliary_loss_clip": 0.01193754,
      "auxiliary_loss_mlp": 0.01028195,
      "balance_loss_clip": 1.05901957,
      "balance_loss_mlp": 1.01794875,
      "epoch": 0.41303432934527745,
      "flos": 23732895715200.0,
      "grad_norm": 1.8068420649923222,
      "language_loss": 0.73223633,
      "learning_rate": 2.6486233419427504e-06,
      "loss": 0.7544558,
      "num_input_tokens_seen": 74183205,
      "step": 3435,
      "time_per_iteration": 2.6130526065826416
    },
    {
      "auxiliary_loss_clip": 0.01197518,
      "auxiliary_loss_mlp": 0.01035794,
      "balance_loss_clip": 1.06019044,
      "balance_loss_mlp": 1.02618575,
      "epoch": 0.41315457223591656,
      "flos": 19755286318080.0,
      "grad_norm": 2.1249146000711194,
      "language_loss": 0.74858332,
      "learning_rate": 2.6478864243124484e-06,
      "loss": 0.77091646,
      "num_input_tokens_seen": 74202870,
      "step": 3436,
      "time_per_iteration": 3.5752084255218506
    },
    {
      "auxiliary_loss_clip": 0.01188788,
      "auxiliary_loss_mlp": 0.01028479,
      "balance_loss_clip": 1.05559742,
      "balance_loss_mlp": 1.0194186,
      "epoch": 0.4132748151265556,
      "flos": 20923316778240.0,
      "grad_norm": 3.0525823676038826,
      "language_loss": 0.85388756,
      "learning_rate": 2.6471494083982903e-06,
      "loss": 0.87606019,
      "num_input_tokens_seen": 74222255,
      "step": 3437,
      "time_per_iteration": 2.7385904788970947
    },
    {
      "auxiliary_loss_clip": 0.0119134,
      "auxiliary_loss_mlp": 0.01033257,
      "balance_loss_clip": 1.05767059,
      "balance_loss_mlp": 1.02329731,
      "epoch": 0.4133950580171947,
      "flos": 32232520016640.0,
      "grad_norm": 1.7778359049199903,
      "language_loss": 0.75130057,
      "learning_rate": 2.6464122943120818e-06,
      "loss": 0.77354646,
      "num_input_tokens_seen": 74242480,
      "step": 3438,
      "time_per_iteration": 2.7111990451812744
    },
    {
      "auxiliary_loss_clip": 0.01194587,
      "auxiliary_loss_mlp": 0.01034874,
      "balance_loss_clip": 1.05933857,
      "balance_loss_mlp": 1.02545619,
      "epoch": 0.41351530090783384,
      "flos": 23292487059840.0,
      "grad_norm": 3.1631393195320987,
      "language_loss": 0.81992149,
      "learning_rate": 2.645675082165642e-06,
      "loss": 0.84221601,
      "num_input_tokens_seen": 74258690,
      "step": 3439,
      "time_per_iteration": 4.388284683227539
    },
    {
      "auxiliary_loss_clip": 0.01194058,
      "auxiliary_loss_mlp": 0.01038149,
      "balance_loss_clip": 1.05855191,
      "balance_loss_mlp": 1.02823043,
      "epoch": 0.4136355437984729,
      "flos": 25593571111680.0,
      "grad_norm": 2.2757122853555263,
      "language_loss": 0.75610626,
      "learning_rate": 2.644937772070806e-06,
      "loss": 0.77842832,
      "num_input_tokens_seen": 74277135,
      "step": 3440,
      "time_per_iteration": 3.6080737113952637
    },
    {
      "auxiliary_loss_clip": 0.01193754,
      "auxiliary_loss_mlp": 0.01029514,
      "balance_loss_clip": 1.06076658,
      "balance_loss_mlp": 1.0203054,
      "epoch": 0.413755786689112,
      "flos": 19828615933440.0,
      "grad_norm": 2.5133801862536522,
      "language_loss": 0.83417529,
      "learning_rate": 2.6442003641394225e-06,
      "loss": 0.85640794,
      "num_input_tokens_seen": 74294730,
      "step": 3441,
      "time_per_iteration": 2.626973867416382
    },
    {
      "auxiliary_loss_clip": 0.0119255,
      "auxiliary_loss_mlp": 0.01032406,
      "balance_loss_clip": 1.06023121,
      "balance_loss_mlp": 1.02362013,
      "epoch": 0.4138760295797511,
      "flos": 26870446759680.0,
      "grad_norm": 1.822932893916961,
      "language_loss": 0.84015775,
      "learning_rate": 2.643462858483356e-06,
      "loss": 0.86240733,
      "num_input_tokens_seen": 74315015,
      "step": 3442,
      "time_per_iteration": 2.6367015838623047
    },
    {
      "auxiliary_loss_clip": 0.0119346,
      "auxiliary_loss_mlp": 0.01034587,
      "balance_loss_clip": 1.05915654,
      "balance_loss_mlp": 1.02495456,
      "epoch": 0.41399627247039017,
      "flos": 16399254798720.0,
      "grad_norm": 1.8117275412957448,
      "language_loss": 0.7278372,
      "learning_rate": 2.6427252552144856e-06,
      "loss": 0.75011772,
      "num_input_tokens_seen": 74333665,
      "step": 3443,
      "time_per_iteration": 2.604604482650757
    },
    {
      "auxiliary_loss_clip": 0.01192411,
      "auxiliary_loss_mlp": 0.01031931,
      "balance_loss_clip": 1.05892515,
      "balance_loss_mlp": 1.02185774,
      "epoch": 0.4141165153610293,
      "flos": 22930220442240.0,
      "grad_norm": 2.7534366383392226,
      "language_loss": 0.7517513,
      "learning_rate": 2.6419875544447044e-06,
      "loss": 0.77399474,
      "num_input_tokens_seen": 74355065,
      "step": 3444,
      "time_per_iteration": 2.671250104904175
    },
    {
      "auxiliary_loss_clip": 0.01193074,
      "auxiliary_loss_mlp": 0.01031007,
      "balance_loss_clip": 1.05867815,
      "balance_loss_mlp": 1.02121997,
      "epoch": 0.4142367582516684,
      "flos": 25192556697600.0,
      "grad_norm": 1.6587139277281973,
      "language_loss": 0.7181201,
      "learning_rate": 2.6412497562859218e-06,
      "loss": 0.74036086,
      "num_input_tokens_seen": 74376345,
      "step": 3445,
      "time_per_iteration": 2.6619043350219727
    },
    {
      "auxiliary_loss_clip": 0.01193297,
      "auxiliary_loss_mlp": 0.01032125,
      "balance_loss_clip": 1.05825591,
      "balance_loss_mlp": 1.02287459,
      "epoch": 0.41435700114230745,
      "flos": 21690476478720.0,
      "grad_norm": 2.581069578309557,
      "language_loss": 0.76489329,
      "learning_rate": 2.6405118608500617e-06,
      "loss": 0.78714758,
      "num_input_tokens_seen": 74395170,
      "step": 3446,
      "time_per_iteration": 2.6263065338134766
    },
    {
      "auxiliary_loss_clip": 0.01190465,
      "auxiliary_loss_mlp": 0.01029382,
      "balance_loss_clip": 1.05896831,
      "balance_loss_mlp": 1.02040505,
      "epoch": 0.41447724403294656,
      "flos": 25995160143360.0,
      "grad_norm": 1.8251609007160576,
      "language_loss": 0.81495142,
      "learning_rate": 2.6397738682490613e-06,
      "loss": 0.83714986,
      "num_input_tokens_seen": 74416070,
      "step": 3447,
      "time_per_iteration": 2.692192316055298
    },
    {
      "auxiliary_loss_clip": 0.01191611,
      "auxiliary_loss_mlp": 0.01035167,
      "balance_loss_clip": 1.05783463,
      "balance_loss_mlp": 1.02552915,
      "epoch": 0.41459748692358567,
      "flos": 18259678800000.0,
      "grad_norm": 2.0247052603835543,
      "language_loss": 0.75408387,
      "learning_rate": 2.6390357785948734e-06,
      "loss": 0.77635169,
      "num_input_tokens_seen": 74433185,
      "step": 3448,
      "time_per_iteration": 2.613187313079834
    },
    {
      "auxiliary_loss_clip": 0.01190476,
      "auxiliary_loss_mlp": 0.01033806,
      "balance_loss_clip": 1.05663037,
      "balance_loss_mlp": 1.02384591,
      "epoch": 0.4147177298142247,
      "flos": 24168456034560.0,
      "grad_norm": 1.618073510596602,
      "language_loss": 0.79909545,
      "learning_rate": 2.6382975919994667e-06,
      "loss": 0.8213383,
      "num_input_tokens_seen": 74453760,
      "step": 3449,
      "time_per_iteration": 2.6368865966796875
    },
    {
      "auxiliary_loss_clip": 0.01190509,
      "auxiliary_loss_mlp": 0.01029664,
      "balance_loss_clip": 1.05924952,
      "balance_loss_mlp": 1.02133155,
      "epoch": 0.41483797270486383,
      "flos": 20084659056000.0,
      "grad_norm": 1.7505219469671274,
      "language_loss": 0.72990179,
      "learning_rate": 2.637559308574822e-06,
      "loss": 0.75210345,
      "num_input_tokens_seen": 74473505,
      "step": 3450,
      "time_per_iteration": 2.5681283473968506
    },
    {
      "auxiliary_loss_clip": 0.01192447,
      "auxiliary_loss_mlp": 0.01028904,
      "balance_loss_clip": 1.05950069,
      "balance_loss_mlp": 1.01998699,
      "epoch": 0.4149582155955029,
      "flos": 30081040110720.0,
      "grad_norm": 1.9625171605516187,
      "language_loss": 0.7154758,
      "learning_rate": 2.6368209284329376e-06,
      "loss": 0.73768938,
      "num_input_tokens_seen": 74494135,
      "step": 3451,
      "time_per_iteration": 2.692988157272339
    },
    {
      "auxiliary_loss_clip": 0.01192539,
      "auxiliary_loss_mlp": 0.0102942,
      "balance_loss_clip": 1.05872798,
      "balance_loss_mlp": 1.02040184,
      "epoch": 0.415078458486142,
      "flos": 16764394504320.0,
      "grad_norm": 2.122432198304349,
      "language_loss": 0.7587136,
      "learning_rate": 2.636082451685825e-06,
      "loss": 0.78093326,
      "num_input_tokens_seen": 74512335,
      "step": 3452,
      "time_per_iteration": 2.6360533237457275
    },
    {
      "auxiliary_loss_clip": 0.01195592,
      "auxiliary_loss_mlp": 0.01027816,
      "balance_loss_clip": 1.06173921,
      "balance_loss_mlp": 1.01889277,
      "epoch": 0.4151987013767811,
      "flos": 26033692458240.0,
      "grad_norm": 1.8160437573454096,
      "language_loss": 0.86643368,
      "learning_rate": 2.6353438784455094e-06,
      "loss": 0.8886677,
      "num_input_tokens_seen": 74535620,
      "step": 3453,
      "time_per_iteration": 2.706620216369629
    },
    {
      "auxiliary_loss_clip": 0.01195049,
      "auxiliary_loss_mlp": 0.0103067,
      "balance_loss_clip": 1.06241393,
      "balance_loss_mlp": 1.02017367,
      "epoch": 0.41531894426742016,
      "flos": 24608002763520.0,
      "grad_norm": 2.109447034103147,
      "language_loss": 0.7180075,
      "learning_rate": 2.6346052088240326e-06,
      "loss": 0.74026465,
      "num_input_tokens_seen": 74555140,
      "step": 3454,
      "time_per_iteration": 2.6789395809173584
    },
    {
      "auxiliary_loss_clip": 0.01191988,
      "auxiliary_loss_mlp": 0.01026992,
      "balance_loss_clip": 1.0587914,
      "balance_loss_mlp": 1.01787233,
      "epoch": 0.4154391871580593,
      "flos": 14975791747200.0,
      "grad_norm": 2.2708997018731916,
      "language_loss": 0.77217805,
      "learning_rate": 2.63386644293345e-06,
      "loss": 0.79436785,
      "num_input_tokens_seen": 74571485,
      "step": 3455,
      "time_per_iteration": 2.6246421337127686
    },
    {
      "auxiliary_loss_clip": 0.011922,
      "auxiliary_loss_mlp": 0.01029855,
      "balance_loss_clip": 1.06018627,
      "balance_loss_mlp": 1.0214324,
      "epoch": 0.4155594300486984,
      "flos": 14647173194880.0,
      "grad_norm": 2.066858306904035,
      "language_loss": 0.82552874,
      "learning_rate": 2.633127580885833e-06,
      "loss": 0.84774929,
      "num_input_tokens_seen": 74585985,
      "step": 3456,
      "time_per_iteration": 2.5832042694091797
    },
    {
      "auxiliary_loss_clip": 0.01192823,
      "auxiliary_loss_mlp": 0.01034079,
      "balance_loss_clip": 1.06106973,
      "balance_loss_mlp": 1.02525449,
      "epoch": 0.41567967293933744,
      "flos": 29497276275840.0,
      "grad_norm": 2.108842252861202,
      "language_loss": 0.64434624,
      "learning_rate": 2.632388622793265e-06,
      "loss": 0.66661531,
      "num_input_tokens_seen": 74605140,
      "step": 3457,
      "time_per_iteration": 2.6901278495788574
    },
    {
      "auxiliary_loss_clip": 0.01196771,
      "auxiliary_loss_mlp": 0.01032247,
      "balance_loss_clip": 1.06329966,
      "balance_loss_mlp": 1.02355647,
      "epoch": 0.41579991582997655,
      "flos": 19238387650560.0,
      "grad_norm": 2.7959084290503586,
      "language_loss": 0.67511857,
      "learning_rate": 2.6316495687678457e-06,
      "loss": 0.69740868,
      "num_input_tokens_seen": 74623790,
      "step": 3458,
      "time_per_iteration": 2.5786185264587402
    },
    {
      "auxiliary_loss_clip": 0.01190455,
      "auxiliary_loss_mlp": 0.01034621,
      "balance_loss_clip": 1.05882251,
      "balance_loss_mlp": 1.02569199,
      "epoch": 0.41592015872061566,
      "flos": 24462061804800.0,
      "grad_norm": 3.0042421394253367,
      "language_loss": 0.77028638,
      "learning_rate": 2.6309104189216887e-06,
      "loss": 0.79253721,
      "num_input_tokens_seen": 74641355,
      "step": 3459,
      "time_per_iteration": 2.5890636444091797
    },
    {
      "auxiliary_loss_clip": 0.01191694,
      "auxiliary_loss_mlp": 0.01030457,
      "balance_loss_clip": 1.05715609,
      "balance_loss_mlp": 1.02114117,
      "epoch": 0.4160404016112547,
      "flos": 20775651966720.0,
      "grad_norm": 3.4998889021100403,
      "language_loss": 0.74864256,
      "learning_rate": 2.630171173366923e-06,
      "loss": 0.77086401,
      "num_input_tokens_seen": 74657155,
      "step": 3460,
      "time_per_iteration": 2.6350173950195312
    },
    {
      "auxiliary_loss_clip": 0.01190858,
      "auxiliary_loss_mlp": 0.01028868,
      "balance_loss_clip": 1.05920517,
      "balance_loss_mlp": 1.01917076,
      "epoch": 0.41616064450189383,
      "flos": 13916462820480.0,
      "grad_norm": 3.358425680662842,
      "language_loss": 0.74608231,
      "learning_rate": 2.629431832215691e-06,
      "loss": 0.76827955,
      "num_input_tokens_seen": 74671960,
      "step": 3461,
      "time_per_iteration": 2.6039674282073975
    },
    {
      "auxiliary_loss_clip": 0.01190065,
      "auxiliary_loss_mlp": 0.0103163,
      "balance_loss_clip": 1.05699646,
      "balance_loss_mlp": 1.02195013,
      "epoch": 0.41628088739253294,
      "flos": 20010826650240.0,
      "grad_norm": 1.8980196375272214,
      "language_loss": 0.87194008,
      "learning_rate": 2.628692395580151e-06,
      "loss": 0.89415705,
      "num_input_tokens_seen": 74692050,
      "step": 3462,
      "time_per_iteration": 3.6121976375579834
    },
    {
      "auxiliary_loss_clip": 0.01192597,
      "auxiliary_loss_mlp": 0.01027196,
      "balance_loss_clip": 1.05835485,
      "balance_loss_mlp": 1.01752186,
      "epoch": 0.416401130283172,
      "flos": 29168801377920.0,
      "grad_norm": 1.945680075380737,
      "language_loss": 0.79275006,
      "learning_rate": 2.6279528635724747e-06,
      "loss": 0.81494796,
      "num_input_tokens_seen": 74712205,
      "step": 3463,
      "time_per_iteration": 2.65120005607605
    },
    {
      "auxiliary_loss_clip": 0.01196763,
      "auxiliary_loss_mlp": 0.01027853,
      "balance_loss_clip": 1.0597291,
      "balance_loss_mlp": 1.01909065,
      "epoch": 0.4165213731738111,
      "flos": 16246813478400.0,
      "grad_norm": 2.5967488230668034,
      "language_loss": 0.77966011,
      "learning_rate": 2.627213236304848e-06,
      "loss": 0.80190629,
      "num_input_tokens_seen": 74729005,
      "step": 3464,
      "time_per_iteration": 2.709496259689331
    },
    {
      "auxiliary_loss_clip": 0.0118976,
      "auxiliary_loss_mlp": 0.01034942,
      "balance_loss_clip": 1.05780172,
      "balance_loss_mlp": 1.02620387,
      "epoch": 0.4166416160644502,
      "flos": 33765438787200.0,
      "grad_norm": 2.993821802839214,
      "language_loss": 0.71199971,
      "learning_rate": 2.626473513889472e-06,
      "loss": 0.73424673,
      "num_input_tokens_seen": 74751385,
      "step": 3465,
      "time_per_iteration": 2.7246084213256836
    },
    {
      "auxiliary_loss_clip": 0.01193507,
      "auxiliary_loss_mlp": 0.01027997,
      "balance_loss_clip": 1.06030655,
      "balance_loss_mlp": 1.01893711,
      "epoch": 0.41676185895508927,
      "flos": 20917498775040.0,
      "grad_norm": 2.522733283748816,
      "language_loss": 0.82769769,
      "learning_rate": 2.625733696438562e-06,
      "loss": 0.84991276,
      "num_input_tokens_seen": 74768890,
      "step": 3466,
      "time_per_iteration": 4.413848161697388
    },
    {
      "auxiliary_loss_clip": 0.01194225,
      "auxiliary_loss_mlp": 0.0103146,
      "balance_loss_clip": 1.0605129,
      "balance_loss_mlp": 1.0223825,
      "epoch": 0.4168821018457284,
      "flos": 18406122549120.0,
      "grad_norm": 2.1945911264789624,
      "language_loss": 0.75194591,
      "learning_rate": 2.6249937840643476e-06,
      "loss": 0.77420276,
      "num_input_tokens_seen": 74787195,
      "step": 3467,
      "time_per_iteration": 2.5999016761779785
    },
    {
      "auxiliary_loss_clip": 0.01195279,
      "auxiliary_loss_mlp": 0.01029094,
      "balance_loss_clip": 1.0622288,
      "balance_loss_mlp": 1.01970053,
      "epoch": 0.41700234473636744,
      "flos": 18698399516160.0,
      "grad_norm": 2.1216898956687715,
      "language_loss": 0.6698947,
      "learning_rate": 2.6242537768790733e-06,
      "loss": 0.69213837,
      "num_input_tokens_seen": 74806350,
      "step": 3468,
      "time_per_iteration": 2.580348014831543
    },
    {
      "auxiliary_loss_clip": 0.01194299,
      "auxiliary_loss_mlp": 0.01033657,
      "balance_loss_clip": 1.05885196,
      "balance_loss_mlp": 1.02311254,
      "epoch": 0.41712258762700655,
      "flos": 31033283616000.0,
      "grad_norm": 1.8908142572421047,
      "language_loss": 0.68952048,
      "learning_rate": 2.6235136749949975e-06,
      "loss": 0.71179998,
      "num_input_tokens_seen": 74829800,
      "step": 3469,
      "time_per_iteration": 2.71278715133667
    },
    {
      "auxiliary_loss_clip": 0.01191017,
      "auxiliary_loss_mlp": 0.01033085,
      "balance_loss_clip": 1.05828476,
      "balance_loss_mlp": 1.02416217,
      "epoch": 0.41724283051764566,
      "flos": 35914763877120.0,
      "grad_norm": 2.2769095845436724,
      "language_loss": 0.61316907,
      "learning_rate": 2.6227734785243924e-06,
      "loss": 0.63541013,
      "num_input_tokens_seen": 74849760,
      "step": 3470,
      "time_per_iteration": 2.6759700775146484
    },
    {
      "auxiliary_loss_clip": 0.0118887,
      "auxiliary_loss_mlp": 0.01034751,
      "balance_loss_clip": 1.05655146,
      "balance_loss_mlp": 1.02635229,
      "epoch": 0.4173630734082847,
      "flos": 25333649320320.0,
      "grad_norm": 2.7242828678192508,
      "language_loss": 0.79337025,
      "learning_rate": 2.6220331875795466e-06,
      "loss": 0.81560647,
      "num_input_tokens_seen": 74869110,
      "step": 3471,
      "time_per_iteration": 2.6653549671173096
    },
    {
      "auxiliary_loss_clip": 0.01192338,
      "auxiliary_loss_mlp": 0.01040683,
      "balance_loss_clip": 1.06040514,
      "balance_loss_mlp": 1.03173614,
      "epoch": 0.4174833162989238,
      "flos": 26685398868480.0,
      "grad_norm": 1.7446934228110114,
      "language_loss": 0.75161892,
      "learning_rate": 2.62129280227276e-06,
      "loss": 0.77394915,
      "num_input_tokens_seen": 74889110,
      "step": 3472,
      "time_per_iteration": 2.635082483291626
    },
    {
      "auxiliary_loss_clip": 0.01193318,
      "auxiliary_loss_mlp": 0.0103398,
      "balance_loss_clip": 1.06054139,
      "balance_loss_mlp": 1.02467537,
      "epoch": 0.41760355918956293,
      "flos": 74739584010240.0,
      "grad_norm": 2.882676446783636,
      "language_loss": 0.68360877,
      "learning_rate": 2.62055232271635e-06,
      "loss": 0.70588171,
      "num_input_tokens_seen": 74916260,
      "step": 3473,
      "time_per_iteration": 3.0407800674438477
    },
    {
      "auxiliary_loss_clip": 0.01192343,
      "auxiliary_loss_mlp": 0.01028194,
      "balance_loss_clip": 1.0597806,
      "balance_loss_mlp": 1.01871085,
      "epoch": 0.417723802080202,
      "flos": 14317513148160.0,
      "grad_norm": 2.1032399937060706,
      "language_loss": 0.87980497,
      "learning_rate": 2.619811749022646e-06,
      "loss": 0.90201038,
      "num_input_tokens_seen": 74931570,
      "step": 3474,
      "time_per_iteration": 2.6185264587402344
    },
    {
      "auxiliary_loss_clip": 0.01195722,
      "auxiliary_loss_mlp": 0.01040033,
      "balance_loss_clip": 1.06188703,
      "balance_loss_mlp": 1.02994812,
      "epoch": 0.4178440449708411,
      "flos": 14643797316480.0,
      "grad_norm": 2.406894312003502,
      "language_loss": 0.71417749,
      "learning_rate": 2.6190710813039917e-06,
      "loss": 0.73653507,
      "num_input_tokens_seen": 74944695,
      "step": 3475,
      "time_per_iteration": 2.792534112930298
    },
    {
      "auxiliary_loss_clip": 0.01191588,
      "auxiliary_loss_mlp": 0.01035504,
      "balance_loss_clip": 1.05709052,
      "balance_loss_mlp": 1.02507353,
      "epoch": 0.4179642878614802,
      "flos": 21507296094720.0,
      "grad_norm": 2.421622879844056,
      "language_loss": 0.83754659,
      "learning_rate": 2.618330319672747e-06,
      "loss": 0.8598175,
      "num_input_tokens_seen": 74964115,
      "step": 3476,
      "time_per_iteration": 2.6707851886749268
    },
    {
      "auxiliary_loss_clip": 0.0119082,
      "auxiliary_loss_mlp": 0.01029697,
      "balance_loss_clip": 1.05813861,
      "balance_loss_mlp": 1.02143538,
      "epoch": 0.41808453075211927,
      "flos": 18441997257600.0,
      "grad_norm": 2.5422910229453692,
      "language_loss": 0.91744339,
      "learning_rate": 2.617589464241284e-06,
      "loss": 0.93964857,
      "num_input_tokens_seen": 74978515,
      "step": 3477,
      "time_per_iteration": 2.6257927417755127
    },
    {
      "auxiliary_loss_clip": 0.01189588,
      "auxiliary_loss_mlp": 0.01033178,
      "balance_loss_clip": 1.05881119,
      "balance_loss_mlp": 1.02451134,
      "epoch": 0.4182047736427584,
      "flos": 20301020628480.0,
      "grad_norm": 1.9203272784999341,
      "language_loss": 0.74301136,
      "learning_rate": 2.6168485151219914e-06,
      "loss": 0.765239,
      "num_input_tokens_seen": 74998135,
      "step": 3478,
      "time_per_iteration": 2.661342144012451
    },
    {
      "auxiliary_loss_clip": 0.01191653,
      "auxiliary_loss_mlp": 0.01038963,
      "balance_loss_clip": 1.05870008,
      "balance_loss_mlp": 1.02874112,
      "epoch": 0.4183250165333975,
      "flos": 18876623823360.0,
      "grad_norm": 2.2631711684846234,
      "language_loss": 0.71466768,
      "learning_rate": 2.616107472427269e-06,
      "loss": 0.73697388,
      "num_input_tokens_seen": 75012830,
      "step": 3479,
      "time_per_iteration": 2.596979856491089
    },
    {
      "auxiliary_loss_clip": 0.01188764,
      "auxiliary_loss_mlp": 0.01025268,
      "balance_loss_clip": 1.05550671,
      "balance_loss_mlp": 1.01597607,
      "epoch": 0.41844525942403654,
      "flos": 17740050698880.0,
      "grad_norm": 2.482767782809504,
      "language_loss": 0.76592743,
      "learning_rate": 2.615366336269533e-06,
      "loss": 0.7880677,
      "num_input_tokens_seen": 75026495,
      "step": 3480,
      "time_per_iteration": 2.621673107147217
    },
    {
      "auxiliary_loss_clip": 0.01193417,
      "auxiliary_loss_mlp": 0.010311,
      "balance_loss_clip": 1.05866194,
      "balance_loss_mlp": 1.02095509,
      "epoch": 0.41856550231467565,
      "flos": 18361377181440.0,
      "grad_norm": 2.5083945469318243,
      "language_loss": 0.80233055,
      "learning_rate": 2.6146251067612126e-06,
      "loss": 0.82457572,
      "num_input_tokens_seen": 75041970,
      "step": 3481,
      "time_per_iteration": 2.5855791568756104
    },
    {
      "auxiliary_loss_clip": 0.01191701,
      "auxiliary_loss_mlp": 0.01031203,
      "balance_loss_clip": 1.05789614,
      "balance_loss_mlp": 1.02278066,
      "epoch": 0.41868574520531476,
      "flos": 22781801445120.0,
      "grad_norm": 1.7748272125103033,
      "language_loss": 0.82904083,
      "learning_rate": 2.6138837840147525e-06,
      "loss": 0.85126984,
      "num_input_tokens_seen": 75061005,
      "step": 3482,
      "time_per_iteration": 2.6383962631225586
    },
    {
      "auxiliary_loss_clip": 0.0118918,
      "auxiliary_loss_mlp": 0.01028202,
      "balance_loss_clip": 1.05574584,
      "balance_loss_mlp": 1.01940417,
      "epoch": 0.4188059880959538,
      "flos": 13699167494400.0,
      "grad_norm": 2.185005035668679,
      "language_loss": 0.76122904,
      "learning_rate": 2.6131423681426103e-06,
      "loss": 0.7834028,
      "num_input_tokens_seen": 75076920,
      "step": 3483,
      "time_per_iteration": 2.6641762256622314
    },
    {
      "auxiliary_loss_clip": 0.01185964,
      "auxiliary_loss_mlp": 0.01028238,
      "balance_loss_clip": 1.05567408,
      "balance_loss_mlp": 1.01944053,
      "epoch": 0.41892623098659293,
      "flos": 37818281220480.0,
      "grad_norm": 1.6622926322129477,
      "language_loss": 0.72864735,
      "learning_rate": 2.6124008592572587e-06,
      "loss": 0.7507894,
      "num_input_tokens_seen": 75100905,
      "step": 3484,
      "time_per_iteration": 2.774610996246338
    },
    {
      "auxiliary_loss_clip": 0.0119135,
      "auxiliary_loss_mlp": 0.01030833,
      "balance_loss_clip": 1.05676484,
      "balance_loss_mlp": 1.02104568,
      "epoch": 0.419046473877232,
      "flos": 23258874908160.0,
      "grad_norm": 2.690664717047782,
      "language_loss": 0.81786907,
      "learning_rate": 2.6116592574711835e-06,
      "loss": 0.84009099,
      "num_input_tokens_seen": 75119205,
      "step": 3485,
      "time_per_iteration": 2.610642671585083
    },
    {
      "auxiliary_loss_clip": 0.01191373,
      "auxiliary_loss_mlp": 0.01036852,
      "balance_loss_clip": 1.05722892,
      "balance_loss_mlp": 1.02685595,
      "epoch": 0.4191667167678711,
      "flos": 20741034234240.0,
      "grad_norm": 1.9069092532920264,
      "language_loss": 0.83960873,
      "learning_rate": 2.6109175628968853e-06,
      "loss": 0.86189097,
      "num_input_tokens_seen": 75138970,
      "step": 3486,
      "time_per_iteration": 2.685246467590332
    },
    {
      "auxiliary_loss_clip": 0.01185659,
      "auxiliary_loss_mlp": 0.0102825,
      "balance_loss_clip": 1.05420256,
      "balance_loss_mlp": 1.01945221,
      "epoch": 0.4192869596585102,
      "flos": 23586416052480.0,
      "grad_norm": 2.0357816516325795,
      "language_loss": 0.82502162,
      "learning_rate": 2.610175775646878e-06,
      "loss": 0.84716076,
      "num_input_tokens_seen": 75157550,
      "step": 3487,
      "time_per_iteration": 2.773300886154175
    },
    {
      "auxiliary_loss_clip": 0.01187238,
      "auxiliary_loss_mlp": 0.01036278,
      "balance_loss_clip": 1.05454087,
      "balance_loss_mlp": 1.02606785,
      "epoch": 0.41940720254914926,
      "flos": 25081269384960.0,
      "grad_norm": 2.953124726026827,
      "language_loss": 0.73412567,
      "learning_rate": 2.6094338958336907e-06,
      "loss": 0.75636083,
      "num_input_tokens_seen": 75176220,
      "step": 3488,
      "time_per_iteration": 3.596184492111206
    },
    {
      "auxiliary_loss_clip": 0.01186586,
      "auxiliary_loss_mlp": 0.010295,
      "balance_loss_clip": 1.05692935,
      "balance_loss_mlp": 1.02073801,
      "epoch": 0.41952744543978837,
      "flos": 15554132628480.0,
      "grad_norm": 3.3322322595800298,
      "language_loss": 0.82279265,
      "learning_rate": 2.608691923569867e-06,
      "loss": 0.84495342,
      "num_input_tokens_seen": 75193095,
      "step": 3489,
      "time_per_iteration": 2.6025404930114746
    },
    {
      "auxiliary_loss_clip": 0.01193049,
      "auxiliary_loss_mlp": 0.01028644,
      "balance_loss_clip": 1.05924988,
      "balance_loss_mlp": 1.01954222,
      "epoch": 0.4196476883304275,
      "flos": 24644775312000.0,
      "grad_norm": 1.6733289658690982,
      "language_loss": 0.75744116,
      "learning_rate": 2.6079498589679616e-06,
      "loss": 0.77965808,
      "num_input_tokens_seen": 75214185,
      "step": 3490,
      "time_per_iteration": 2.632301092147827
    },
    {
      "auxiliary_loss_clip": 0.01193421,
      "auxiliary_loss_mlp": 0.01029987,
      "balance_loss_clip": 1.0580287,
      "balance_loss_mlp": 1.02025366,
      "epoch": 0.41976793122106654,
      "flos": 24531333183360.0,
      "grad_norm": 1.9064204724603433,
      "language_loss": 0.76209211,
      "learning_rate": 2.6072077021405465e-06,
      "loss": 0.7843262,
      "num_input_tokens_seen": 75233020,
      "step": 3491,
      "time_per_iteration": 2.694455146789551
    },
    {
      "auxiliary_loss_clip": 0.01186119,
      "auxiliary_loss_mlp": 0.01031157,
      "balance_loss_clip": 1.05520892,
      "balance_loss_mlp": 1.02280605,
      "epoch": 0.41988817411170565,
      "flos": 21175301664000.0,
      "grad_norm": 2.319161562169994,
      "language_loss": 0.69333625,
      "learning_rate": 2.6064654532002054e-06,
      "loss": 0.71550906,
      "num_input_tokens_seen": 75252030,
      "step": 3492,
      "time_per_iteration": 3.5053462982177734
    },
    {
      "auxiliary_loss_clip": 0.01188743,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.05727172,
      "balance_loss_mlp": 1.01726985,
      "epoch": 0.42000841700234476,
      "flos": 31649402626560.0,
      "grad_norm": 1.433911372786238,
      "language_loss": 0.75914383,
      "learning_rate": 2.6057231122595375e-06,
      "loss": 0.78129089,
      "num_input_tokens_seen": 75273340,
      "step": 3493,
      "time_per_iteration": 3.632870674133301
    },
    {
      "auxiliary_loss_clip": 0.01185754,
      "auxiliary_loss_mlp": 0.01026587,
      "balance_loss_clip": 1.05397964,
      "balance_loss_mlp": 1.01762295,
      "epoch": 0.4201286598929838,
      "flos": 21281525159040.0,
      "grad_norm": 1.7080845045796362,
      "language_loss": 0.7274577,
      "learning_rate": 2.604980679431154e-06,
      "loss": 0.7495811,
      "num_input_tokens_seen": 75291580,
      "step": 3494,
      "time_per_iteration": 2.6323471069335938
    },
    {
      "auxiliary_loss_clip": 0.01186698,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.05416226,
      "balance_loss_mlp": 1.01728892,
      "epoch": 0.4202489027836229,
      "flos": 18546532813440.0,
      "grad_norm": 2.1178722706854467,
      "language_loss": 0.74102265,
      "learning_rate": 2.604238154827684e-06,
      "loss": 0.76315057,
      "num_input_tokens_seen": 75308205,
      "step": 3495,
      "time_per_iteration": 2.5475001335144043
    },
    {
      "auxiliary_loss_clip": 0.01186939,
      "auxiliary_loss_mlp": 0.01029292,
      "balance_loss_clip": 1.05745661,
      "balance_loss_mlp": 1.02060199,
      "epoch": 0.42036914567426203,
      "flos": 19317643009920.0,
      "grad_norm": 2.0156584875475736,
      "language_loss": 0.72746605,
      "learning_rate": 2.6034955385617656e-06,
      "loss": 0.74962831,
      "num_input_tokens_seen": 75326535,
      "step": 3496,
      "time_per_iteration": 2.609637498855591
    },
    {
      "auxiliary_loss_clip": 0.01077183,
      "auxiliary_loss_mlp": 0.01006687,
      "balance_loss_clip": 1.01222289,
      "balance_loss_mlp": 1.00567997,
      "epoch": 0.4204893885649011,
      "flos": 67842942935040.0,
      "grad_norm": 0.7267516217670024,
      "language_loss": 0.61659956,
      "learning_rate": 2.6027528307460544e-06,
      "loss": 0.63743818,
      "num_input_tokens_seen": 75390540,
      "step": 3497,
      "time_per_iteration": 3.298961639404297
    },
    {
      "auxiliary_loss_clip": 0.01187726,
      "auxiliary_loss_mlp": 0.01029116,
      "balance_loss_clip": 1.05497324,
      "balance_loss_mlp": 1.02035999,
      "epoch": 0.4206096314555402,
      "flos": 21908777385600.0,
      "grad_norm": 1.93086659760114,
      "language_loss": 0.86779177,
      "learning_rate": 2.602010031493217e-06,
      "loss": 0.88996023,
      "num_input_tokens_seen": 75408770,
      "step": 3498,
      "time_per_iteration": 2.689462184906006
    },
    {
      "auxiliary_loss_clip": 0.01185787,
      "auxiliary_loss_mlp": 0.0103032,
      "balance_loss_clip": 1.05391932,
      "balance_loss_mlp": 1.02120113,
      "epoch": 0.42072987434617926,
      "flos": 29278185269760.0,
      "grad_norm": 1.9022371300737475,
      "language_loss": 0.869681,
      "learning_rate": 2.6012671409159367e-06,
      "loss": 0.89184201,
      "num_input_tokens_seen": 75430105,
      "step": 3499,
      "time_per_iteration": 2.660930871963501
    },
    {
      "auxiliary_loss_clip": 0.01185405,
      "auxiliary_loss_mlp": 0.01024473,
      "balance_loss_clip": 1.0530448,
      "balance_loss_mlp": 1.0151093,
      "epoch": 0.42085011723681837,
      "flos": 27600726170880.0,
      "grad_norm": 1.7160747414596602,
      "language_loss": 0.81489199,
      "learning_rate": 2.6005241591269097e-06,
      "loss": 0.83699077,
      "num_input_tokens_seen": 75449475,
      "step": 3500,
      "time_per_iteration": 2.6488006114959717
    },
    {
      "auxiliary_loss_clip": 0.01187554,
      "auxiliary_loss_mlp": 0.0102925,
      "balance_loss_clip": 1.05793238,
      "balance_loss_mlp": 1.02132881,
      "epoch": 0.4209703601274575,
      "flos": 27818632028160.0,
      "grad_norm": 1.704313132907032,
      "language_loss": 0.7983501,
      "learning_rate": 2.5997810862388454e-06,
      "loss": 0.82051814,
      "num_input_tokens_seen": 75469315,
      "step": 3501,
      "time_per_iteration": 2.71652889251709
    },
    {
      "auxiliary_loss_clip": 0.01189135,
      "auxiliary_loss_mlp": 0.01031339,
      "balance_loss_clip": 1.05695081,
      "balance_loss_mlp": 1.02243984,
      "epoch": 0.42109060301809653,
      "flos": 27525529048320.0,
      "grad_norm": 2.2622294415810886,
      "language_loss": 0.75888598,
      "learning_rate": 2.599037922364467e-06,
      "loss": 0.7810908,
      "num_input_tokens_seen": 75488215,
      "step": 3502,
      "time_per_iteration": 2.65828275680542
    },
    {
      "auxiliary_loss_clip": 0.01185723,
      "auxiliary_loss_mlp": 0.01028601,
      "balance_loss_clip": 1.05381238,
      "balance_loss_mlp": 1.019732,
      "epoch": 0.42121084590873564,
      "flos": 29314275459840.0,
      "grad_norm": 2.578171961318932,
      "language_loss": 0.75657439,
      "learning_rate": 2.5982946676165112e-06,
      "loss": 0.77871764,
      "num_input_tokens_seen": 75507985,
      "step": 3503,
      "time_per_iteration": 2.6955151557922363
    },
    {
      "auxiliary_loss_clip": 0.01077306,
      "auxiliary_loss_mlp": 0.01001453,
      "balance_loss_clip": 1.0126586,
      "balance_loss_mlp": 1.00049293,
      "epoch": 0.42133108879937475,
      "flos": 67398835178880.0,
      "grad_norm": 0.7302928367033481,
      "language_loss": 0.57598025,
      "learning_rate": 2.5975513221077313e-06,
      "loss": 0.59676784,
      "num_input_tokens_seen": 75571955,
      "step": 3504,
      "time_per_iteration": 3.273569107055664
    },
    {
      "auxiliary_loss_clip": 0.011856,
      "auxiliary_loss_mlp": 0.01028153,
      "balance_loss_clip": 1.05465579,
      "balance_loss_mlp": 1.01943851,
      "epoch": 0.4214513316900138,
      "flos": 23106038538240.0,
      "grad_norm": 2.267966818784064,
      "language_loss": 0.88164556,
      "learning_rate": 2.5968078859508897e-06,
      "loss": 0.90378314,
      "num_input_tokens_seen": 75589155,
      "step": 3505,
      "time_per_iteration": 2.6184146404266357
    },
    {
      "auxiliary_loss_clip": 0.0118422,
      "auxiliary_loss_mlp": 0.01034679,
      "balance_loss_clip": 1.05292726,
      "balance_loss_mlp": 1.02591097,
      "epoch": 0.4215715745806529,
      "flos": 15336190857600.0,
      "grad_norm": 2.16068201063552,
      "language_loss": 0.80326319,
      "learning_rate": 2.5960643592587673e-06,
      "loss": 0.82545221,
      "num_input_tokens_seen": 75606565,
      "step": 3506,
      "time_per_iteration": 2.58591365814209
    },
    {
      "auxiliary_loss_clip": 0.011858,
      "auxiliary_loss_mlp": 0.01026148,
      "balance_loss_clip": 1.05446732,
      "balance_loss_mlp": 1.01774395,
      "epoch": 0.42169181747129203,
      "flos": 22127257860480.0,
      "grad_norm": 2.4462938006307606,
      "language_loss": 0.81519115,
      "learning_rate": 2.5953207421441553e-06,
      "loss": 0.83731061,
      "num_input_tokens_seen": 75625165,
      "step": 3507,
      "time_per_iteration": 2.649350166320801
    },
    {
      "auxiliary_loss_clip": 0.01187363,
      "auxiliary_loss_mlp": 0.0102894,
      "balance_loss_clip": 1.05683076,
      "balance_loss_mlp": 1.02016592,
      "epoch": 0.4218120603619311,
      "flos": 22630724841600.0,
      "grad_norm": 2.3005514356470336,
      "language_loss": 0.75210071,
      "learning_rate": 2.5945770347198603e-06,
      "loss": 0.77426374,
      "num_input_tokens_seen": 75643320,
      "step": 3508,
      "time_per_iteration": 2.6578235626220703
    },
    {
      "auxiliary_loss_clip": 0.01183298,
      "auxiliary_loss_mlp": 0.01028265,
      "balance_loss_clip": 1.05386138,
      "balance_loss_mlp": 1.02029538,
      "epoch": 0.4219323032525702,
      "flos": 19682818629120.0,
      "grad_norm": 1.8228567503304327,
      "language_loss": 0.81971651,
      "learning_rate": 2.593833237098701e-06,
      "loss": 0.84183216,
      "num_input_tokens_seen": 75660920,
      "step": 3509,
      "time_per_iteration": 2.6642467975616455
    },
    {
      "auxiliary_loss_clip": 0.01187975,
      "auxiliary_loss_mlp": 0.01039851,
      "balance_loss_clip": 1.05454612,
      "balance_loss_mlp": 1.03063011,
      "epoch": 0.4220525461432093,
      "flos": 30190747224960.0,
      "grad_norm": 1.9509240673577213,
      "language_loss": 0.62631565,
      "learning_rate": 2.593089349393512e-06,
      "loss": 0.6485939,
      "num_input_tokens_seen": 75681410,
      "step": 3510,
      "time_per_iteration": 2.7263998985290527
    },
    {
      "auxiliary_loss_clip": 0.0118948,
      "auxiliary_loss_mlp": 0.01035949,
      "balance_loss_clip": 1.05590045,
      "balance_loss_mlp": 1.026824,
      "epoch": 0.42217278903384836,
      "flos": 24315941278080.0,
      "grad_norm": 3.26651733011721,
      "language_loss": 0.83749962,
      "learning_rate": 2.592345371717141e-06,
      "loss": 0.85975397,
      "num_input_tokens_seen": 75700940,
      "step": 3511,
      "time_per_iteration": 2.6659462451934814
    },
    {
      "auxiliary_loss_clip": 0.01191182,
      "auxiliary_loss_mlp": 0.01040481,
      "balance_loss_clip": 1.05938339,
      "balance_loss_mlp": 1.03061652,
      "epoch": 0.42229303192448747,
      "flos": 17092474352640.0,
      "grad_norm": 1.9493702361273062,
      "language_loss": 0.72222459,
      "learning_rate": 2.591601304182448e-06,
      "loss": 0.74454123,
      "num_input_tokens_seen": 75718910,
      "step": 3512,
      "time_per_iteration": 2.5971221923828125
    },
    {
      "auxiliary_loss_clip": 0.01186213,
      "auxiliary_loss_mlp": 0.01029569,
      "balance_loss_clip": 1.05721438,
      "balance_loss_mlp": 1.02115846,
      "epoch": 0.4224132748151266,
      "flos": 22784530878720.0,
      "grad_norm": 1.7203837052946238,
      "language_loss": 0.79059482,
      "learning_rate": 2.5908571469023067e-06,
      "loss": 0.8127526,
      "num_input_tokens_seen": 75738395,
      "step": 3513,
      "time_per_iteration": 2.627725839614868
    },
    {
      "auxiliary_loss_clip": 0.01186673,
      "auxiliary_loss_mlp": 0.0103895,
      "balance_loss_clip": 1.05516171,
      "balance_loss_mlp": 1.02969289,
      "epoch": 0.42253351770576564,
      "flos": 17819090576640.0,
      "grad_norm": 2.8155853547052963,
      "language_loss": 0.75909138,
      "learning_rate": 2.5901128999896067e-06,
      "loss": 0.78134757,
      "num_input_tokens_seen": 75753825,
      "step": 3514,
      "time_per_iteration": 2.5583040714263916
    },
    {
      "auxiliary_loss_clip": 0.01186151,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.05483806,
      "balance_loss_mlp": 1.0201695,
      "epoch": 0.42265376059640475,
      "flos": 28512390286080.0,
      "grad_norm": 1.7807035998089011,
      "language_loss": 0.68333328,
      "learning_rate": 2.5893685635572487e-06,
      "loss": 0.70547807,
      "num_input_tokens_seen": 75774675,
      "step": 3515,
      "time_per_iteration": 3.5554709434509277
    },
    {
      "auxiliary_loss_clip": 0.01186262,
      "auxiliary_loss_mlp": 0.01026795,
      "balance_loss_clip": 1.05489004,
      "balance_loss_mlp": 1.01729441,
      "epoch": 0.4227740034870438,
      "flos": 16253349753600.0,
      "grad_norm": 2.5511454716078923,
      "language_loss": 0.69031906,
      "learning_rate": 2.5886241377181483e-06,
      "loss": 0.71244967,
      "num_input_tokens_seen": 75793545,
      "step": 3516,
      "time_per_iteration": 2.6152350902557373
    },
    {
      "auxiliary_loss_clip": 0.01192207,
      "auxiliary_loss_mlp": 0.01029003,
      "balance_loss_clip": 1.05777431,
      "balance_loss_mlp": 1.01938868,
      "epoch": 0.4228942463776829,
      "flos": 25295691623040.0,
      "grad_norm": 2.803741186292814,
      "language_loss": 0.80874383,
      "learning_rate": 2.587879622585234e-06,
      "loss": 0.83095586,
      "num_input_tokens_seen": 75812145,
      "step": 3517,
      "time_per_iteration": 2.6788010597229004
    },
    {
      "auxiliary_loss_clip": 0.01186285,
      "auxiliary_loss_mlp": 0.01035312,
      "balance_loss_clip": 1.05589342,
      "balance_loss_mlp": 1.02615082,
      "epoch": 0.423014489268322,
      "flos": 26395779507840.0,
      "grad_norm": 2.0416785781485536,
      "language_loss": 0.76084447,
      "learning_rate": 2.5871350182714486e-06,
      "loss": 0.78306043,
      "num_input_tokens_seen": 75833025,
      "step": 3518,
      "time_per_iteration": 3.6226959228515625
    },
    {
      "auxiliary_loss_clip": 0.01184628,
      "auxiliary_loss_mlp": 0.01028449,
      "balance_loss_clip": 1.05518758,
      "balance_loss_mlp": 1.02072978,
      "epoch": 0.4231347321589611,
      "flos": 17274002711040.0,
      "grad_norm": 2.3250938796059915,
      "language_loss": 0.80410969,
      "learning_rate": 2.586390324889748e-06,
      "loss": 0.82624042,
      "num_input_tokens_seen": 75848925,
      "step": 3519,
      "time_per_iteration": 2.5989301204681396
    },
    {
      "auxiliary_loss_clip": 0.01188178,
      "auxiliary_loss_mlp": 0.01031512,
      "balance_loss_clip": 1.05620062,
      "balance_loss_mlp": 1.02264881,
      "epoch": 0.4232549750496002,
      "flos": 22999635475200.0,
      "grad_norm": 1.7706333312057152,
      "language_loss": 0.67418009,
      "learning_rate": 2.5856455425531003e-06,
      "loss": 0.69637698,
      "num_input_tokens_seen": 75870400,
      "step": 3520,
      "time_per_iteration": 3.601487398147583
    },
    {
      "auxiliary_loss_clip": 0.0118703,
      "auxiliary_loss_mlp": 0.01027724,
      "balance_loss_clip": 1.05706191,
      "balance_loss_mlp": 1.01928425,
      "epoch": 0.4233752179402393,
      "flos": 21248343970560.0,
      "grad_norm": 1.9355759895611029,
      "language_loss": 0.80734992,
      "learning_rate": 2.5849006713744902e-06,
      "loss": 0.82949752,
      "num_input_tokens_seen": 75889195,
      "step": 3521,
      "time_per_iteration": 2.7208917140960693
    },
    {
      "auxiliary_loss_clip": 0.01191817,
      "auxiliary_loss_mlp": 0.01026509,
      "balance_loss_clip": 1.05934739,
      "balance_loss_mlp": 1.01756835,
      "epoch": 0.42349546083087836,
      "flos": 20704297599360.0,
      "grad_norm": 2.2824772984540944,
      "language_loss": 0.73177373,
      "learning_rate": 2.5841557114669135e-06,
      "loss": 0.75395703,
      "num_input_tokens_seen": 75906055,
      "step": 3522,
      "time_per_iteration": 2.5961661338806152
    },
    {
      "auxiliary_loss_clip": 0.01188997,
      "auxiliary_loss_mlp": 0.01029188,
      "balance_loss_clip": 1.05464506,
      "balance_loss_mlp": 1.01904881,
      "epoch": 0.42361570372151747,
      "flos": 18585065128320.0,
      "grad_norm": 2.937822074680316,
      "language_loss": 0.68014616,
      "learning_rate": 2.58341066294338e-06,
      "loss": 0.70232797,
      "num_input_tokens_seen": 75922720,
      "step": 3523,
      "time_per_iteration": 2.6102356910705566
    },
    {
      "auxiliary_loss_clip": 0.01191101,
      "auxiliary_loss_mlp": 0.01032695,
      "balance_loss_clip": 1.06025267,
      "balance_loss_mlp": 1.02423692,
      "epoch": 0.4237359466121566,
      "flos": 20959478795520.0,
      "grad_norm": 2.2571494974630006,
      "language_loss": 0.85251784,
      "learning_rate": 2.5826655259169124e-06,
      "loss": 0.87475586,
      "num_input_tokens_seen": 75941375,
      "step": 3524,
      "time_per_iteration": 2.6425819396972656
    },
    {
      "auxiliary_loss_clip": 0.01190406,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.05991411,
      "balance_loss_mlp": 1.02017426,
      "epoch": 0.42385618950279563,
      "flos": 18038181582720.0,
      "grad_norm": 8.425411828221387,
      "language_loss": 0.90594864,
      "learning_rate": 2.5819203005005475e-06,
      "loss": 0.92813706,
      "num_input_tokens_seen": 75958710,
      "step": 3525,
      "time_per_iteration": 2.668330192565918
    },
    {
      "auxiliary_loss_clip": 0.01182945,
      "auxiliary_loss_mlp": 0.01028001,
      "balance_loss_clip": 1.05358195,
      "balance_loss_mlp": 1.01951277,
      "epoch": 0.42397643239343474,
      "flos": 23769129559680.0,
      "grad_norm": 1.6585295813425522,
      "language_loss": 0.78603256,
      "learning_rate": 2.581174986807336e-06,
      "loss": 0.80814195,
      "num_input_tokens_seen": 75978945,
      "step": 3526,
      "time_per_iteration": 2.5948379039764404
    },
    {
      "auxiliary_loss_clip": 0.01186852,
      "auxiliary_loss_mlp": 0.01030354,
      "balance_loss_clip": 1.0558449,
      "balance_loss_mlp": 1.02244461,
      "epoch": 0.42409667528407385,
      "flos": 16545088016640.0,
      "grad_norm": 2.1798048145510975,
      "language_loss": 0.9142189,
      "learning_rate": 2.580429584950341e-06,
      "loss": 0.936391,
      "num_input_tokens_seen": 75994695,
      "step": 3527,
      "time_per_iteration": 2.6519618034362793
    },
    {
      "auxiliary_loss_clip": 0.011886,
      "auxiliary_loss_mlp": 0.01030729,
      "balance_loss_clip": 1.05561781,
      "balance_loss_mlp": 1.02128172,
      "epoch": 0.4242169181747129,
      "flos": 16034186920320.0,
      "grad_norm": 2.271043676410259,
      "language_loss": 0.66451406,
      "learning_rate": 2.5796840950426397e-06,
      "loss": 0.68670726,
      "num_input_tokens_seen": 76011780,
      "step": 3528,
      "time_per_iteration": 2.583564043045044
    },
    {
      "auxiliary_loss_clip": 0.01189447,
      "auxiliary_loss_mlp": 0.01027134,
      "balance_loss_clip": 1.05827332,
      "balance_loss_mlp": 1.01873016,
      "epoch": 0.424337161065352,
      "flos": 20084012611200.0,
      "grad_norm": 1.7404879793155414,
      "language_loss": 0.65613127,
      "learning_rate": 2.578938517197322e-06,
      "loss": 0.6782971,
      "num_input_tokens_seen": 76029875,
      "step": 3529,
      "time_per_iteration": 2.6364991664886475
    },
    {
      "auxiliary_loss_clip": 0.01185302,
      "auxiliary_loss_mlp": 0.01027649,
      "balance_loss_clip": 1.05680013,
      "balance_loss_mlp": 1.01877987,
      "epoch": 0.4244574039559911,
      "flos": 23878369797120.0,
      "grad_norm": 2.4034184090482578,
      "language_loss": 0.62837642,
      "learning_rate": 2.5781928515274916e-06,
      "loss": 0.6505059,
      "num_input_tokens_seen": 76048595,
      "step": 3530,
      "time_per_iteration": 2.635011911392212
    },
    {
      "auxiliary_loss_clip": 0.01189253,
      "auxiliary_loss_mlp": 0.01025549,
      "balance_loss_clip": 1.05834675,
      "balance_loss_mlp": 1.01670396,
      "epoch": 0.4245776468466302,
      "flos": 17565920542080.0,
      "grad_norm": 1.9420184404317427,
      "language_loss": 0.67797476,
      "learning_rate": 2.577447098146265e-06,
      "loss": 0.70012277,
      "num_input_tokens_seen": 76065770,
      "step": 3531,
      "time_per_iteration": 2.588550329208374
    },
    {
      "auxiliary_loss_clip": 0.01188433,
      "auxiliary_loss_mlp": 0.01022878,
      "balance_loss_clip": 1.05759799,
      "balance_loss_mlp": 1.01459265,
      "epoch": 0.4246978897372693,
      "flos": 27776256958080.0,
      "grad_norm": 1.9939114555753452,
      "language_loss": 0.79251361,
      "learning_rate": 2.5767012571667724e-06,
      "loss": 0.81462669,
      "num_input_tokens_seen": 76085250,
      "step": 3532,
      "time_per_iteration": 2.6679484844207764
    },
    {
      "auxiliary_loss_clip": 0.01190027,
      "auxiliary_loss_mlp": 0.01030838,
      "balance_loss_clip": 1.05642295,
      "balance_loss_mlp": 1.02220082,
      "epoch": 0.42481813262790835,
      "flos": 15596615439360.0,
      "grad_norm": 2.122001244515103,
      "language_loss": 0.68446624,
      "learning_rate": 2.5759553287021587e-06,
      "loss": 0.70667493,
      "num_input_tokens_seen": 76103580,
      "step": 3533,
      "time_per_iteration": 2.696516513824463
    },
    {
      "auxiliary_loss_clip": 0.01190099,
      "auxiliary_loss_mlp": 0.01033809,
      "balance_loss_clip": 1.05781388,
      "balance_loss_mlp": 1.02493358,
      "epoch": 0.42493837551854746,
      "flos": 23951088881280.0,
      "grad_norm": 1.9747526374288324,
      "language_loss": 0.77753466,
      "learning_rate": 2.5752093128655786e-06,
      "loss": 0.79977375,
      "num_input_tokens_seen": 76121825,
      "step": 3534,
      "time_per_iteration": 2.658592700958252
    },
    {
      "auxiliary_loss_clip": 0.0118597,
      "auxiliary_loss_mlp": 0.01030528,
      "balance_loss_clip": 1.05586648,
      "balance_loss_mlp": 1.02146852,
      "epoch": 0.4250586184091866,
      "flos": 20813466009600.0,
      "grad_norm": 2.0211459932427553,
      "language_loss": 0.74052417,
      "learning_rate": 2.574463209770204e-06,
      "loss": 0.76268911,
      "num_input_tokens_seen": 76141140,
      "step": 3535,
      "time_per_iteration": 2.609804153442383
    },
    {
      "auxiliary_loss_clip": 0.01191233,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 1.05816722,
      "balance_loss_mlp": 1.01828182,
      "epoch": 0.42517886129982563,
      "flos": 30371018607360.0,
      "grad_norm": 2.8927105204462156,
      "language_loss": 0.79830408,
      "learning_rate": 2.5737170195292165e-06,
      "loss": 0.82048666,
      "num_input_tokens_seen": 76164475,
      "step": 3536,
      "time_per_iteration": 2.752197265625
    },
    {
      "auxiliary_loss_clip": 0.01189245,
      "auxiliary_loss_mlp": 0.01035271,
      "balance_loss_clip": 1.05617321,
      "balance_loss_mlp": 1.02656877,
      "epoch": 0.42529910419046474,
      "flos": 20080636732800.0,
      "grad_norm": 2.763725242464535,
      "language_loss": 0.7799086,
      "learning_rate": 2.572970742255814e-06,
      "loss": 0.80215371,
      "num_input_tokens_seen": 76182965,
      "step": 3537,
      "time_per_iteration": 2.5792882442474365
    },
    {
      "auxiliary_loss_clip": 0.01186839,
      "auxiliary_loss_mlp": 0.01029581,
      "balance_loss_clip": 1.05659735,
      "balance_loss_mlp": 1.02139759,
      "epoch": 0.42541934708110385,
      "flos": 22632448694400.0,
      "grad_norm": 1.873931280328867,
      "language_loss": 0.81551844,
      "learning_rate": 2.5722243780632046e-06,
      "loss": 0.83768266,
      "num_input_tokens_seen": 76201230,
      "step": 3538,
      "time_per_iteration": 2.6806588172912598
    },
    {
      "auxiliary_loss_clip": 0.01079685,
      "auxiliary_loss_mlp": 0.01003288,
      "balance_loss_clip": 1.01511002,
      "balance_loss_mlp": 1.00217366,
      "epoch": 0.4255395899717429,
      "flos": 66200676186240.0,
      "grad_norm": 0.7598229072191545,
      "language_loss": 0.60469323,
      "learning_rate": 2.5714779270646125e-06,
      "loss": 0.62552303,
      "num_input_tokens_seen": 76262000,
      "step": 3539,
      "time_per_iteration": 3.1194238662719727
    },
    {
      "auxiliary_loss_clip": 0.01189938,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.05695629,
      "balance_loss_mlp": 1.01723933,
      "epoch": 0.425659832862382,
      "flos": 17931814433280.0,
      "grad_norm": 2.2477919616908184,
      "language_loss": 0.77963924,
      "learning_rate": 2.5707313893732735e-06,
      "loss": 0.80180454,
      "num_input_tokens_seen": 76280540,
      "step": 3540,
      "time_per_iteration": 2.614143133163452
    },
    {
      "auxiliary_loss_clip": 0.01189886,
      "auxiliary_loss_mlp": 0.01028051,
      "balance_loss_clip": 1.05673647,
      "balance_loss_mlp": 1.01973605,
      "epoch": 0.4257800757530211,
      "flos": 24022550989440.0,
      "grad_norm": 1.7335953298440492,
      "language_loss": 0.77149141,
      "learning_rate": 2.5699847651024364e-06,
      "loss": 0.79367077,
      "num_input_tokens_seen": 76301180,
      "step": 3541,
      "time_per_iteration": 3.5981452465057373
    },
    {
      "auxiliary_loss_clip": 0.01187368,
      "auxiliary_loss_mlp": 0.01031838,
      "balance_loss_clip": 1.05820727,
      "balance_loss_mlp": 1.02358878,
      "epoch": 0.4259003186436602,
      "flos": 23696015425920.0,
      "grad_norm": 2.233949720968532,
      "language_loss": 0.77321756,
      "learning_rate": 2.5692380543653627e-06,
      "loss": 0.79540956,
      "num_input_tokens_seen": 76319335,
      "step": 3542,
      "time_per_iteration": 2.602423906326294
    },
    {
      "auxiliary_loss_clip": 0.01190001,
      "auxiliary_loss_mlp": 0.01029996,
      "balance_loss_clip": 1.05793417,
      "balance_loss_mlp": 1.02060843,
      "epoch": 0.4260205615342993,
      "flos": 15259772672640.0,
      "grad_norm": 2.3930739820621714,
      "language_loss": 0.70149171,
      "learning_rate": 2.5684912572753293e-06,
      "loss": 0.72369164,
      "num_input_tokens_seen": 76335010,
      "step": 3543,
      "time_per_iteration": 2.59236216545105
    },
    {
      "auxiliary_loss_clip": 0.01182372,
      "auxiliary_loss_mlp": 0.01026828,
      "balance_loss_clip": 1.05357122,
      "balance_loss_mlp": 1.01841223,
      "epoch": 0.4261408044249384,
      "flos": 30665306736000.0,
      "grad_norm": 2.09928606426383,
      "language_loss": 0.84017885,
      "learning_rate": 2.5677443739456245e-06,
      "loss": 0.86227083,
      "num_input_tokens_seen": 76356670,
      "step": 3544,
      "time_per_iteration": 3.636901617050171
    },
    {
      "auxiliary_loss_clip": 0.0118769,
      "auxiliary_loss_mlp": 0.01031526,
      "balance_loss_clip": 1.05640984,
      "balance_loss_mlp": 1.02278233,
      "epoch": 0.42626104731557746,
      "flos": 23257905240960.0,
      "grad_norm": 2.8292092326950335,
      "language_loss": 0.79940432,
      "learning_rate": 2.5669974044895495e-06,
      "loss": 0.8215965,
      "num_input_tokens_seen": 76373065,
      "step": 3545,
      "time_per_iteration": 2.599322557449341
    },
    {
      "auxiliary_loss_clip": 0.0119041,
      "auxiliary_loss_mlp": 0.01028093,
      "balance_loss_clip": 1.05854988,
      "balance_loss_mlp": 1.01965272,
      "epoch": 0.42638129020621657,
      "flos": 25884770670720.0,
      "grad_norm": 2.1801798437552384,
      "language_loss": 0.79657567,
      "learning_rate": 2.5662503490204187e-06,
      "loss": 0.81876075,
      "num_input_tokens_seen": 76393230,
      "step": 3546,
      "time_per_iteration": 4.449785470962524
    },
    {
      "auxiliary_loss_clip": 0.01189535,
      "auxiliary_loss_mlp": 0.01030617,
      "balance_loss_clip": 1.05661345,
      "balance_loss_mlp": 1.02202201,
      "epoch": 0.4265015330968556,
      "flos": 26502362138880.0,
      "grad_norm": 1.9874033852775417,
      "language_loss": 0.76374465,
      "learning_rate": 2.5655032076515603e-06,
      "loss": 0.78594619,
      "num_input_tokens_seen": 76412555,
      "step": 3547,
      "time_per_iteration": 2.6657426357269287
    },
    {
      "auxiliary_loss_clip": 0.01185317,
      "auxiliary_loss_mlp": 0.01032113,
      "balance_loss_clip": 1.05536532,
      "balance_loss_mlp": 1.02305913,
      "epoch": 0.42662177598749473,
      "flos": 24389522288640.0,
      "grad_norm": 2.108117600801587,
      "language_loss": 0.82074296,
      "learning_rate": 2.5647559804963155e-06,
      "loss": 0.84291732,
      "num_input_tokens_seen": 76432485,
      "step": 3548,
      "time_per_iteration": 2.6783716678619385
    },
    {
      "auxiliary_loss_clip": 0.01192215,
      "auxiliary_loss_mlp": 0.01032205,
      "balance_loss_clip": 1.05899036,
      "balance_loss_mlp": 1.02274597,
      "epoch": 0.42674201887813384,
      "flos": 23148629089920.0,
      "grad_norm": 1.9478730801875468,
      "language_loss": 0.78624654,
      "learning_rate": 2.5640086676680364e-06,
      "loss": 0.80849069,
      "num_input_tokens_seen": 76453980,
      "step": 3549,
      "time_per_iteration": 2.6668615341186523
    },
    {
      "auxiliary_loss_clip": 0.0119016,
      "auxiliary_loss_mlp": 0.01030291,
      "balance_loss_clip": 1.05763698,
      "balance_loss_mlp": 1.02096844,
      "epoch": 0.4268622617687729,
      "flos": 21689614552320.0,
      "grad_norm": 2.349504877371345,
      "language_loss": 0.80736381,
      "learning_rate": 2.5632612692800923e-06,
      "loss": 0.82956827,
      "num_input_tokens_seen": 76473045,
      "step": 3550,
      "time_per_iteration": 2.670011043548584
    },
    {
      "auxiliary_loss_clip": 0.01193502,
      "auxiliary_loss_mlp": 0.01032605,
      "balance_loss_clip": 1.05859351,
      "balance_loss_mlp": 1.02234077,
      "epoch": 0.426982504659412,
      "flos": 23440151871360.0,
      "grad_norm": 1.9403166962448286,
      "language_loss": 0.75364375,
      "learning_rate": 2.5625137854458603e-06,
      "loss": 0.77590477,
      "num_input_tokens_seen": 76492060,
      "step": 3551,
      "time_per_iteration": 2.607088327407837
    },
    {
      "auxiliary_loss_clip": 0.01186515,
      "auxiliary_loss_mlp": 0.01026417,
      "balance_loss_clip": 1.0574882,
      "balance_loss_mlp": 1.0185194,
      "epoch": 0.4271027475500511,
      "flos": 18916556768640.0,
      "grad_norm": 1.9151716477821323,
      "language_loss": 0.80084366,
      "learning_rate": 2.561766216278735e-06,
      "loss": 0.82297295,
      "num_input_tokens_seen": 76509655,
      "step": 3552,
      "time_per_iteration": 2.5931508541107178
    },
    {
      "auxiliary_loss_clip": 0.01192415,
      "auxiliary_loss_mlp": 0.01028078,
      "balance_loss_clip": 1.06110299,
      "balance_loss_mlp": 1.01878583,
      "epoch": 0.4272229904406902,
      "flos": 26870554500480.0,
      "grad_norm": 1.9681891668051255,
      "language_loss": 0.8109073,
      "learning_rate": 2.561018561892121e-06,
      "loss": 0.83311224,
      "num_input_tokens_seen": 76528795,
      "step": 3553,
      "time_per_iteration": 2.6629791259765625
    },
    {
      "auxiliary_loss_clip": 0.01185105,
      "auxiliary_loss_mlp": 0.01023267,
      "balance_loss_clip": 1.05355918,
      "balance_loss_mlp": 1.01445723,
      "epoch": 0.4273432333313293,
      "flos": 23951376190080.0,
      "grad_norm": 1.7231781923566185,
      "language_loss": 0.76267362,
      "learning_rate": 2.5602708223994363e-06,
      "loss": 0.78475732,
      "num_input_tokens_seen": 76550660,
      "step": 3554,
      "time_per_iteration": 2.6345388889312744
    },
    {
      "auxiliary_loss_clip": 0.01185265,
      "auxiliary_loss_mlp": 0.01028692,
      "balance_loss_clip": 1.05439866,
      "balance_loss_mlp": 1.01993024,
      "epoch": 0.4274634762219684,
      "flos": 29570354496000.0,
      "grad_norm": 2.5639929427300348,
      "language_loss": 0.68139446,
      "learning_rate": 2.559522997914115e-06,
      "loss": 0.70353401,
      "num_input_tokens_seen": 76570240,
      "step": 3555,
      "time_per_iteration": 2.715611219406128
    },
    {
      "auxiliary_loss_clip": 0.01187216,
      "auxiliary_loss_mlp": 0.01029513,
      "balance_loss_clip": 1.05859017,
      "balance_loss_mlp": 1.02116215,
      "epoch": 0.42758371911260745,
      "flos": 21434146047360.0,
      "grad_norm": 2.216882894502074,
      "language_loss": 0.84564942,
      "learning_rate": 2.558775088549599e-06,
      "loss": 0.86781669,
      "num_input_tokens_seen": 76589820,
      "step": 3556,
      "time_per_iteration": 2.61592960357666
    },
    {
      "auxiliary_loss_clip": 0.01191693,
      "auxiliary_loss_mlp": 0.01035133,
      "balance_loss_clip": 1.05834734,
      "balance_loss_mlp": 1.02621579,
      "epoch": 0.42770396200324656,
      "flos": 14752822072320.0,
      "grad_norm": 2.3212191235829414,
      "language_loss": 0.66430342,
      "learning_rate": 2.5580270944193467e-06,
      "loss": 0.6865716,
      "num_input_tokens_seen": 76606640,
      "step": 3557,
      "time_per_iteration": 2.6251676082611084
    },
    {
      "auxiliary_loss_clip": 0.01080573,
      "auxiliary_loss_mlp": 0.01003201,
      "balance_loss_clip": 1.01601219,
      "balance_loss_mlp": 1.00204456,
      "epoch": 0.4278242048938857,
      "flos": 70654712601600.0,
      "grad_norm": 0.7460470424893635,
      "language_loss": 0.55531454,
      "learning_rate": 2.557279015636827e-06,
      "loss": 0.57615221,
      "num_input_tokens_seen": 76667050,
      "step": 3558,
      "time_per_iteration": 3.17830491065979
    },
    {
      "auxiliary_loss_clip": 0.01080423,
      "auxiliary_loss_mlp": 0.01002303,
      "balance_loss_clip": 1.01589155,
      "balance_loss_mlp": 1.00113475,
      "epoch": 0.42794444778452473,
      "flos": 69366165033600.0,
      "grad_norm": 0.7664431407853612,
      "language_loss": 0.61203176,
      "learning_rate": 2.5565308523155245e-06,
      "loss": 0.63285899,
      "num_input_tokens_seen": 76726650,
      "step": 3559,
      "time_per_iteration": 3.1474192142486572
    },
    {
      "auxiliary_loss_clip": 0.01190001,
      "auxiliary_loss_mlp": 0.0102798,
      "balance_loss_clip": 1.05906343,
      "balance_loss_mlp": 1.01840115,
      "epoch": 0.42806469067516384,
      "flos": 18215328481920.0,
      "grad_norm": 2.6754994352565884,
      "language_loss": 0.8247683,
      "learning_rate": 2.5557826045689336e-06,
      "loss": 0.84694815,
      "num_input_tokens_seen": 76742890,
      "step": 3560,
      "time_per_iteration": 2.6292057037353516
    },
    {
      "auxiliary_loss_clip": 0.01080123,
      "auxiliary_loss_mlp": 0.01001619,
      "balance_loss_clip": 1.01577044,
      "balance_loss_mlp": 1.00046301,
      "epoch": 0.4281849335658029,
      "flos": 54535814432640.0,
      "grad_norm": 0.8159135027821586,
      "language_loss": 0.58782244,
      "learning_rate": 2.5550342725105643e-06,
      "loss": 0.6086399,
      "num_input_tokens_seen": 76801055,
      "step": 3561,
      "time_per_iteration": 3.188852548599243
    },
    {
      "auxiliary_loss_clip": 0.01192704,
      "auxiliary_loss_mlp": 0.01029862,
      "balance_loss_clip": 1.05995905,
      "balance_loss_mlp": 1.02095735,
      "epoch": 0.428305176456442,
      "flos": 17274828723840.0,
      "grad_norm": 2.059896838694362,
      "language_loss": 0.80888677,
      "learning_rate": 2.554285856253937e-06,
      "loss": 0.83111238,
      "num_input_tokens_seen": 76819890,
      "step": 3562,
      "time_per_iteration": 2.7027664184570312
    },
    {
      "auxiliary_loss_clip": 0.01188356,
      "auxiliary_loss_mlp": 0.01030719,
      "balance_loss_clip": 1.05735695,
      "balance_loss_mlp": 1.02219605,
      "epoch": 0.4284254193470811,
      "flos": 26359509749760.0,
      "grad_norm": 1.951057222257977,
      "language_loss": 0.7804935,
      "learning_rate": 2.5535373559125855e-06,
      "loss": 0.80268419,
      "num_input_tokens_seen": 76840255,
      "step": 3563,
      "time_per_iteration": 2.671109199523926
    },
    {
      "auxiliary_loss_clip": 0.01189195,
      "auxiliary_loss_mlp": 0.01038116,
      "balance_loss_clip": 1.05643415,
      "balance_loss_mlp": 1.02871609,
      "epoch": 0.42854566223772017,
      "flos": 29714248379520.0,
      "grad_norm": 1.7382319459240474,
      "language_loss": 0.82092285,
      "learning_rate": 2.552788771600057e-06,
      "loss": 0.84319592,
      "num_input_tokens_seen": 76860565,
      "step": 3564,
      "time_per_iteration": 2.711153030395508
    },
    {
      "auxiliary_loss_clip": 0.01194778,
      "auxiliary_loss_mlp": 0.01031978,
      "balance_loss_clip": 1.06152368,
      "balance_loss_mlp": 1.02326417,
      "epoch": 0.4286659051283593,
      "flos": 22018161277440.0,
      "grad_norm": 1.958498725247404,
      "language_loss": 0.82438123,
      "learning_rate": 2.5520401034299118e-06,
      "loss": 0.84664875,
      "num_input_tokens_seen": 76878325,
      "step": 3565,
      "time_per_iteration": 2.681713104248047
    },
    {
      "auxiliary_loss_clip": 0.01189618,
      "auxiliary_loss_mlp": 0.0102768,
      "balance_loss_clip": 1.0546124,
      "balance_loss_mlp": 1.0176425,
      "epoch": 0.4287861480189984,
      "flos": 13334422838400.0,
      "grad_norm": 2.157236847965602,
      "language_loss": 0.87629294,
      "learning_rate": 2.551291351515722e-06,
      "loss": 0.89846599,
      "num_input_tokens_seen": 76895340,
      "step": 3566,
      "time_per_iteration": 2.6410880088806152
    },
    {
      "auxiliary_loss_clip": 0.01184444,
      "auxiliary_loss_mlp": 0.0102528,
      "balance_loss_clip": 1.05355573,
      "balance_loss_mlp": 1.01664925,
      "epoch": 0.42890639090963745,
      "flos": 26651535321600.0,
      "grad_norm": 1.6587821357880554,
      "language_loss": 0.85865283,
      "learning_rate": 2.5505425159710726e-06,
      "loss": 0.88075006,
      "num_input_tokens_seen": 76915150,
      "step": 3567,
      "time_per_iteration": 3.6143124103546143
    },
    {
      "auxiliary_loss_clip": 0.01191584,
      "auxiliary_loss_mlp": 0.0102634,
      "balance_loss_clip": 1.0588522,
      "balance_loss_mlp": 1.01635039,
      "epoch": 0.42902663380027656,
      "flos": 24055768091520.0,
      "grad_norm": 2.403449583806297,
      "language_loss": 0.8252821,
      "learning_rate": 2.549793596909561e-06,
      "loss": 0.84746128,
      "num_input_tokens_seen": 76933770,
      "step": 3568,
      "time_per_iteration": 2.7147903442382812
    },
    {
      "auxiliary_loss_clip": 0.01186504,
      "auxiliary_loss_mlp": 0.01028727,
      "balance_loss_clip": 1.05460358,
      "balance_loss_mlp": 1.02011991,
      "epoch": 0.42914687669091567,
      "flos": 15632561975040.0,
      "grad_norm": 3.2780756438086813,
      "language_loss": 0.66170555,
      "learning_rate": 2.5490445944447976e-06,
      "loss": 0.68385786,
      "num_input_tokens_seen": 76952265,
      "step": 3569,
      "time_per_iteration": 2.6039509773254395
    },
    {
      "auxiliary_loss_clip": 0.01189605,
      "auxiliary_loss_mlp": 0.01028378,
      "balance_loss_clip": 1.05737031,
      "balance_loss_mlp": 1.01953912,
      "epoch": 0.4292671195815547,
      "flos": 31467802440960.0,
      "grad_norm": 2.5885833783929013,
      "language_loss": 0.6536172,
      "learning_rate": 2.548295508690406e-06,
      "loss": 0.67579705,
      "num_input_tokens_seen": 76973560,
      "step": 3570,
      "time_per_iteration": 2.702125310897827
    },
    {
      "auxiliary_loss_clip": 0.01188917,
      "auxiliary_loss_mlp": 0.01032624,
      "balance_loss_clip": 1.05701423,
      "balance_loss_mlp": 1.02364194,
      "epoch": 0.42938736247219383,
      "flos": 30257756046720.0,
      "grad_norm": 1.8121504180809562,
      "language_loss": 0.76572144,
      "learning_rate": 2.5475463397600217e-06,
      "loss": 0.78793681,
      "num_input_tokens_seen": 76993640,
      "step": 3571,
      "time_per_iteration": 3.554278612136841
    },
    {
      "auxiliary_loss_clip": 0.01194245,
      "auxiliary_loss_mlp": 0.01034054,
      "balance_loss_clip": 1.06028795,
      "balance_loss_mlp": 1.02436817,
      "epoch": 0.42950760536283294,
      "flos": 29349683291520.0,
      "grad_norm": 2.7027756522588215,
      "language_loss": 0.77336836,
      "learning_rate": 2.546797087767293e-06,
      "loss": 0.79565132,
      "num_input_tokens_seen": 77013765,
      "step": 3572,
      "time_per_iteration": 3.566810369491577
    },
    {
      "auxiliary_loss_clip": 0.01187014,
      "auxiliary_loss_mlp": 0.01029956,
      "balance_loss_clip": 1.05581141,
      "balance_loss_mlp": 1.02094364,
      "epoch": 0.429627848253472,
      "flos": 26869943969280.0,
      "grad_norm": 4.413990422319219,
      "language_loss": 0.87428468,
      "learning_rate": 2.546047752825881e-06,
      "loss": 0.89645445,
      "num_input_tokens_seen": 77034370,
      "step": 3573,
      "time_per_iteration": 3.5486855506896973
    },
    {
      "auxiliary_loss_clip": 0.0118715,
      "auxiliary_loss_mlp": 0.01033275,
      "balance_loss_clip": 1.0547471,
      "balance_loss_mlp": 1.0241909,
      "epoch": 0.4297480911441111,
      "flos": 13881270470400.0,
      "grad_norm": 2.7677268407371534,
      "language_loss": 0.93260169,
      "learning_rate": 2.5452983350494595e-06,
      "loss": 0.95480591,
      "num_input_tokens_seen": 77049925,
      "step": 3574,
      "time_per_iteration": 2.6645748615264893
    },
    {
      "auxiliary_loss_clip": 0.01189035,
      "auxiliary_loss_mlp": 0.0103045,
      "balance_loss_clip": 1.05753469,
      "balance_loss_mlp": 1.02133656,
      "epoch": 0.4298683340347502,
      "flos": 20741141975040.0,
      "grad_norm": 2.389278395808295,
      "language_loss": 0.65144253,
      "learning_rate": 2.544548834551713e-06,
      "loss": 0.67363733,
      "num_input_tokens_seen": 77068930,
      "step": 3575,
      "time_per_iteration": 2.5828847885131836
    },
    {
      "auxiliary_loss_clip": 0.01189658,
      "auxiliary_loss_mlp": 0.01028259,
      "balance_loss_clip": 1.05786979,
      "balance_loss_mlp": 1.01975393,
      "epoch": 0.4299885769253893,
      "flos": 20882126856960.0,
      "grad_norm": 2.534935563401723,
      "language_loss": 0.94164789,
      "learning_rate": 2.5437992514463424e-06,
      "loss": 0.96382713,
      "num_input_tokens_seen": 77082255,
      "step": 3576,
      "time_per_iteration": 2.6378562450408936
    },
    {
      "auxiliary_loss_clip": 0.01192169,
      "auxiliary_loss_mlp": 0.01027391,
      "balance_loss_clip": 1.05824184,
      "balance_loss_mlp": 1.01798546,
      "epoch": 0.4301088198160284,
      "flos": 25484618183040.0,
      "grad_norm": 1.7472386578102441,
      "language_loss": 0.88139707,
      "learning_rate": 2.5430495858470565e-06,
      "loss": 0.90359271,
      "num_input_tokens_seen": 77101725,
      "step": 3577,
      "time_per_iteration": 2.6737232208251953
    },
    {
      "auxiliary_loss_clip": 0.01188749,
      "auxiliary_loss_mlp": 0.01032322,
      "balance_loss_clip": 1.05683768,
      "balance_loss_mlp": 1.02333331,
      "epoch": 0.43022906270666744,
      "flos": 18259427404800.0,
      "grad_norm": 3.030633405577258,
      "language_loss": 0.77619326,
      "learning_rate": 2.54229983786758e-06,
      "loss": 0.79840404,
      "num_input_tokens_seen": 77119670,
      "step": 3578,
      "time_per_iteration": 2.6567609310150146
    },
    {
      "auxiliary_loss_clip": 0.01193898,
      "auxiliary_loss_mlp": 0.0102971,
      "balance_loss_clip": 1.06071579,
      "balance_loss_mlp": 1.02113867,
      "epoch": 0.43034930559730655,
      "flos": 23399536567680.0,
      "grad_norm": 2.0775126282002465,
      "language_loss": 0.84927672,
      "learning_rate": 2.541550007621651e-06,
      "loss": 0.87151277,
      "num_input_tokens_seen": 77138160,
      "step": 3579,
      "time_per_iteration": 2.605764389038086
    },
    {
      "auxiliary_loss_clip": 0.01189597,
      "auxiliary_loss_mlp": 0.01031606,
      "balance_loss_clip": 1.05994678,
      "balance_loss_mlp": 1.02302337,
      "epoch": 0.43046954848794566,
      "flos": 28184382264960.0,
      "grad_norm": 2.846754665617626,
      "language_loss": 0.8038559,
      "learning_rate": 2.5408000952230156e-06,
      "loss": 0.82606792,
      "num_input_tokens_seen": 77156950,
      "step": 3580,
      "time_per_iteration": 2.6751582622528076
    },
    {
      "auxiliary_loss_clip": 0.01190796,
      "auxiliary_loss_mlp": 0.01032559,
      "balance_loss_clip": 1.05771852,
      "balance_loss_mlp": 1.02323103,
      "epoch": 0.4305897913785847,
      "flos": 28580476515840.0,
      "grad_norm": 1.9967968498692328,
      "language_loss": 0.90552497,
      "learning_rate": 2.5400501007854357e-06,
      "loss": 0.92775851,
      "num_input_tokens_seen": 77176395,
      "step": 3581,
      "time_per_iteration": 2.682058572769165
    },
    {
      "auxiliary_loss_clip": 0.01184877,
      "auxiliary_loss_mlp": 0.01027319,
      "balance_loss_clip": 1.05534816,
      "balance_loss_mlp": 1.01882505,
      "epoch": 0.43071003426922383,
      "flos": 20448721353600.0,
      "grad_norm": 1.9317624351965517,
      "language_loss": 0.75434446,
      "learning_rate": 2.539300024422685e-06,
      "loss": 0.77646637,
      "num_input_tokens_seen": 77194340,
      "step": 3582,
      "time_per_iteration": 2.6199378967285156
    },
    {
      "auxiliary_loss_clip": 0.01081749,
      "auxiliary_loss_mlp": 0.01003148,
      "balance_loss_clip": 1.01760125,
      "balance_loss_mlp": 1.0018605,
      "epoch": 0.43083027715986294,
      "flos": 51997969883520.0,
      "grad_norm": 0.790557795530441,
      "language_loss": 0.60902822,
      "learning_rate": 2.538549866248549e-06,
      "loss": 0.62987715,
      "num_input_tokens_seen": 77249320,
      "step": 3583,
      "time_per_iteration": 3.066729784011841
    },
    {
      "auxiliary_loss_clip": 0.01190464,
      "auxiliary_loss_mlp": 0.01026852,
      "balance_loss_clip": 1.05754781,
      "balance_loss_mlp": 1.01735127,
      "epoch": 0.430950520050502,
      "flos": 16690885320960.0,
      "grad_norm": 2.369299495416214,
      "language_loss": 0.81099689,
      "learning_rate": 2.5377996263768274e-06,
      "loss": 0.83317006,
      "num_input_tokens_seen": 77267400,
      "step": 3584,
      "time_per_iteration": 2.6041958332061768
    },
    {
      "auxiliary_loss_clip": 0.01192096,
      "auxiliary_loss_mlp": 0.01032602,
      "balance_loss_clip": 1.06114411,
      "balance_loss_mlp": 1.02360773,
      "epoch": 0.4310707629411411,
      "flos": 24608433726720.0,
      "grad_norm": 1.7169695104095903,
      "language_loss": 0.68611407,
      "learning_rate": 2.5370493049213293e-06,
      "loss": 0.70836109,
      "num_input_tokens_seen": 77287045,
      "step": 3585,
      "time_per_iteration": 2.688886880874634
    },
    {
      "auxiliary_loss_clip": 0.01190567,
      "auxiliary_loss_mlp": 0.01030471,
      "balance_loss_clip": 1.05659461,
      "balance_loss_mlp": 1.02110088,
      "epoch": 0.4311910058317802,
      "flos": 26432983019520.0,
      "grad_norm": 4.118739783505985,
      "language_loss": 0.79938871,
      "learning_rate": 2.536298901995878e-06,
      "loss": 0.82159913,
      "num_input_tokens_seen": 77306255,
      "step": 3586,
      "time_per_iteration": 2.6521482467651367
    },
    {
      "auxiliary_loss_clip": 0.0119109,
      "auxiliary_loss_mlp": 0.01027802,
      "balance_loss_clip": 1.05880654,
      "balance_loss_mlp": 1.01861143,
      "epoch": 0.43131124872241927,
      "flos": 25155891889920.0,
      "grad_norm": 1.72421165307327,
      "language_loss": 0.80360186,
      "learning_rate": 2.535548417714311e-06,
      "loss": 0.82579076,
      "num_input_tokens_seen": 77325555,
      "step": 3587,
      "time_per_iteration": 2.690544366836548
    },
    {
      "auxiliary_loss_clip": 0.01189297,
      "auxiliary_loss_mlp": 0.01028746,
      "balance_loss_clip": 1.05580509,
      "balance_loss_mlp": 1.02031231,
      "epoch": 0.4314314916130584,
      "flos": 21614812479360.0,
      "grad_norm": 1.9694510035839707,
      "language_loss": 0.87371147,
      "learning_rate": 2.534797852190474e-06,
      "loss": 0.8958919,
      "num_input_tokens_seen": 77345735,
      "step": 3588,
      "time_per_iteration": 2.649150848388672
    },
    {
      "auxiliary_loss_clip": 0.01189633,
      "auxiliary_loss_mlp": 0.01034559,
      "balance_loss_clip": 1.05907953,
      "balance_loss_mlp": 1.02581549,
      "epoch": 0.4315517345036975,
      "flos": 19275016544640.0,
      "grad_norm": 1.8985407843290552,
      "language_loss": 0.81624752,
      "learning_rate": 2.5340472055382283e-06,
      "loss": 0.83848941,
      "num_input_tokens_seen": 77361765,
      "step": 3589,
      "time_per_iteration": 2.7195498943328857
    },
    {
      "auxiliary_loss_clip": 0.01189084,
      "auxiliary_loss_mlp": 0.01025954,
      "balance_loss_clip": 1.05910015,
      "balance_loss_mlp": 1.01771629,
      "epoch": 0.43167197739433655,
      "flos": 24273853516800.0,
      "grad_norm": 3.244614124574233,
      "language_loss": 0.80981135,
      "learning_rate": 2.5332964778714468e-06,
      "loss": 0.83196169,
      "num_input_tokens_seen": 77378950,
      "step": 3590,
      "time_per_iteration": 2.656728744506836
    },
    {
      "auxiliary_loss_clip": 0.01188466,
      "auxiliary_loss_mlp": 0.01037002,
      "balance_loss_clip": 1.05790925,
      "balance_loss_mlp": 1.0279839,
      "epoch": 0.43179222028497566,
      "flos": 16867816738560.0,
      "grad_norm": 1.4968638153236122,
      "language_loss": 0.65984368,
      "learning_rate": 2.5325456693040123e-06,
      "loss": 0.68209833,
      "num_input_tokens_seen": 77396145,
      "step": 3591,
      "time_per_iteration": 2.689135789871216
    },
    {
      "auxiliary_loss_clip": 0.01190764,
      "auxiliary_loss_mlp": 0.0102824,
      "balance_loss_clip": 1.05725837,
      "balance_loss_mlp": 1.01862001,
      "epoch": 0.43191246317561477,
      "flos": 17639214243840.0,
      "grad_norm": 2.1534300194138494,
      "language_loss": 0.75121123,
      "learning_rate": 2.531794779949824e-06,
      "loss": 0.77340126,
      "num_input_tokens_seen": 77414045,
      "step": 3592,
      "time_per_iteration": 2.631518840789795
    },
    {
      "auxiliary_loss_clip": 0.01186825,
      "auxiliary_loss_mlp": 0.01028891,
      "balance_loss_clip": 1.05641091,
      "balance_loss_mlp": 1.02051103,
      "epoch": 0.4320327060662538,
      "flos": 23878800760320.0,
      "grad_norm": 1.681318496444379,
      "language_loss": 0.87903517,
      "learning_rate": 2.5310438099227903e-06,
      "loss": 0.90119231,
      "num_input_tokens_seen": 77431310,
      "step": 3593,
      "time_per_iteration": 2.6817078590393066
    },
    {
      "auxiliary_loss_clip": 0.01081678,
      "auxiliary_loss_mlp": 0.01001226,
      "balance_loss_clip": 1.01748419,
      "balance_loss_mlp": 0.99992096,
      "epoch": 0.43215294895689293,
      "flos": 66394917959040.0,
      "grad_norm": 0.8412432541091005,
      "language_loss": 0.5336206,
      "learning_rate": 2.530292759336833e-06,
      "loss": 0.55444962,
      "num_input_tokens_seen": 77492045,
      "step": 3594,
      "time_per_iteration": 4.252472639083862
    },
    {
      "auxiliary_loss_clip": 0.01188871,
      "auxiliary_loss_mlp": 0.01029387,
      "balance_loss_clip": 1.05560434,
      "balance_loss_mlp": 1.01956463,
      "epoch": 0.432273191847532,
      "flos": 20594267262720.0,
      "grad_norm": 2.8859138272282396,
      "language_loss": 0.69824034,
      "learning_rate": 2.5295416283058855e-06,
      "loss": 0.72042292,
      "num_input_tokens_seen": 77510910,
      "step": 3595,
      "time_per_iteration": 2.6084110736846924
    },
    {
      "auxiliary_loss_clip": 0.01190297,
      "auxiliary_loss_mlp": 0.01035935,
      "balance_loss_clip": 1.06038451,
      "balance_loss_mlp": 1.02777481,
      "epoch": 0.4323934347381711,
      "flos": 19282127437440.0,
      "grad_norm": 1.5644753943130265,
      "language_loss": 0.65702462,
      "learning_rate": 2.5287904169438943e-06,
      "loss": 0.6792869,
      "num_input_tokens_seen": 77530115,
      "step": 3596,
      "time_per_iteration": 2.6829590797424316
    },
    {
      "auxiliary_loss_clip": 0.01198078,
      "auxiliary_loss_mlp": 0.01031184,
      "balance_loss_clip": 1.06242836,
      "balance_loss_mlp": 1.02178478,
      "epoch": 0.4325136776288102,
      "flos": 21726315273600.0,
      "grad_norm": 3.2253692830966476,
      "language_loss": 0.63437152,
      "learning_rate": 2.528039125364817e-06,
      "loss": 0.65666413,
      "num_input_tokens_seen": 77548920,
      "step": 3597,
      "time_per_iteration": 3.6744091510772705
    },
    {
      "auxiliary_loss_clip": 0.01190997,
      "auxiliary_loss_mlp": 0.01032087,
      "balance_loss_clip": 1.05774188,
      "balance_loss_mlp": 1.02241361,
      "epoch": 0.43263392051944927,
      "flos": 22340746344960.0,
      "grad_norm": 3.899337335862752,
      "language_loss": 0.7580204,
      "learning_rate": 2.5272877536826246e-06,
      "loss": 0.78025126,
      "num_input_tokens_seen": 77567715,
      "step": 3598,
      "time_per_iteration": 3.5006728172302246
    },
    {
      "auxiliary_loss_clip": 0.01186527,
      "auxiliary_loss_mlp": 0.01031094,
      "balance_loss_clip": 1.05334973,
      "balance_loss_mlp": 1.02240944,
      "epoch": 0.4327541634100884,
      "flos": 29168406328320.0,
      "grad_norm": 2.887356227645135,
      "language_loss": 0.70516384,
      "learning_rate": 2.5265363020112986e-06,
      "loss": 0.72733998,
      "num_input_tokens_seen": 77588035,
      "step": 3599,
      "time_per_iteration": 3.704576015472412
    },
    {
      "auxiliary_loss_clip": 0.01191342,
      "auxiliary_loss_mlp": 0.0102753,
      "balance_loss_clip": 1.05886877,
      "balance_loss_mlp": 1.01780844,
      "epoch": 0.4328744063007275,
      "flos": 26067448264320.0,
      "grad_norm": 1.8421041525535575,
      "language_loss": 0.83480847,
      "learning_rate": 2.5257847704648344e-06,
      "loss": 0.85699719,
      "num_input_tokens_seen": 77609265,
      "step": 3600,
      "time_per_iteration": 2.651115894317627
    },
    {
      "auxiliary_loss_clip": 0.0118962,
      "auxiliary_loss_mlp": 0.01023552,
      "balance_loss_clip": 1.0587827,
      "balance_loss_mlp": 1.01495099,
      "epoch": 0.43299464919136654,
      "flos": 16581357774720.0,
      "grad_norm": 1.8600362183166408,
      "language_loss": 0.75492573,
      "learning_rate": 2.525033159157239e-06,
      "loss": 0.77705741,
      "num_input_tokens_seen": 77625580,
      "step": 3601,
      "time_per_iteration": 2.567551851272583
    },
    {
      "auxiliary_loss_clip": 0.01192193,
      "auxiliary_loss_mlp": 0.01032813,
      "balance_loss_clip": 1.05891705,
      "balance_loss_mlp": 1.02265048,
      "epoch": 0.43311489208200565,
      "flos": 16107265140480.0,
      "grad_norm": 2.213054246053201,
      "language_loss": 0.77384305,
      "learning_rate": 2.52428146820253e-06,
      "loss": 0.79609311,
      "num_input_tokens_seen": 77643835,
      "step": 3602,
      "time_per_iteration": 2.6175966262817383
    },
    {
      "auxiliary_loss_clip": 0.01188877,
      "auxiliary_loss_mlp": 0.01034378,
      "balance_loss_clip": 1.0572629,
      "balance_loss_mlp": 1.02500784,
      "epoch": 0.43323513497264476,
      "flos": 22930220442240.0,
      "grad_norm": 2.4698832034524183,
      "language_loss": 0.81861955,
      "learning_rate": 2.52352969771474e-06,
      "loss": 0.84085214,
      "num_input_tokens_seen": 77663060,
      "step": 3603,
      "time_per_iteration": 2.6706154346466064
    },
    {
      "auxiliary_loss_clip": 0.01187794,
      "auxiliary_loss_mlp": 0.01030273,
      "balance_loss_clip": 1.05855727,
      "balance_loss_mlp": 1.02190471,
      "epoch": 0.4333553778632838,
      "flos": 25299031587840.0,
      "grad_norm": 1.968859151866394,
      "language_loss": 0.88826984,
      "learning_rate": 2.5227778478079106e-06,
      "loss": 0.91045058,
      "num_input_tokens_seen": 77682470,
      "step": 3604,
      "time_per_iteration": 2.6685659885406494
    },
    {
      "auxiliary_loss_clip": 0.01188408,
      "auxiliary_loss_mlp": 0.01029605,
      "balance_loss_clip": 1.05876076,
      "balance_loss_mlp": 1.02092075,
      "epoch": 0.43347562075392293,
      "flos": 19387165783680.0,
      "grad_norm": 2.023271424877265,
      "language_loss": 0.76796114,
      "learning_rate": 2.522025918596098e-06,
      "loss": 0.79014128,
      "num_input_tokens_seen": 77700770,
      "step": 3605,
      "time_per_iteration": 2.645402431488037
    },
    {
      "auxiliary_loss_clip": 0.01186396,
      "auxiliary_loss_mlp": 0.01024868,
      "balance_loss_clip": 1.05859995,
      "balance_loss_mlp": 1.01680994,
      "epoch": 0.43359586364456204,
      "flos": 26325969425280.0,
      "grad_norm": 1.4976733236516742,
      "language_loss": 0.65234113,
      "learning_rate": 2.521273910193368e-06,
      "loss": 0.67445374,
      "num_input_tokens_seen": 77723950,
      "step": 3606,
      "time_per_iteration": 2.741093873977661
    },
    {
      "auxiliary_loss_clip": 0.01191568,
      "auxiliary_loss_mlp": 0.01029321,
      "balance_loss_clip": 1.05904078,
      "balance_loss_mlp": 1.02036881,
      "epoch": 0.4337161065352011,
      "flos": 15989261984640.0,
      "grad_norm": 2.0048691506464933,
      "language_loss": 0.87034887,
      "learning_rate": 2.5205218227138006e-06,
      "loss": 0.89255774,
      "num_input_tokens_seen": 77736905,
      "step": 3607,
      "time_per_iteration": 2.580034017562866
    },
    {
      "auxiliary_loss_clip": 0.01188764,
      "auxiliary_loss_mlp": 0.01030071,
      "balance_loss_clip": 1.0576365,
      "balance_loss_mlp": 1.02213168,
      "epoch": 0.4338363494258402,
      "flos": 20224710184320.0,
      "grad_norm": 1.8816334560348746,
      "language_loss": 0.79000139,
      "learning_rate": 2.519769656271486e-06,
      "loss": 0.81218982,
      "num_input_tokens_seen": 77754325,
      "step": 3608,
      "time_per_iteration": 2.6499826908111572
    },
    {
      "auxiliary_loss_clip": 0.01189997,
      "auxiliary_loss_mlp": 0.0103172,
      "balance_loss_clip": 1.05824757,
      "balance_loss_mlp": 1.02280903,
      "epoch": 0.43395659231647926,
      "flos": 20083904870400.0,
      "grad_norm": 1.9989201537336019,
      "language_loss": 0.67684138,
      "learning_rate": 2.5190174109805285e-06,
      "loss": 0.69905853,
      "num_input_tokens_seen": 77774150,
      "step": 3609,
      "time_per_iteration": 2.5732734203338623
    },
    {
      "auxiliary_loss_clip": 0.01187347,
      "auxiliary_loss_mlp": 0.01031006,
      "balance_loss_clip": 1.05587566,
      "balance_loss_mlp": 1.02245235,
      "epoch": 0.43407683520711837,
      "flos": 19901801894400.0,
      "grad_norm": 1.9703487258556371,
      "language_loss": 0.64211929,
      "learning_rate": 2.518265086955042e-06,
      "loss": 0.66430277,
      "num_input_tokens_seen": 77791870,
      "step": 3610,
      "time_per_iteration": 2.6276438236236572
    },
    {
      "auxiliary_loss_clip": 0.01187788,
      "auxiliary_loss_mlp": 0.01027796,
      "balance_loss_clip": 1.05717909,
      "balance_loss_mlp": 1.01848555,
      "epoch": 0.4341970780977575,
      "flos": 23108732058240.0,
      "grad_norm": 1.9036291185662368,
      "language_loss": 0.83245653,
      "learning_rate": 2.5175126843091534e-06,
      "loss": 0.85461235,
      "num_input_tokens_seen": 77811240,
      "step": 3611,
      "time_per_iteration": 2.632328510284424
    },
    {
      "auxiliary_loss_clip": 0.01185891,
      "auxiliary_loss_mlp": 0.01032036,
      "balance_loss_clip": 1.05608404,
      "balance_loss_mlp": 1.02335739,
      "epoch": 0.43431732098839654,
      "flos": 37408288406400.0,
      "grad_norm": 2.327219263366736,
      "language_loss": 0.75603104,
      "learning_rate": 2.5167602031570034e-06,
      "loss": 0.77821028,
      "num_input_tokens_seen": 77831425,
      "step": 3612,
      "time_per_iteration": 2.6857481002807617
    },
    {
      "auxiliary_loss_clip": 0.01190347,
      "auxiliary_loss_mlp": 0.01031812,
      "balance_loss_clip": 1.05867887,
      "balance_loss_mlp": 1.02263856,
      "epoch": 0.43443756387903565,
      "flos": 31868206323840.0,
      "grad_norm": 1.753491407996452,
      "language_loss": 0.73660147,
      "learning_rate": 2.51600764361274e-06,
      "loss": 0.75882304,
      "num_input_tokens_seen": 77852950,
      "step": 3613,
      "time_per_iteration": 2.663727283477783
    },
    {
      "auxiliary_loss_clip": 0.01189194,
      "auxiliary_loss_mlp": 0.01027383,
      "balance_loss_clip": 1.05808783,
      "balance_loss_mlp": 1.01851392,
      "epoch": 0.43455780676967476,
      "flos": 23477139901440.0,
      "grad_norm": 3.191084749501579,
      "language_loss": 0.78919274,
      "learning_rate": 2.5152550057905283e-06,
      "loss": 0.81135845,
      "num_input_tokens_seen": 77872840,
      "step": 3614,
      "time_per_iteration": 2.6893835067749023
    },
    {
      "auxiliary_loss_clip": 0.01191649,
      "auxiliary_loss_mlp": 0.01033746,
      "balance_loss_clip": 1.05823553,
      "balance_loss_mlp": 1.02447164,
      "epoch": 0.4346780496603138,
      "flos": 24207060176640.0,
      "grad_norm": 2.401163646567461,
      "language_loss": 0.76958823,
      "learning_rate": 2.5145022898045415e-06,
      "loss": 0.79184216,
      "num_input_tokens_seen": 77892025,
      "step": 3615,
      "time_per_iteration": 2.628498077392578
    },
    {
      "auxiliary_loss_clip": 0.01187785,
      "auxiliary_loss_mlp": 0.01034418,
      "balance_loss_clip": 1.05536699,
      "balance_loss_mlp": 1.02463078,
      "epoch": 0.4347982925509529,
      "flos": 17092366611840.0,
      "grad_norm": 2.396185283319455,
      "language_loss": 0.8973707,
      "learning_rate": 2.5137494957689664e-06,
      "loss": 0.91959268,
      "num_input_tokens_seen": 77907635,
      "step": 3616,
      "time_per_iteration": 2.5979363918304443
    },
    {
      "auxiliary_loss_clip": 0.01082977,
      "auxiliary_loss_mlp": 0.0100187,
      "balance_loss_clip": 1.01900506,
      "balance_loss_mlp": 1.00048149,
      "epoch": 0.43491853544159204,
      "flos": 60945544696320.0,
      "grad_norm": 0.7653905671883223,
      "language_loss": 0.57389367,
      "learning_rate": 2.5129966237980016e-06,
      "loss": 0.59474218,
      "num_input_tokens_seen": 77970630,
      "step": 3617,
      "time_per_iteration": 3.161264181137085
    },
    {
      "auxiliary_loss_clip": 0.01187242,
      "auxiliary_loss_mlp": 0.01026693,
      "balance_loss_clip": 1.05757618,
      "balance_loss_mlp": 1.01806796,
      "epoch": 0.4350387783322311,
      "flos": 21944652094080.0,
      "grad_norm": 1.7097512840587263,
      "language_loss": 0.78334248,
      "learning_rate": 2.512243674005857e-06,
      "loss": 0.80548191,
      "num_input_tokens_seen": 77989995,
      "step": 3618,
      "time_per_iteration": 2.62204909324646
    },
    {
      "auxiliary_loss_clip": 0.01191047,
      "auxiliary_loss_mlp": 0.01034506,
      "balance_loss_clip": 1.06167483,
      "balance_loss_mlp": 1.02570879,
      "epoch": 0.4351590212228702,
      "flos": 25082705928960.0,
      "grad_norm": 2.062821001698054,
      "language_loss": 0.86430335,
      "learning_rate": 2.5114906465067537e-06,
      "loss": 0.88655889,
      "num_input_tokens_seen": 78010980,
      "step": 3619,
      "time_per_iteration": 2.6739089488983154
    },
    {
      "auxiliary_loss_clip": 0.01189895,
      "auxiliary_loss_mlp": 0.01024981,
      "balance_loss_clip": 1.05873787,
      "balance_loss_mlp": 1.01637983,
      "epoch": 0.4352792641135093,
      "flos": 21506541909120.0,
      "grad_norm": 2.642245344875597,
      "language_loss": 0.74966419,
      "learning_rate": 2.5107375414149264e-06,
      "loss": 0.77181292,
      "num_input_tokens_seen": 78030225,
      "step": 3620,
      "time_per_iteration": 3.5652523040771484
    },
    {
      "auxiliary_loss_clip": 0.01187689,
      "auxiliary_loss_mlp": 0.01028639,
      "balance_loss_clip": 1.05519962,
      "balance_loss_mlp": 1.01951981,
      "epoch": 0.43539950700414837,
      "flos": 16253457494400.0,
      "grad_norm": 2.3518908592376646,
      "language_loss": 0.71794617,
      "learning_rate": 2.5099843588446197e-06,
      "loss": 0.74010944,
      "num_input_tokens_seen": 78048545,
      "step": 3621,
      "time_per_iteration": 2.6740198135375977
    },
    {
      "auxiliary_loss_clip": 0.01190689,
      "auxiliary_loss_mlp": 0.0103277,
      "balance_loss_clip": 1.05922556,
      "balance_loss_mlp": 1.0238409,
      "epoch": 0.4355197498947875,
      "flos": 16691819074560.0,
      "grad_norm": 1.5892000685374459,
      "language_loss": 0.61200607,
      "learning_rate": 2.509231098910091e-06,
      "loss": 0.63424063,
      "num_input_tokens_seen": 78068415,
      "step": 3622,
      "time_per_iteration": 2.6874165534973145
    },
    {
      "auxiliary_loss_clip": 0.01186429,
      "auxiliary_loss_mlp": 0.01026289,
      "balance_loss_clip": 1.05616403,
      "balance_loss_mlp": 1.01795042,
      "epoch": 0.4356399927854266,
      "flos": 16362733645440.0,
      "grad_norm": 8.692042558874896,
      "language_loss": 0.74727935,
      "learning_rate": 2.508477761725611e-06,
      "loss": 0.7694065,
      "num_input_tokens_seen": 78086690,
      "step": 3623,
      "time_per_iteration": 2.637345314025879
    },
    {
      "auxiliary_loss_clip": 0.01185393,
      "auxiliary_loss_mlp": 0.01026609,
      "balance_loss_clip": 1.05574667,
      "balance_loss_mlp": 1.0173521,
      "epoch": 0.43576023567606564,
      "flos": 17202037812480.0,
      "grad_norm": 1.944299192421025,
      "language_loss": 0.80887282,
      "learning_rate": 2.507724347405458e-06,
      "loss": 0.83099294,
      "num_input_tokens_seen": 78104640,
      "step": 3624,
      "time_per_iteration": 3.523383617401123
    },
    {
      "auxiliary_loss_clip": 0.01184771,
      "auxiliary_loss_mlp": 0.01030367,
      "balance_loss_clip": 1.05554616,
      "balance_loss_mlp": 1.02209401,
      "epoch": 0.43588047856670475,
      "flos": 15917656222080.0,
      "grad_norm": 1.9613156756443746,
      "language_loss": 0.81648195,
      "learning_rate": 2.5069708560639243e-06,
      "loss": 0.83863336,
      "num_input_tokens_seen": 78122550,
      "step": 3625,
      "time_per_iteration": 3.5114145278930664
    },
    {
      "auxiliary_loss_clip": 0.01188952,
      "auxiliary_loss_mlp": 0.01024533,
      "balance_loss_clip": 1.05792344,
      "balance_loss_mlp": 1.01579452,
      "epoch": 0.4360007214573438,
      "flos": 23659566099840.0,
      "grad_norm": 2.263918557909108,
      "language_loss": 0.61054134,
      "learning_rate": 2.5062172878153158e-06,
      "loss": 0.63267624,
      "num_input_tokens_seen": 78141825,
      "step": 3626,
      "time_per_iteration": 3.579195737838745
    },
    {
      "auxiliary_loss_clip": 0.01186879,
      "auxiliary_loss_mlp": 0.01030402,
      "balance_loss_clip": 1.05552959,
      "balance_loss_mlp": 1.02147317,
      "epoch": 0.4361209643479829,
      "flos": 21978767036160.0,
      "grad_norm": 1.9173410135987952,
      "language_loss": 0.87501585,
      "learning_rate": 2.505463642773947e-06,
      "loss": 0.89718866,
      "num_input_tokens_seen": 78161790,
      "step": 3627,
      "time_per_iteration": 2.6219351291656494
    },
    {
      "auxiliary_loss_clip": 0.01186709,
      "auxiliary_loss_mlp": 0.01028341,
      "balance_loss_clip": 1.05535197,
      "balance_loss_mlp": 1.01951981,
      "epoch": 0.43624120723862203,
      "flos": 17420159151360.0,
      "grad_norm": 2.7391203677781366,
      "language_loss": 0.75029594,
      "learning_rate": 2.504709921054146e-06,
      "loss": 0.77244639,
      "num_input_tokens_seen": 78178605,
      "step": 3628,
      "time_per_iteration": 2.6041274070739746
    },
    {
      "auxiliary_loss_clip": 0.0119032,
      "auxiliary_loss_mlp": 0.01029931,
      "balance_loss_clip": 1.05895567,
      "balance_loss_mlp": 1.02069199,
      "epoch": 0.4363614501292611,
      "flos": 17895293280000.0,
      "grad_norm": 2.1989932914957184,
      "language_loss": 0.83781981,
      "learning_rate": 2.50395612277025e-06,
      "loss": 0.86002237,
      "num_input_tokens_seen": 78194460,
      "step": 3629,
      "time_per_iteration": 2.631589889526367
    },
    {
      "auxiliary_loss_clip": 0.01184996,
      "auxiliary_loss_mlp": 0.01025392,
      "balance_loss_clip": 1.05444169,
      "balance_loss_mlp": 1.01729155,
      "epoch": 0.4364816930199002,
      "flos": 20302888135680.0,
      "grad_norm": 2.8431491967480937,
      "language_loss": 0.72892523,
      "learning_rate": 2.503202248036612e-06,
      "loss": 0.75102907,
      "num_input_tokens_seen": 78213315,
      "step": 3630,
      "time_per_iteration": 2.5862691402435303
    },
    {
      "auxiliary_loss_clip": 0.01187018,
      "auxiliary_loss_mlp": 0.01027542,
      "balance_loss_clip": 1.05736268,
      "balance_loss_mlp": 1.01884556,
      "epoch": 0.4366019359105393,
      "flos": 24061334699520.0,
      "grad_norm": 1.7599665224870067,
      "language_loss": 0.73478138,
      "learning_rate": 2.5024482969675927e-06,
      "loss": 0.75692695,
      "num_input_tokens_seen": 78233270,
      "step": 3631,
      "time_per_iteration": 2.6865994930267334
    },
    {
      "auxiliary_loss_clip": 0.01181912,
      "auxiliary_loss_mlp": 0.01024053,
      "balance_loss_clip": 1.05468535,
      "balance_loss_mlp": 1.01601839,
      "epoch": 0.43672217880117836,
      "flos": 21754109422080.0,
      "grad_norm": 2.8133241825197697,
      "language_loss": 0.84695905,
      "learning_rate": 2.501694269677566e-06,
      "loss": 0.86901867,
      "num_input_tokens_seen": 78251040,
      "step": 3632,
      "time_per_iteration": 2.628760814666748
    },
    {
      "auxiliary_loss_clip": 0.01184014,
      "auxiliary_loss_mlp": 0.01026328,
      "balance_loss_clip": 1.0541563,
      "balance_loss_mlp": 1.01804304,
      "epoch": 0.4368424216918175,
      "flos": 18035200753920.0,
      "grad_norm": 2.750490777157947,
      "language_loss": 0.80261678,
      "learning_rate": 2.500940166280918e-06,
      "loss": 0.8247202,
      "num_input_tokens_seen": 78269470,
      "step": 3633,
      "time_per_iteration": 2.6736929416656494
    },
    {
      "auxiliary_loss_clip": 0.01184097,
      "auxiliary_loss_mlp": 0.01030334,
      "balance_loss_clip": 1.05618572,
      "balance_loss_mlp": 1.02197134,
      "epoch": 0.4369626645824566,
      "flos": 25447127362560.0,
      "grad_norm": 1.8614448610534422,
      "language_loss": 0.79645532,
      "learning_rate": 2.500185986892045e-06,
      "loss": 0.81859964,
      "num_input_tokens_seen": 78288955,
      "step": 3634,
      "time_per_iteration": 2.72904372215271
    },
    {
      "auxiliary_loss_clip": 0.0118913,
      "auxiliary_loss_mlp": 0.01026979,
      "balance_loss_clip": 1.0573281,
      "balance_loss_mlp": 1.0175612,
      "epoch": 0.43708290747309564,
      "flos": 25302694775040.0,
      "grad_norm": 2.009312240978406,
      "language_loss": 0.77552116,
      "learning_rate": 2.499431731625355e-06,
      "loss": 0.79768229,
      "num_input_tokens_seen": 78307980,
      "step": 3635,
      "time_per_iteration": 2.626199245452881
    },
    {
      "auxiliary_loss_clip": 0.0118616,
      "auxiliary_loss_mlp": 0.01035324,
      "balance_loss_clip": 1.05526459,
      "balance_loss_mlp": 1.02678895,
      "epoch": 0.43720315036373475,
      "flos": 31575103344000.0,
      "grad_norm": 1.940804283734395,
      "language_loss": 0.79720676,
      "learning_rate": 2.4986774005952686e-06,
      "loss": 0.81942159,
      "num_input_tokens_seen": 78330355,
      "step": 3636,
      "time_per_iteration": 2.7361843585968018
    },
    {
      "auxiliary_loss_clip": 0.01184243,
      "auxiliary_loss_mlp": 0.01028706,
      "balance_loss_clip": 1.05530882,
      "balance_loss_mlp": 1.01999187,
      "epoch": 0.43732339325437386,
      "flos": 23112000195840.0,
      "grad_norm": 2.0620673300909678,
      "language_loss": 0.84869218,
      "learning_rate": 2.4979229939162166e-06,
      "loss": 0.87082171,
      "num_input_tokens_seen": 78349135,
      "step": 3637,
      "time_per_iteration": 2.657927989959717
    },
    {
      "auxiliary_loss_clip": 0.01182064,
      "auxiliary_loss_mlp": 0.01024369,
      "balance_loss_clip": 1.054721,
      "balance_loss_mlp": 1.01626241,
      "epoch": 0.4374436361450129,
      "flos": 27746272080000.0,
      "grad_norm": 1.567559255316952,
      "language_loss": 0.80485207,
      "learning_rate": 2.4971685117026433e-06,
      "loss": 0.8269164,
      "num_input_tokens_seen": 78368900,
      "step": 3638,
      "time_per_iteration": 2.699002504348755
    },
    {
      "auxiliary_loss_clip": 0.01185687,
      "auxiliary_loss_mlp": 0.01022824,
      "balance_loss_clip": 1.05694151,
      "balance_loss_mlp": 1.01413953,
      "epoch": 0.437563879035652,
      "flos": 24172370616960.0,
      "grad_norm": 1.6307561300528282,
      "language_loss": 0.7665987,
      "learning_rate": 2.4964139540690018e-06,
      "loss": 0.78868377,
      "num_input_tokens_seen": 78392235,
      "step": 3639,
      "time_per_iteration": 2.6439883708953857
    },
    {
      "auxiliary_loss_clip": 0.01184344,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.05497205,
      "balance_loss_mlp": 1.0159204,
      "epoch": 0.4376841219262911,
      "flos": 23477211728640.0,
      "grad_norm": 4.777321652032218,
      "language_loss": 0.72698963,
      "learning_rate": 2.495659321129758e-06,
      "loss": 0.74908715,
      "num_input_tokens_seen": 78409980,
      "step": 3640,
      "time_per_iteration": 2.628192663192749
    },
    {
      "auxiliary_loss_clip": 0.011833,
      "auxiliary_loss_mlp": 0.01031154,
      "balance_loss_clip": 1.05509329,
      "balance_loss_mlp": 1.02344704,
      "epoch": 0.4378043648169302,
      "flos": 25447809720960.0,
      "grad_norm": 7.079670224172516,
      "language_loss": 0.75673687,
      "learning_rate": 2.494904612999389e-06,
      "loss": 0.77888137,
      "num_input_tokens_seen": 78428690,
      "step": 3641,
      "time_per_iteration": 2.634920597076416
    },
    {
      "auxiliary_loss_clip": 0.01083056,
      "auxiliary_loss_mlp": 0.01004115,
      "balance_loss_clip": 1.0192821,
      "balance_loss_mlp": 1.00294673,
      "epoch": 0.4379246077075693,
      "flos": 53914056986880.0,
      "grad_norm": 0.7543024252391651,
      "language_loss": 0.56499314,
      "learning_rate": 2.4941498297923843e-06,
      "loss": 0.5858649,
      "num_input_tokens_seen": 78489260,
      "step": 3642,
      "time_per_iteration": 3.1626901626586914
    },
    {
      "auxiliary_loss_clip": 0.01181546,
      "auxiliary_loss_mlp": 0.01024765,
      "balance_loss_clip": 1.05431557,
      "balance_loss_mlp": 1.01573455,
      "epoch": 0.43804485059820836,
      "flos": 20588305605120.0,
      "grad_norm": 3.5038742342397353,
      "language_loss": 0.69889009,
      "learning_rate": 2.4933949716232424e-06,
      "loss": 0.72095317,
      "num_input_tokens_seen": 78506785,
      "step": 3643,
      "time_per_iteration": 2.6190497875213623
    },
    {
      "auxiliary_loss_clip": 0.0118725,
      "auxiliary_loss_mlp": 0.01031512,
      "balance_loss_clip": 1.05944741,
      "balance_loss_mlp": 1.02328086,
      "epoch": 0.43816509348884747,
      "flos": 23876214981120.0,
      "grad_norm": 2.368825327611096,
      "language_loss": 0.73788762,
      "learning_rate": 2.492640038606476e-06,
      "loss": 0.76007521,
      "num_input_tokens_seen": 78525150,
      "step": 3644,
      "time_per_iteration": 2.689021348953247
    },
    {
      "auxiliary_loss_clip": 0.01185335,
      "auxiliary_loss_mlp": 0.01022133,
      "balance_loss_clip": 1.05538821,
      "balance_loss_mlp": 1.01367533,
      "epoch": 0.4382853363794866,
      "flos": 14684448533760.0,
      "grad_norm": 2.608952519431401,
      "language_loss": 0.78621292,
      "learning_rate": 2.491885030856608e-06,
      "loss": 0.80828762,
      "num_input_tokens_seen": 78543245,
      "step": 3645,
      "time_per_iteration": 2.651944637298584
    },
    {
      "auxiliary_loss_clip": 0.01185941,
      "auxiliary_loss_mlp": 0.01025628,
      "balance_loss_clip": 1.05679512,
      "balance_loss_mlp": 1.01657367,
      "epoch": 0.43840557927012563,
      "flos": 17165301177600.0,
      "grad_norm": 2.8829608245726854,
      "language_loss": 0.82952267,
      "learning_rate": 2.4911299484881713e-06,
      "loss": 0.85163832,
      "num_input_tokens_seen": 78560775,
      "step": 3646,
      "time_per_iteration": 2.652435064315796
    },
    {
      "auxiliary_loss_clip": 0.01182791,
      "auxiliary_loss_mlp": 0.01025766,
      "balance_loss_clip": 1.05419481,
      "balance_loss_mlp": 1.01766837,
      "epoch": 0.43852582216076474,
      "flos": 19390685316480.0,
      "grad_norm": 1.6757388443944785,
      "language_loss": 0.80991256,
      "learning_rate": 2.490374791615712e-06,
      "loss": 0.83199817,
      "num_input_tokens_seen": 78580800,
      "step": 3647,
      "time_per_iteration": 3.5753843784332275
    },
    {
      "auxiliary_loss_clip": 0.01189481,
      "auxiliary_loss_mlp": 0.01030273,
      "balance_loss_clip": 1.05791283,
      "balance_loss_mlp": 1.02110541,
      "epoch": 0.43864606505140386,
      "flos": 18075133699200.0,
      "grad_norm": 3.708346074926722,
      "language_loss": 0.77932763,
      "learning_rate": 2.4896195603537867e-06,
      "loss": 0.80152512,
      "num_input_tokens_seen": 78595410,
      "step": 3648,
      "time_per_iteration": 2.6293106079101562
    },
    {
      "auxiliary_loss_clip": 0.01188724,
      "auxiliary_loss_mlp": 0.01027631,
      "balance_loss_clip": 1.05988944,
      "balance_loss_mlp": 1.01941705,
      "epoch": 0.4387663079420429,
      "flos": 19644896845440.0,
      "grad_norm": 1.9982578701618607,
      "language_loss": 0.73937744,
      "learning_rate": 2.488864254816964e-06,
      "loss": 0.76154101,
      "num_input_tokens_seen": 78614100,
      "step": 3649,
      "time_per_iteration": 2.6761624813079834
    },
    {
      "auxiliary_loss_clip": 0.0118921,
      "auxiliary_loss_mlp": 0.01029915,
      "balance_loss_clip": 1.05753946,
      "balance_loss_mlp": 1.02028859,
      "epoch": 0.438886550832682,
      "flos": 19719339782400.0,
      "grad_norm": 2.8126029985992345,
      "language_loss": 0.68237364,
      "learning_rate": 2.4881088751198218e-06,
      "loss": 0.70456493,
      "num_input_tokens_seen": 78632260,
      "step": 3650,
      "time_per_iteration": 3.4957592487335205
    },
    {
      "auxiliary_loss_clip": 0.01189341,
      "auxiliary_loss_mlp": 0.01028095,
      "balance_loss_clip": 1.05739903,
      "balance_loss_mlp": 1.01870108,
      "epoch": 0.43900679372332113,
      "flos": 14536675981440.0,
      "grad_norm": 2.9543857182169937,
      "language_loss": 0.6417675,
      "learning_rate": 2.4873534213769517e-06,
      "loss": 0.66394192,
      "num_input_tokens_seen": 78647490,
      "step": 3651,
      "time_per_iteration": 3.547456979751587
    },
    {
      "auxiliary_loss_clip": 0.01186943,
      "auxiliary_loss_mlp": 0.01026991,
      "balance_loss_clip": 1.05853403,
      "balance_loss_mlp": 1.01830077,
      "epoch": 0.4391270366139602,
      "flos": 24056234968320.0,
      "grad_norm": 2.467251142755227,
      "language_loss": 0.71830875,
      "learning_rate": 2.4865978937029547e-06,
      "loss": 0.74044812,
      "num_input_tokens_seen": 78666470,
      "step": 3652,
      "time_per_iteration": 3.6004908084869385
    },
    {
      "auxiliary_loss_clip": 0.01186752,
      "auxiliary_loss_mlp": 0.01031839,
      "balance_loss_clip": 1.05860925,
      "balance_loss_mlp": 1.02311897,
      "epoch": 0.4392472795045993,
      "flos": 31538510363520.0,
      "grad_norm": 1.8354710585429828,
      "language_loss": 0.66270304,
      "learning_rate": 2.485842292212445e-06,
      "loss": 0.68488896,
      "num_input_tokens_seen": 78687685,
      "step": 3653,
      "time_per_iteration": 2.757328510284424
    },
    {
      "auxiliary_loss_clip": 0.01189779,
      "auxiliary_loss_mlp": 0.01029932,
      "balance_loss_clip": 1.05950892,
      "balance_loss_mlp": 1.02225447,
      "epoch": 0.4393675223952384,
      "flos": 14866300114560.0,
      "grad_norm": 2.430187111447149,
      "language_loss": 0.80443525,
      "learning_rate": 2.485086617020045e-06,
      "loss": 0.82663238,
      "num_input_tokens_seen": 78706180,
      "step": 3654,
      "time_per_iteration": 2.67834734916687
    },
    {
      "auxiliary_loss_clip": 0.0118466,
      "auxiliary_loss_mlp": 0.01030092,
      "balance_loss_clip": 1.05282915,
      "balance_loss_mlp": 1.02063262,
      "epoch": 0.43948776528587746,
      "flos": 14825900292480.0,
      "grad_norm": 2.2518073301154597,
      "language_loss": 0.81941885,
      "learning_rate": 2.4843308682403903e-06,
      "loss": 0.84156632,
      "num_input_tokens_seen": 78723095,
      "step": 3655,
      "time_per_iteration": 2.6457812786102295
    },
    {
      "auxiliary_loss_clip": 0.0118441,
      "auxiliary_loss_mlp": 0.010279,
      "balance_loss_clip": 1.05589318,
      "balance_loss_mlp": 1.01932287,
      "epoch": 0.4396080081765166,
      "flos": 13914523486080.0,
      "grad_norm": 2.981113743471009,
      "language_loss": 0.82932007,
      "learning_rate": 2.4835750459881294e-06,
      "loss": 0.85144317,
      "num_input_tokens_seen": 78739720,
      "step": 3656,
      "time_per_iteration": 2.6193385124206543
    },
    {
      "auxiliary_loss_clip": 0.01188027,
      "auxiliary_loss_mlp": 0.01031826,
      "balance_loss_clip": 1.05849111,
      "balance_loss_mlp": 1.02349889,
      "epoch": 0.43972825106715563,
      "flos": 18222978078720.0,
      "grad_norm": 3.978050200517701,
      "language_loss": 0.82455146,
      "learning_rate": 2.4828191503779177e-06,
      "loss": 0.84675002,
      "num_input_tokens_seen": 78757820,
      "step": 3657,
      "time_per_iteration": 2.566823959350586
    },
    {
      "auxiliary_loss_clip": 0.0118499,
      "auxiliary_loss_mlp": 0.0103205,
      "balance_loss_clip": 1.05476749,
      "balance_loss_mlp": 1.0234549,
      "epoch": 0.43984849395779474,
      "flos": 16873239692160.0,
      "grad_norm": 2.8920886513857496,
      "language_loss": 0.89606154,
      "learning_rate": 2.482063181524425e-06,
      "loss": 0.91823202,
      "num_input_tokens_seen": 78773720,
      "step": 3658,
      "time_per_iteration": 2.636204957962036
    },
    {
      "auxiliary_loss_clip": 0.01189585,
      "auxiliary_loss_mlp": 0.01030737,
      "balance_loss_clip": 1.05877399,
      "balance_loss_mlp": 1.02178442,
      "epoch": 0.43996873684843385,
      "flos": 18691504104960.0,
      "grad_norm": 3.1743827719234394,
      "language_loss": 0.80889624,
      "learning_rate": 2.4813071395423307e-06,
      "loss": 0.83109945,
      "num_input_tokens_seen": 78791285,
      "step": 3659,
      "time_per_iteration": 2.692241907119751
    },
    {
      "auxiliary_loss_clip": 0.01183756,
      "auxiliary_loss_mlp": 0.01029473,
      "balance_loss_clip": 1.05361509,
      "balance_loss_mlp": 1.02032351,
      "epoch": 0.4400889797390729,
      "flos": 23653460787840.0,
      "grad_norm": 1.6708962943928907,
      "language_loss": 0.64214849,
      "learning_rate": 2.4805510245463263e-06,
      "loss": 0.66428077,
      "num_input_tokens_seen": 78811440,
      "step": 3660,
      "time_per_iteration": 2.6693427562713623
    },
    {
      "auxiliary_loss_clip": 0.01184473,
      "auxiliary_loss_mlp": 0.01027864,
      "balance_loss_clip": 1.0550878,
      "balance_loss_mlp": 1.0192517,
      "epoch": 0.440209222629712,
      "flos": 23149203707520.0,
      "grad_norm": 2.212400071839482,
      "language_loss": 0.60220659,
      "learning_rate": 2.4797948366511137e-06,
      "loss": 0.62432992,
      "num_input_tokens_seen": 78831150,
      "step": 3661,
      "time_per_iteration": 2.683852195739746
    },
    {
      "auxiliary_loss_clip": 0.011831,
      "auxiliary_loss_mlp": 0.01029159,
      "balance_loss_clip": 1.05344915,
      "balance_loss_mlp": 1.02023625,
      "epoch": 0.4403294655203511,
      "flos": 24823394668800.0,
      "grad_norm": 2.0434034718738,
      "language_loss": 0.75959384,
      "learning_rate": 2.4790385759714055e-06,
      "loss": 0.78171647,
      "num_input_tokens_seen": 78850215,
      "step": 3662,
      "time_per_iteration": 2.6666858196258545
    },
    {
      "auxiliary_loss_clip": 0.01184987,
      "auxiliary_loss_mlp": 0.01030456,
      "balance_loss_clip": 1.05624866,
      "balance_loss_mlp": 1.02226603,
      "epoch": 0.4404497084109902,
      "flos": 22565080736640.0,
      "grad_norm": 1.7792053196785178,
      "language_loss": 0.70979691,
      "learning_rate": 2.478282242621926e-06,
      "loss": 0.7319513,
      "num_input_tokens_seen": 78870675,
      "step": 3663,
      "time_per_iteration": 2.6798884868621826
    },
    {
      "auxiliary_loss_clip": 0.0108853,
      "auxiliary_loss_mlp": 0.01004214,
      "balance_loss_clip": 1.02414274,
      "balance_loss_mlp": 1.00295687,
      "epoch": 0.4405699513016293,
      "flos": 64967073448320.0,
      "grad_norm": 0.8454370344528872,
      "language_loss": 0.59543729,
      "learning_rate": 2.477525836717411e-06,
      "loss": 0.61636472,
      "num_input_tokens_seen": 78938440,
      "step": 3664,
      "time_per_iteration": 3.3220624923706055
    },
    {
      "auxiliary_loss_clip": 0.01186022,
      "auxiliary_loss_mlp": 0.01030929,
      "balance_loss_clip": 1.0550611,
      "balance_loss_mlp": 1.02257252,
      "epoch": 0.4406901941922684,
      "flos": 35661952978560.0,
      "grad_norm": 2.2248679759505334,
      "language_loss": 0.79741251,
      "learning_rate": 2.476769358372606e-06,
      "loss": 0.81958199,
      "num_input_tokens_seen": 78960090,
      "step": 3665,
      "time_per_iteration": 2.7371861934661865
    },
    {
      "auxiliary_loss_clip": 0.01184187,
      "auxiliary_loss_mlp": 0.01030721,
      "balance_loss_clip": 1.05687594,
      "balance_loss_mlp": 1.02217984,
      "epoch": 0.44081043708290746,
      "flos": 18040767361920.0,
      "grad_norm": 3.0183125517269125,
      "language_loss": 0.74937576,
      "learning_rate": 2.4760128077022683e-06,
      "loss": 0.77152485,
      "num_input_tokens_seen": 78978225,
      "step": 3666,
      "time_per_iteration": 2.5798823833465576
    },
    {
      "auxiliary_loss_clip": 0.01184063,
      "auxiliary_loss_mlp": 0.01024964,
      "balance_loss_clip": 1.0559535,
      "balance_loss_mlp": 1.016765,
      "epoch": 0.44093067997354657,
      "flos": 30153507799680.0,
      "grad_norm": 1.5557457296296127,
      "language_loss": 0.68660426,
      "learning_rate": 2.4752561848211672e-06,
      "loss": 0.70869452,
      "num_input_tokens_seen": 79000625,
      "step": 3667,
      "time_per_iteration": 2.7067177295684814
    },
    {
      "auxiliary_loss_clip": 0.01183492,
      "auxiliary_loss_mlp": 0.01027736,
      "balance_loss_clip": 1.05796039,
      "balance_loss_mlp": 1.02027655,
      "epoch": 0.4410509228641857,
      "flos": 23255068066560.0,
      "grad_norm": 2.13251340134267,
      "language_loss": 0.71222979,
      "learning_rate": 2.4744994898440797e-06,
      "loss": 0.73434204,
      "num_input_tokens_seen": 79019415,
      "step": 3668,
      "time_per_iteration": 2.6065945625305176
    },
    {
      "auxiliary_loss_clip": 0.01189683,
      "auxiliary_loss_mlp": 0.01030692,
      "balance_loss_clip": 1.05739784,
      "balance_loss_mlp": 1.02166784,
      "epoch": 0.44117116575482473,
      "flos": 19500571998720.0,
      "grad_norm": 4.929284794945199,
      "language_loss": 0.83887166,
      "learning_rate": 2.473742722885797e-06,
      "loss": 0.8610754,
      "num_input_tokens_seen": 79038435,
      "step": 3669,
      "time_per_iteration": 2.6686689853668213
    },
    {
      "auxiliary_loss_clip": 0.01187932,
      "auxiliary_loss_mlp": 0.01033626,
      "balance_loss_clip": 1.05740595,
      "balance_loss_mlp": 1.02444124,
      "epoch": 0.44129140864546385,
      "flos": 27053124353280.0,
      "grad_norm": 2.8468959053714595,
      "language_loss": 0.65130079,
      "learning_rate": 2.4729858840611197e-06,
      "loss": 0.67351633,
      "num_input_tokens_seen": 79057345,
      "step": 3670,
      "time_per_iteration": 2.7572882175445557
    },
    {
      "auxiliary_loss_clip": 0.01185341,
      "auxiliary_loss_mlp": 0.01027057,
      "balance_loss_clip": 1.05798006,
      "balance_loss_mlp": 1.01849747,
      "epoch": 0.4414116515361029,
      "flos": 26102101910400.0,
      "grad_norm": 2.4061181101964904,
      "language_loss": 0.72671986,
      "learning_rate": 2.4722289734848605e-06,
      "loss": 0.74884385,
      "num_input_tokens_seen": 79077810,
      "step": 3671,
      "time_per_iteration": 2.677884340286255
    },
    {
      "auxiliary_loss_clip": 0.01184009,
      "auxiliary_loss_mlp": 0.01026022,
      "balance_loss_clip": 1.05536449,
      "balance_loss_mlp": 1.01801729,
      "epoch": 0.441531894426742,
      "flos": 21906083865600.0,
      "grad_norm": 2.51725107701066,
      "language_loss": 0.7775234,
      "learning_rate": 2.471471991271841e-06,
      "loss": 0.79962379,
      "num_input_tokens_seen": 79094935,
      "step": 3672,
      "time_per_iteration": 2.645136594772339
    },
    {
      "auxiliary_loss_clip": 0.0118644,
      "auxiliary_loss_mlp": 0.01035064,
      "balance_loss_clip": 1.05583,
      "balance_loss_mlp": 1.02590275,
      "epoch": 0.4416521373173811,
      "flos": 23437099215360.0,
      "grad_norm": 1.930535032416054,
      "language_loss": 0.79160905,
      "learning_rate": 2.470714937536896e-06,
      "loss": 0.81382418,
      "num_input_tokens_seen": 79113660,
      "step": 3673,
      "time_per_iteration": 3.5766491889953613
    },
    {
      "auxiliary_loss_clip": 0.01187667,
      "auxiliary_loss_mlp": 0.01026853,
      "balance_loss_clip": 1.05718684,
      "balance_loss_mlp": 1.01769221,
      "epoch": 0.4417723802080202,
      "flos": 20334345471360.0,
      "grad_norm": 2.1422343106684463,
      "language_loss": 0.70288426,
      "learning_rate": 2.469957812394868e-06,
      "loss": 0.72502947,
      "num_input_tokens_seen": 79132470,
      "step": 3674,
      "time_per_iteration": 2.6293556690216064
    },
    {
      "auxiliary_loss_clip": 0.01184375,
      "auxiliary_loss_mlp": 0.01031217,
      "balance_loss_clip": 1.05719054,
      "balance_loss_mlp": 1.02312875,
      "epoch": 0.4418926230986593,
      "flos": 18880682060160.0,
      "grad_norm": 2.37260470619038,
      "language_loss": 0.76085258,
      "learning_rate": 2.4692006159606148e-06,
      "loss": 0.78300846,
      "num_input_tokens_seen": 79150000,
      "step": 3675,
      "time_per_iteration": 2.6215009689331055
    },
    {
      "auxiliary_loss_clip": 0.01186466,
      "auxiliary_loss_mlp": 0.01030998,
      "balance_loss_clip": 1.056916,
      "balance_loss_mlp": 1.02241504,
      "epoch": 0.4420128659892984,
      "flos": 19464409981440.0,
      "grad_norm": 2.051280108128422,
      "language_loss": 0.78622532,
      "learning_rate": 2.468443348349e-06,
      "loss": 0.80839992,
      "num_input_tokens_seen": 79167875,
      "step": 3676,
      "time_per_iteration": 3.6533243656158447
    },
    {
      "auxiliary_loss_clip": 0.01189052,
      "auxiliary_loss_mlp": 0.01030875,
      "balance_loss_clip": 1.05801129,
      "balance_loss_mlp": 1.02229762,
      "epoch": 0.44213310887993745,
      "flos": 17894359526400.0,
      "grad_norm": 2.3608096797334825,
      "language_loss": 0.82494187,
      "learning_rate": 2.467686009674902e-06,
      "loss": 0.84714115,
      "num_input_tokens_seen": 79182325,
      "step": 3677,
      "time_per_iteration": 3.5600433349609375
    },
    {
      "auxiliary_loss_clip": 0.01188526,
      "auxiliary_loss_mlp": 0.01030495,
      "balance_loss_clip": 1.05684614,
      "balance_loss_mlp": 1.02092886,
      "epoch": 0.44225335177057656,
      "flos": 19204667758080.0,
      "grad_norm": 2.070612757103046,
      "language_loss": 0.85379362,
      "learning_rate": 2.466928600053209e-06,
      "loss": 0.87598389,
      "num_input_tokens_seen": 79197630,
      "step": 3678,
      "time_per_iteration": 2.619086980819702
    },
    {
      "auxiliary_loss_clip": 0.01185697,
      "auxiliary_loss_mlp": 0.01031833,
      "balance_loss_clip": 1.0555315,
      "balance_loss_mlp": 1.02354836,
      "epoch": 0.4423735946612157,
      "flos": 23471321898240.0,
      "grad_norm": 3.9626889990095573,
      "language_loss": 0.71402037,
      "learning_rate": 2.466171119598818e-06,
      "loss": 0.73619568,
      "num_input_tokens_seen": 79217600,
      "step": 3679,
      "time_per_iteration": 3.568192720413208
    },
    {
      "auxiliary_loss_clip": 0.01190876,
      "auxiliary_loss_mlp": 0.01024678,
      "balance_loss_clip": 1.05706728,
      "balance_loss_mlp": 1.01504588,
      "epoch": 0.44249383755185473,
      "flos": 26685398868480.0,
      "grad_norm": 5.245782769939176,
      "language_loss": 0.76940596,
      "learning_rate": 2.465413568426639e-06,
      "loss": 0.79156148,
      "num_input_tokens_seen": 79238550,
      "step": 3680,
      "time_per_iteration": 2.6591362953186035
    },
    {
      "auxiliary_loss_clip": 0.01181952,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.05396235,
      "balance_loss_mlp": 1.01770616,
      "epoch": 0.44261408044249384,
      "flos": 23147659422720.0,
      "grad_norm": 1.6296400618949323,
      "language_loss": 0.81127834,
      "learning_rate": 2.464655946651591e-06,
      "loss": 0.83335072,
      "num_input_tokens_seen": 79257555,
      "step": 3681,
      "time_per_iteration": 2.588738203048706
    },
    {
      "auxiliary_loss_clip": 0.01186977,
      "auxiliary_loss_mlp": 0.01028385,
      "balance_loss_clip": 1.05631721,
      "balance_loss_mlp": 1.01968908,
      "epoch": 0.44273432333313295,
      "flos": 24462564595200.0,
      "grad_norm": 2.0323811767754862,
      "language_loss": 0.80968773,
      "learning_rate": 2.4638982543886065e-06,
      "loss": 0.83184135,
      "num_input_tokens_seen": 79277595,
      "step": 3682,
      "time_per_iteration": 2.6728246212005615
    },
    {
      "auxiliary_loss_clip": 0.01189919,
      "auxiliary_loss_mlp": 0.01029348,
      "balance_loss_clip": 1.05880737,
      "balance_loss_mlp": 1.02068734,
      "epoch": 0.442854566223772,
      "flos": 17528932512000.0,
      "grad_norm": 2.2979403352964116,
      "language_loss": 0.87003332,
      "learning_rate": 2.4631404917526254e-06,
      "loss": 0.89222598,
      "num_input_tokens_seen": 79294550,
      "step": 3683,
      "time_per_iteration": 2.606444835662842
    },
    {
      "auxiliary_loss_clip": 0.0118285,
      "auxiliary_loss_mlp": 0.0102706,
      "balance_loss_clip": 1.05452335,
      "balance_loss_mlp": 1.01904297,
      "epoch": 0.4429748091144111,
      "flos": 24896293320960.0,
      "grad_norm": 1.642446334930872,
      "language_loss": 0.79167461,
      "learning_rate": 2.4623826588586e-06,
      "loss": 0.81377375,
      "num_input_tokens_seen": 79314820,
      "step": 3684,
      "time_per_iteration": 2.6609041690826416
    },
    {
      "auxiliary_loss_clip": 0.0118616,
      "auxiliary_loss_mlp": 0.0102768,
      "balance_loss_clip": 1.05443549,
      "balance_loss_mlp": 1.01859629,
      "epoch": 0.4430950520050502,
      "flos": 21614704738560.0,
      "grad_norm": 1.7795338149094155,
      "language_loss": 0.82834995,
      "learning_rate": 2.461624755821492e-06,
      "loss": 0.85048831,
      "num_input_tokens_seen": 79334300,
      "step": 3685,
      "time_per_iteration": 2.602560520172119
    },
    {
      "auxiliary_loss_clip": 0.01184493,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.05699599,
      "balance_loss_mlp": 1.01852655,
      "epoch": 0.4432152948956893,
      "flos": 24572271709440.0,
      "grad_norm": 1.8754520350455997,
      "language_loss": 0.76350182,
      "learning_rate": 2.4608667827562763e-06,
      "loss": 0.78561962,
      "num_input_tokens_seen": 79353630,
      "step": 3686,
      "time_per_iteration": 2.714421033859253
    },
    {
      "auxiliary_loss_clip": 0.01194658,
      "auxiliary_loss_mlp": 0.01029314,
      "balance_loss_clip": 1.06030858,
      "balance_loss_mlp": 1.02069557,
      "epoch": 0.4433355377863284,
      "flos": 21762261809280.0,
      "grad_norm": 2.6391718067215066,
      "language_loss": 0.90032893,
      "learning_rate": 2.460108739777936e-06,
      "loss": 0.92256862,
      "num_input_tokens_seen": 79372765,
      "step": 3687,
      "time_per_iteration": 2.596418619155884
    },
    {
      "auxiliary_loss_clip": 0.01186586,
      "auxiliary_loss_mlp": 0.0103074,
      "balance_loss_clip": 1.05674851,
      "balance_loss_mlp": 1.02257156,
      "epoch": 0.44345578067696745,
      "flos": 20084479488000.0,
      "grad_norm": 2.1703939148471845,
      "language_loss": 0.76189727,
      "learning_rate": 2.4593506270014656e-06,
      "loss": 0.78407049,
      "num_input_tokens_seen": 79391735,
      "step": 3688,
      "time_per_iteration": 2.6619279384613037
    },
    {
      "auxiliary_loss_clip": 0.01185577,
      "auxiliary_loss_mlp": 0.01029153,
      "balance_loss_clip": 1.05525017,
      "balance_loss_mlp": 1.02037907,
      "epoch": 0.44357602356760656,
      "flos": 24169497528960.0,
      "grad_norm": 1.92297405526017,
      "language_loss": 0.82165837,
      "learning_rate": 2.45859244454187e-06,
      "loss": 0.84380567,
      "num_input_tokens_seen": 79411525,
      "step": 3689,
      "time_per_iteration": 2.6611881256103516
    },
    {
      "auxiliary_loss_clip": 0.01186124,
      "auxiliary_loss_mlp": 0.01027431,
      "balance_loss_clip": 1.05727553,
      "balance_loss_mlp": 1.020123,
      "epoch": 0.44369626645824567,
      "flos": 22707717644160.0,
      "grad_norm": 1.683571224862422,
      "language_loss": 0.66404784,
      "learning_rate": 2.4578341925141655e-06,
      "loss": 0.68618345,
      "num_input_tokens_seen": 79430740,
      "step": 3690,
      "time_per_iteration": 2.7449162006378174
    },
    {
      "auxiliary_loss_clip": 0.01189684,
      "auxiliary_loss_mlp": 0.01029047,
      "balance_loss_clip": 1.05860019,
      "balance_loss_mlp": 1.01958752,
      "epoch": 0.4438165093488847,
      "flos": 38030225420160.0,
      "grad_norm": 3.386179797200325,
      "language_loss": 0.72449803,
      "learning_rate": 2.457075871033378e-06,
      "loss": 0.74668533,
      "num_input_tokens_seen": 79452615,
      "step": 3691,
      "time_per_iteration": 2.793870449066162
    },
    {
      "auxiliary_loss_clip": 0.01183085,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.05486798,
      "balance_loss_mlp": 1.02211106,
      "epoch": 0.44393675223952384,
      "flos": 15523213996800.0,
      "grad_norm": 2.5329848722634103,
      "language_loss": 0.88811862,
      "learning_rate": 2.4563174802145445e-06,
      "loss": 0.91024995,
      "num_input_tokens_seen": 79469865,
      "step": 3692,
      "time_per_iteration": 2.622969388961792
    },
    {
      "auxiliary_loss_clip": 0.01089516,
      "auxiliary_loss_mlp": 0.01001213,
      "balance_loss_clip": 1.02517891,
      "balance_loss_mlp": 0.99998552,
      "epoch": 0.44405699513016295,
      "flos": 64574893779840.0,
      "grad_norm": 0.6463750988088204,
      "language_loss": 0.48587126,
      "learning_rate": 2.455559020172712e-06,
      "loss": 0.50677854,
      "num_input_tokens_seen": 79537220,
      "step": 3693,
      "time_per_iteration": 3.2980589866638184
    },
    {
      "auxiliary_loss_clip": 0.0118973,
      "auxiliary_loss_mlp": 0.01029531,
      "balance_loss_clip": 1.06022656,
      "balance_loss_mlp": 1.02038205,
      "epoch": 0.444177238020802,
      "flos": 23987394552960.0,
      "grad_norm": 2.4754683447516896,
      "language_loss": 0.89816129,
      "learning_rate": 2.4548004910229385e-06,
      "loss": 0.92035389,
      "num_input_tokens_seen": 79554795,
      "step": 3694,
      "time_per_iteration": 2.647275447845459
    },
    {
      "auxiliary_loss_clip": 0.01186281,
      "auxiliary_loss_mlp": 0.01029672,
      "balance_loss_clip": 1.05589366,
      "balance_loss_mlp": 1.02099395,
      "epoch": 0.4442974809114411,
      "flos": 22563069575040.0,
      "grad_norm": 2.2813747916609866,
      "language_loss": 0.87047935,
      "learning_rate": 2.4540418928802913e-06,
      "loss": 0.89263892,
      "num_input_tokens_seen": 79573530,
      "step": 3695,
      "time_per_iteration": 2.6584930419921875
    },
    {
      "auxiliary_loss_clip": 0.01190178,
      "auxiliary_loss_mlp": 0.01037669,
      "balance_loss_clip": 1.05873132,
      "balance_loss_mlp": 1.02866864,
      "epoch": 0.4444177238020802,
      "flos": 17675699483520.0,
      "grad_norm": 3.254078294379325,
      "language_loss": 0.65906358,
      "learning_rate": 2.4532832258598506e-06,
      "loss": 0.68134201,
      "num_input_tokens_seen": 79591360,
      "step": 3696,
      "time_per_iteration": 2.757523536682129
    },
    {
      "auxiliary_loss_clip": 0.01184983,
      "auxiliary_loss_mlp": 0.0102218,
      "balance_loss_clip": 1.05787146,
      "balance_loss_mlp": 1.01454508,
      "epoch": 0.4445379666927193,
      "flos": 28621594609920.0,
      "grad_norm": 1.736172249568541,
      "language_loss": 0.80989379,
      "learning_rate": 2.4525244900767047e-06,
      "loss": 0.83196545,
      "num_input_tokens_seen": 79612175,
      "step": 3697,
      "time_per_iteration": 2.668424129486084
    },
    {
      "auxiliary_loss_clip": 0.01087728,
      "auxiliary_loss_mlp": 0.01000803,
      "balance_loss_clip": 1.0233953,
      "balance_loss_mlp": 0.99961054,
      "epoch": 0.4446582095833584,
      "flos": 70487370115200.0,
      "grad_norm": 0.7709074960997224,
      "language_loss": 0.60500032,
      "learning_rate": 2.4517656856459536e-06,
      "loss": 0.62588561,
      "num_input_tokens_seen": 79678020,
      "step": 3698,
      "time_per_iteration": 3.2708048820495605
    },
    {
      "auxiliary_loss_clip": 0.0118189,
      "auxiliary_loss_mlp": 0.01024638,
      "balance_loss_clip": 1.05407929,
      "balance_loss_mlp": 1.01581645,
      "epoch": 0.4447784524739975,
      "flos": 26505199313280.0,
      "grad_norm": 1.7820078638729737,
      "language_loss": 0.6808877,
      "learning_rate": 2.4510068126827073e-06,
      "loss": 0.70295298,
      "num_input_tokens_seen": 79699020,
      "step": 3699,
      "time_per_iteration": 2.662149667739868
    },
    {
      "auxiliary_loss_clip": 0.01188393,
      "auxiliary_loss_mlp": 0.01024418,
      "balance_loss_clip": 1.05856824,
      "balance_loss_mlp": 1.01618052,
      "epoch": 0.44489869536463655,
      "flos": 11656209553920.0,
      "grad_norm": 2.7525232536367,
      "language_loss": 0.81646407,
      "learning_rate": 2.450247871302086e-06,
      "loss": 0.83859217,
      "num_input_tokens_seen": 79716795,
      "step": 3700,
      "time_per_iteration": 3.605025291442871
    },
    {
      "auxiliary_loss_clip": 0.01184442,
      "auxiliary_loss_mlp": 0.01030271,
      "balance_loss_clip": 1.05535924,
      "balance_loss_mlp": 1.02246916,
      "epoch": 0.44501893825527566,
      "flos": 20448469958400.0,
      "grad_norm": 2.543211756593869,
      "language_loss": 0.83716178,
      "learning_rate": 2.44948886161922e-06,
      "loss": 0.85930896,
      "num_input_tokens_seen": 79735810,
      "step": 3701,
      "time_per_iteration": 2.545248508453369
    },
    {
      "auxiliary_loss_clip": 0.01185889,
      "auxiliary_loss_mlp": 0.01030134,
      "balance_loss_clip": 1.05776525,
      "balance_loss_mlp": 1.02220345,
      "epoch": 0.4451391811459148,
      "flos": 18261079430400.0,
      "grad_norm": 1.7571141031438289,
      "language_loss": 0.84744203,
      "learning_rate": 2.4487297837492524e-06,
      "loss": 0.8696022,
      "num_input_tokens_seen": 79754975,
      "step": 3702,
      "time_per_iteration": 2.6081502437591553
    },
    {
      "auxiliary_loss_clip": 0.01184973,
      "auxiliary_loss_mlp": 0.01028477,
      "balance_loss_clip": 1.05474639,
      "balance_loss_mlp": 1.02000117,
      "epoch": 0.44525942403655383,
      "flos": 16910155895040.0,
      "grad_norm": 2.305304798312761,
      "language_loss": 0.61864948,
      "learning_rate": 2.4479706378073323e-06,
      "loss": 0.64078397,
      "num_input_tokens_seen": 79773515,
      "step": 3703,
      "time_per_iteration": 3.5102686882019043
    },
    {
      "auxiliary_loss_clip": 0.01183873,
      "auxiliary_loss_mlp": 0.01027467,
      "balance_loss_clip": 1.05645525,
      "balance_loss_mlp": 1.01998377,
      "epoch": 0.44537966692719294,
      "flos": 23258838994560.0,
      "grad_norm": 1.5385162577075542,
      "language_loss": 0.83871984,
      "learning_rate": 2.447211423908623e-06,
      "loss": 0.86083329,
      "num_input_tokens_seen": 79793560,
      "step": 3704,
      "time_per_iteration": 3.5131094455718994
    },
    {
      "auxiliary_loss_clip": 0.0118286,
      "auxiliary_loss_mlp": 0.01034098,
      "balance_loss_clip": 1.05559707,
      "balance_loss_mlp": 1.02595568,
      "epoch": 0.445499909817832,
      "flos": 21724160457600.0,
      "grad_norm": 1.9315754666965845,
      "language_loss": 0.74713707,
      "learning_rate": 2.4464521421682966e-06,
      "loss": 0.76930666,
      "num_input_tokens_seen": 79811150,
      "step": 3705,
      "time_per_iteration": 3.530097723007202
    },
    {
      "auxiliary_loss_clip": 0.01182404,
      "auxiliary_loss_mlp": 0.01032664,
      "balance_loss_clip": 1.05699873,
      "balance_loss_mlp": 1.02489769,
      "epoch": 0.4456201527084711,
      "flos": 23987969170560.0,
      "grad_norm": 3.819825125044915,
      "language_loss": 0.87528175,
      "learning_rate": 2.4456927927015345e-06,
      "loss": 0.89743245,
      "num_input_tokens_seen": 79832190,
      "step": 3706,
      "time_per_iteration": 2.633157253265381
    },
    {
      "auxiliary_loss_clip": 0.01190506,
      "auxiliary_loss_mlp": 0.01028574,
      "balance_loss_clip": 1.058007,
      "balance_loss_mlp": 1.01919806,
      "epoch": 0.4457403955991102,
      "flos": 18807065136000.0,
      "grad_norm": 3.1888083521312653,
      "language_loss": 0.76290679,
      "learning_rate": 2.4449333756235307e-06,
      "loss": 0.7850976,
      "num_input_tokens_seen": 79848905,
      "step": 3707,
      "time_per_iteration": 2.6932260990142822
    },
    {
      "auxiliary_loss_clip": 0.01184723,
      "auxiliary_loss_mlp": 0.010275,
      "balance_loss_clip": 1.0576278,
      "balance_loss_mlp": 1.01945305,
      "epoch": 0.4458606384897493,
      "flos": 19207756327680.0,
      "grad_norm": 2.315261354666771,
      "language_loss": 0.78616685,
      "learning_rate": 2.4441738910494876e-06,
      "loss": 0.80828905,
      "num_input_tokens_seen": 79863640,
      "step": 3708,
      "time_per_iteration": 2.6049892902374268
    },
    {
      "auxiliary_loss_clip": 0.01186922,
      "auxiliary_loss_mlp": 0.01030023,
      "balance_loss_clip": 1.05714524,
      "balance_loss_mlp": 1.02155948,
      "epoch": 0.4459808813803884,
      "flos": 21361283308800.0,
      "grad_norm": 2.92812099157235,
      "language_loss": 0.81834173,
      "learning_rate": 2.4434143390946176e-06,
      "loss": 0.8405112,
      "num_input_tokens_seen": 79882450,
      "step": 3709,
      "time_per_iteration": 2.6023759841918945
    },
    {
      "auxiliary_loss_clip": 0.01184957,
      "auxiliary_loss_mlp": 0.01024366,
      "balance_loss_clip": 1.05695629,
      "balance_loss_mlp": 1.01633453,
      "epoch": 0.4461011242710275,
      "flos": 23288967527040.0,
      "grad_norm": 1.900671332487748,
      "language_loss": 0.85626203,
      "learning_rate": 2.4426547198741457e-06,
      "loss": 0.87835521,
      "num_input_tokens_seen": 79900655,
      "step": 3710,
      "time_per_iteration": 2.640902519226074
    },
    {
      "auxiliary_loss_clip": 0.01192322,
      "auxiliary_loss_mlp": 0.01032906,
      "balance_loss_clip": 1.06014609,
      "balance_loss_mlp": 1.02429342,
      "epoch": 0.44622136716166655,
      "flos": 20193001453440.0,
      "grad_norm": 2.3638588454936045,
      "language_loss": 0.7500819,
      "learning_rate": 2.441895033503305e-06,
      "loss": 0.77233422,
      "num_input_tokens_seen": 79918575,
      "step": 3711,
      "time_per_iteration": 2.829312801361084
    },
    {
      "auxiliary_loss_clip": 0.01186603,
      "auxiliary_loss_mlp": 0.01031585,
      "balance_loss_clip": 1.05763173,
      "balance_loss_mlp": 1.02246535,
      "epoch": 0.44634161005230566,
      "flos": 21283033530240.0,
      "grad_norm": 1.7234532048897397,
      "language_loss": 0.82111192,
      "learning_rate": 2.4411352800973375e-06,
      "loss": 0.84329373,
      "num_input_tokens_seen": 79937010,
      "step": 3712,
      "time_per_iteration": 2.727015256881714
    },
    {
      "auxiliary_loss_clip": 0.01185822,
      "auxiliary_loss_mlp": 0.01029632,
      "balance_loss_clip": 1.05597484,
      "balance_loss_mlp": 1.02101874,
      "epoch": 0.44646185294294477,
      "flos": 22929358515840.0,
      "grad_norm": 4.223047741424549,
      "language_loss": 0.75344288,
      "learning_rate": 2.4403754597715005e-06,
      "loss": 0.77559745,
      "num_input_tokens_seen": 79956455,
      "step": 3713,
      "time_per_iteration": 2.645366668701172
    },
    {
      "auxiliary_loss_clip": 0.01189101,
      "auxiliary_loss_mlp": 0.01031118,
      "balance_loss_clip": 1.05727959,
      "balance_loss_mlp": 1.02201653,
      "epoch": 0.4465820958335838,
      "flos": 22637692080000.0,
      "grad_norm": 2.1911667579740093,
      "language_loss": 0.93093467,
      "learning_rate": 2.4396155726410553e-06,
      "loss": 0.95313686,
      "num_input_tokens_seen": 79975065,
      "step": 3714,
      "time_per_iteration": 2.6798152923583984
    },
    {
      "auxiliary_loss_clip": 0.01187572,
      "auxiliary_loss_mlp": 0.01027786,
      "balance_loss_clip": 1.05631256,
      "balance_loss_mlp": 1.01927483,
      "epoch": 0.44670233872422294,
      "flos": 22672525294080.0,
      "grad_norm": 2.8138289164716848,
      "language_loss": 0.9090516,
      "learning_rate": 2.438855618821278e-06,
      "loss": 0.93120515,
      "num_input_tokens_seen": 79990865,
      "step": 3715,
      "time_per_iteration": 2.606753349304199
    },
    {
      "auxiliary_loss_clip": 0.01182584,
      "auxiliary_loss_mlp": 0.01023551,
      "balance_loss_clip": 1.05362415,
      "balance_loss_mlp": 1.0150336,
      "epoch": 0.44682258161486205,
      "flos": 23582178247680.0,
      "grad_norm": 2.01216786411553,
      "language_loss": 0.67855978,
      "learning_rate": 2.4380955984274517e-06,
      "loss": 0.70062113,
      "num_input_tokens_seen": 80009520,
      "step": 3716,
      "time_per_iteration": 2.664210319519043
    },
    {
      "auxiliary_loss_clip": 0.01189392,
      "auxiliary_loss_mlp": 0.01028316,
      "balance_loss_clip": 1.05843019,
      "balance_loss_mlp": 1.019256,
      "epoch": 0.4469428245055011,
      "flos": 26501356558080.0,
      "grad_norm": 2.0973596955619174,
      "language_loss": 0.7700125,
      "learning_rate": 2.4373355115748716e-06,
      "loss": 0.7921896,
      "num_input_tokens_seen": 80030350,
      "step": 3717,
      "time_per_iteration": 2.6463944911956787
    },
    {
      "auxiliary_loss_clip": 0.011854,
      "auxiliary_loss_mlp": 0.01026763,
      "balance_loss_clip": 1.05710948,
      "balance_loss_mlp": 1.01819229,
      "epoch": 0.4470630673961402,
      "flos": 21504925797120.0,
      "grad_norm": 3.054623927137393,
      "language_loss": 0.72399449,
      "learning_rate": 2.436575358378842e-06,
      "loss": 0.74611616,
      "num_input_tokens_seen": 80049840,
      "step": 3718,
      "time_per_iteration": 2.6552534103393555
    },
    {
      "auxiliary_loss_clip": 0.01187602,
      "auxiliary_loss_mlp": 0.01029716,
      "balance_loss_clip": 1.0569135,
      "balance_loss_mlp": 1.02081144,
      "epoch": 0.44718331028677927,
      "flos": 16173986653440.0,
      "grad_norm": 11.184793427012105,
      "language_loss": 0.82677335,
      "learning_rate": 2.4358151389546782e-06,
      "loss": 0.84894651,
      "num_input_tokens_seen": 80066525,
      "step": 3719,
      "time_per_iteration": 2.618352174758911
    },
    {
      "auxiliary_loss_clip": 0.01188091,
      "auxiliary_loss_mlp": 0.01028242,
      "balance_loss_clip": 1.05839658,
      "balance_loss_mlp": 1.01999855,
      "epoch": 0.4473035531774184,
      "flos": 19681238430720.0,
      "grad_norm": 6.646569617663685,
      "language_loss": 0.75609708,
      "learning_rate": 2.4350548534177035e-06,
      "loss": 0.77826035,
      "num_input_tokens_seen": 80083355,
      "step": 3720,
      "time_per_iteration": 2.6465675830841064
    },
    {
      "auxiliary_loss_clip": 0.01181804,
      "auxiliary_loss_mlp": 0.01028461,
      "balance_loss_clip": 1.05493188,
      "balance_loss_mlp": 1.02064037,
      "epoch": 0.4474237960680575,
      "flos": 41427590515200.0,
      "grad_norm": 3.9661516808402757,
      "language_loss": 0.66523373,
      "learning_rate": 2.434294501883254e-06,
      "loss": 0.68733639,
      "num_input_tokens_seen": 80106450,
      "step": 3721,
      "time_per_iteration": 2.7891957759857178
    },
    {
      "auxiliary_loss_clip": 0.01184642,
      "auxiliary_loss_mlp": 0.01029965,
      "balance_loss_clip": 1.05704451,
      "balance_loss_mlp": 1.02094698,
      "epoch": 0.44754403895869654,
      "flos": 22891328991360.0,
      "grad_norm": 1.7094134141925459,
      "language_loss": 0.65816826,
      "learning_rate": 2.433534084466674e-06,
      "loss": 0.6803143,
      "num_input_tokens_seen": 80125670,
      "step": 3722,
      "time_per_iteration": 2.6509487628936768
    },
    {
      "auxiliary_loss_clip": 0.01182085,
      "auxiliary_loss_mlp": 0.01026006,
      "balance_loss_clip": 1.05508375,
      "balance_loss_mlp": 1.0179801,
      "epoch": 0.44766428184933565,
      "flos": 25630271832960.0,
      "grad_norm": 2.9561556141561653,
      "language_loss": 0.70998746,
      "learning_rate": 2.4327736012833178e-06,
      "loss": 0.73206836,
      "num_input_tokens_seen": 80147390,
      "step": 3723,
      "time_per_iteration": 2.6831820011138916
    },
    {
      "auxiliary_loss_clip": 0.0118515,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 1.05534279,
      "balance_loss_mlp": 1.01842523,
      "epoch": 0.44778452473997477,
      "flos": 20448972748800.0,
      "grad_norm": 2.482690887240029,
      "language_loss": 0.7688151,
      "learning_rate": 2.4320130524485506e-06,
      "loss": 0.79093283,
      "num_input_tokens_seen": 80166185,
      "step": 3724,
      "time_per_iteration": 2.7339937686920166
    },
    {
      "auxiliary_loss_clip": 0.01183678,
      "auxiliary_loss_mlp": 0.01031228,
      "balance_loss_clip": 1.05651927,
      "balance_loss_mlp": 1.02314603,
      "epoch": 0.4479047676306138,
      "flos": 21975462984960.0,
      "grad_norm": 1.8962047060940108,
      "language_loss": 0.7956993,
      "learning_rate": 2.431252438077746e-06,
      "loss": 0.81784838,
      "num_input_tokens_seen": 80185685,
      "step": 3725,
      "time_per_iteration": 2.6325860023498535
    },
    {
      "auxiliary_loss_clip": 0.01185089,
      "auxiliary_loss_mlp": 0.01022245,
      "balance_loss_clip": 1.05576789,
      "balance_loss_mlp": 1.01362002,
      "epoch": 0.44802501052125293,
      "flos": 21467219495040.0,
      "grad_norm": 2.9256410117756158,
      "language_loss": 0.76877248,
      "learning_rate": 2.4304917582862906e-06,
      "loss": 0.79084581,
      "num_input_tokens_seen": 80204865,
      "step": 3726,
      "time_per_iteration": 3.5873658657073975
    },
    {
      "auxiliary_loss_clip": 0.01184377,
      "auxiliary_loss_mlp": 0.01026429,
      "balance_loss_clip": 1.05691314,
      "balance_loss_mlp": 1.01804292,
      "epoch": 0.44814525341189204,
      "flos": 22126970551680.0,
      "grad_norm": 2.046801906825566,
      "language_loss": 0.87789291,
      "learning_rate": 2.4297310131895774e-06,
      "loss": 0.90000093,
      "num_input_tokens_seen": 80223410,
      "step": 3727,
      "time_per_iteration": 2.654317855834961
    },
    {
      "auxiliary_loss_clip": 0.01186648,
      "auxiliary_loss_mlp": 0.01030653,
      "balance_loss_clip": 1.05622554,
      "balance_loss_mlp": 1.02214181,
      "epoch": 0.4482654963025311,
      "flos": 16653933204480.0,
      "grad_norm": 2.292173457997496,
      "language_loss": 0.74235833,
      "learning_rate": 2.4289702029030113e-06,
      "loss": 0.76453137,
      "num_input_tokens_seen": 80240880,
      "step": 3728,
      "time_per_iteration": 2.603353261947632
    },
    {
      "auxiliary_loss_clip": 0.01189496,
      "auxiliary_loss_mlp": 0.01031708,
      "balance_loss_clip": 1.05875683,
      "balance_loss_mlp": 1.0227493,
      "epoch": 0.4483857391931702,
      "flos": 18841251905280.0,
      "grad_norm": 1.9071861699033077,
      "language_loss": 0.83059454,
      "learning_rate": 2.4282093275420057e-06,
      "loss": 0.85280657,
      "num_input_tokens_seen": 80259910,
      "step": 3729,
      "time_per_iteration": 3.5142786502838135
    },
    {
      "auxiliary_loss_clip": 0.01186955,
      "auxiliary_loss_mlp": 0.01030564,
      "balance_loss_clip": 1.05745649,
      "balance_loss_mlp": 1.02242208,
      "epoch": 0.4485059820838093,
      "flos": 20372590477440.0,
      "grad_norm": 2.7470919904164233,
      "language_loss": 0.70741713,
      "learning_rate": 2.4274483872219863e-06,
      "loss": 0.72959232,
      "num_input_tokens_seen": 80277270,
      "step": 3730,
      "time_per_iteration": 3.575108528137207
    },
    {
      "auxiliary_loss_clip": 0.01183579,
      "auxiliary_loss_mlp": 0.01027139,
      "balance_loss_clip": 1.05626047,
      "balance_loss_mlp": 1.01891935,
      "epoch": 0.4486262249744484,
      "flos": 20047742853120.0,
      "grad_norm": 2.235196286663565,
      "language_loss": 0.93461633,
      "learning_rate": 2.426687382058386e-06,
      "loss": 0.95672351,
      "num_input_tokens_seen": 80295550,
      "step": 3731,
      "time_per_iteration": 3.4959540367126465
    },
    {
      "auxiliary_loss_clip": 0.01086706,
      "auxiliary_loss_mlp": 0.01001857,
      "balance_loss_clip": 1.02276301,
      "balance_loss_mlp": 1.00058174,
      "epoch": 0.4487464678650875,
      "flos": 64595684776320.0,
      "grad_norm": 0.8640294163240405,
      "language_loss": 0.59786785,
      "learning_rate": 2.425926312166649e-06,
      "loss": 0.61875355,
      "num_input_tokens_seen": 80348425,
      "step": 3732,
      "time_per_iteration": 3.0437378883361816
    },
    {
      "auxiliary_loss_clip": 0.01188217,
      "auxiliary_loss_mlp": 0.01033556,
      "balance_loss_clip": 1.05636835,
      "balance_loss_mlp": 1.02438903,
      "epoch": 0.4488667107557266,
      "flos": 20769798049920.0,
      "grad_norm": 5.617543385046393,
      "language_loss": 0.73163784,
      "learning_rate": 2.42516517766223e-06,
      "loss": 0.75385553,
      "num_input_tokens_seen": 80366505,
      "step": 3733,
      "time_per_iteration": 2.6830689907073975
    },
    {
      "auxiliary_loss_clip": 0.01187772,
      "auxiliary_loss_mlp": 0.01032554,
      "balance_loss_clip": 1.05954814,
      "balance_loss_mlp": 1.023947,
      "epoch": 0.44898695364636565,
      "flos": 23951735326080.0,
      "grad_norm": 1.9098481135684553,
      "language_loss": 0.68047547,
      "learning_rate": 2.4244039786605907e-06,
      "loss": 0.70267868,
      "num_input_tokens_seen": 80387510,
      "step": 3734,
      "time_per_iteration": 2.7096846103668213
    },
    {
      "auxiliary_loss_clip": 0.01183034,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.05346417,
      "balance_loss_mlp": 1.02074742,
      "epoch": 0.44910719653700476,
      "flos": 18624351628800.0,
      "grad_norm": 2.4483564305228054,
      "language_loss": 0.82521641,
      "learning_rate": 2.4236427152772055e-06,
      "loss": 0.8473385,
      "num_input_tokens_seen": 80405915,
      "step": 3735,
      "time_per_iteration": 2.6373608112335205
    },
    {
      "auxiliary_loss_clip": 0.01084667,
      "auxiliary_loss_mlp": 0.01002283,
      "balance_loss_clip": 1.02077806,
      "balance_loss_mlp": 1.00101948,
      "epoch": 0.4492274394276438,
      "flos": 57033435749760.0,
      "grad_norm": 0.8522842979281546,
      "language_loss": 0.57377613,
      "learning_rate": 2.422881387627557e-06,
      "loss": 0.59464562,
      "num_input_tokens_seen": 80458365,
      "step": 3736,
      "time_per_iteration": 2.9478931427001953
    },
    {
      "auxiliary_loss_clip": 0.01185171,
      "auxiliary_loss_mlp": 0.01029307,
      "balance_loss_clip": 1.05681336,
      "balance_loss_mlp": 1.02029502,
      "epoch": 0.4493476823182829,
      "flos": 23254888498560.0,
      "grad_norm": 1.5709463820215182,
      "language_loss": 0.77395523,
      "learning_rate": 2.422119995827139e-06,
      "loss": 0.79610002,
      "num_input_tokens_seen": 80478490,
      "step": 3737,
      "time_per_iteration": 2.6290688514709473
    },
    {
      "auxiliary_loss_clip": 0.01185279,
      "auxiliary_loss_mlp": 0.01028912,
      "balance_loss_clip": 1.05776167,
      "balance_loss_mlp": 1.01995349,
      "epoch": 0.44946792520892204,
      "flos": 15815131827840.0,
      "grad_norm": 2.854072221603449,
      "language_loss": 0.74420512,
      "learning_rate": 2.4213585399914528e-06,
      "loss": 0.76634705,
      "num_input_tokens_seen": 80495695,
      "step": 3738,
      "time_per_iteration": 2.6101725101470947
    },
    {
      "auxiliary_loss_clip": 0.01185213,
      "auxiliary_loss_mlp": 0.01023575,
      "balance_loss_clip": 1.05654228,
      "balance_loss_mlp": 1.01608837,
      "epoch": 0.4495881680995611,
      "flos": 19610063631360.0,
      "grad_norm": 2.5481261353039355,
      "language_loss": 0.85263354,
      "learning_rate": 2.4205970202360113e-06,
      "loss": 0.87472141,
      "num_input_tokens_seen": 80515260,
      "step": 3739,
      "time_per_iteration": 2.642836809158325
    },
    {
      "auxiliary_loss_clip": 0.0118563,
      "auxiliary_loss_mlp": 0.01029165,
      "balance_loss_clip": 1.05632734,
      "balance_loss_mlp": 1.02050495,
      "epoch": 0.4497084109902002,
      "flos": 26031465815040.0,
      "grad_norm": 2.491410833413112,
      "language_loss": 0.78233343,
      "learning_rate": 2.4198354366763354e-06,
      "loss": 0.80448139,
      "num_input_tokens_seen": 80533900,
      "step": 3740,
      "time_per_iteration": 2.6345691680908203
    },
    {
      "auxiliary_loss_clip": 0.01186014,
      "auxiliary_loss_mlp": 0.01030274,
      "balance_loss_clip": 1.05691791,
      "balance_loss_mlp": 1.02135086,
      "epoch": 0.4498286538808393,
      "flos": 14793688771200.0,
      "grad_norm": 2.035723370851809,
      "language_loss": 0.78461719,
      "learning_rate": 2.4190737894279587e-06,
      "loss": 0.80677998,
      "num_input_tokens_seen": 80551270,
      "step": 3741,
      "time_per_iteration": 2.6369311809539795
    },
    {
      "auxiliary_loss_clip": 0.01183217,
      "auxiliary_loss_mlp": 0.01027908,
      "balance_loss_clip": 1.05624926,
      "balance_loss_mlp": 1.01963484,
      "epoch": 0.44994889677147837,
      "flos": 15450171690240.0,
      "grad_norm": 2.304537952815193,
      "language_loss": 0.8013376,
      "learning_rate": 2.4183120786064203e-06,
      "loss": 0.82344878,
      "num_input_tokens_seen": 80568145,
      "step": 3742,
      "time_per_iteration": 2.5906338691711426
    },
    {
      "auxiliary_loss_clip": 0.01185013,
      "auxiliary_loss_mlp": 0.01035667,
      "balance_loss_clip": 1.05564749,
      "balance_loss_mlp": 1.02684546,
      "epoch": 0.4500691396621175,
      "flos": 21798316085760.0,
      "grad_norm": 2.3218419953275236,
      "language_loss": 0.85493493,
      "learning_rate": 2.417550304327273e-06,
      "loss": 0.87714171,
      "num_input_tokens_seen": 80586185,
      "step": 3743,
      "time_per_iteration": 2.6787188053131104
    },
    {
      "auxiliary_loss_clip": 0.01190018,
      "auxiliary_loss_mlp": 0.01030841,
      "balance_loss_clip": 1.05675209,
      "balance_loss_mlp": 1.02072573,
      "epoch": 0.4501893825527566,
      "flos": 32382016421760.0,
      "grad_norm": 1.9095119686062363,
      "language_loss": 0.75629145,
      "learning_rate": 2.4167884667060763e-06,
      "loss": 0.77850008,
      "num_input_tokens_seen": 80608895,
      "step": 3744,
      "time_per_iteration": 2.718067169189453
    },
    {
      "auxiliary_loss_clip": 0.0119183,
      "auxiliary_loss_mlp": 0.01031668,
      "balance_loss_clip": 1.05913544,
      "balance_loss_mlp": 1.02223253,
      "epoch": 0.45030962544339564,
      "flos": 16544944362240.0,
      "grad_norm": 2.3134133368974776,
      "language_loss": 0.87108767,
      "learning_rate": 2.4160265658584e-06,
      "loss": 0.89332259,
      "num_input_tokens_seen": 80623785,
      "step": 3745,
      "time_per_iteration": 2.646925926208496
    },
    {
      "auxiliary_loss_clip": 0.01188213,
      "auxiliary_loss_mlp": 0.01028085,
      "balance_loss_clip": 1.05950856,
      "balance_loss_mlp": 1.01959157,
      "epoch": 0.45042986833403476,
      "flos": 19573039687680.0,
      "grad_norm": 2.0658149634642395,
      "language_loss": 0.68519127,
      "learning_rate": 2.4152646018998253e-06,
      "loss": 0.70735425,
      "num_input_tokens_seen": 80642735,
      "step": 3746,
      "time_per_iteration": 2.633648157119751
    },
    {
      "auxiliary_loss_clip": 0.01187053,
      "auxiliary_loss_mlp": 0.01029365,
      "balance_loss_clip": 1.0576148,
      "balance_loss_mlp": 1.02042449,
      "epoch": 0.45055011122467387,
      "flos": 23112467072640.0,
      "grad_norm": 1.8055834538954216,
      "language_loss": 0.71554792,
      "learning_rate": 2.4145025749459403e-06,
      "loss": 0.73771214,
      "num_input_tokens_seen": 80663760,
      "step": 3747,
      "time_per_iteration": 2.6933417320251465
    },
    {
      "auxiliary_loss_clip": 0.01188411,
      "auxiliary_loss_mlp": 0.01029629,
      "balance_loss_clip": 1.05752957,
      "balance_loss_mlp": 1.02067077,
      "epoch": 0.4506703541153129,
      "flos": 19934623946880.0,
      "grad_norm": 2.0611536757565663,
      "language_loss": 0.70332175,
      "learning_rate": 2.413740485112344e-06,
      "loss": 0.72550219,
      "num_input_tokens_seen": 80682100,
      "step": 3748,
      "time_per_iteration": 2.7108170986175537
    },
    {
      "auxiliary_loss_clip": 0.01184759,
      "auxiliary_loss_mlp": 0.01025864,
      "balance_loss_clip": 1.05683756,
      "balance_loss_mlp": 1.01807427,
      "epoch": 0.45079059700595203,
      "flos": 19499530504320.0,
      "grad_norm": 5.092081906429002,
      "language_loss": 0.82313263,
      "learning_rate": 2.412978332514646e-06,
      "loss": 0.84523892,
      "num_input_tokens_seen": 80700880,
      "step": 3749,
      "time_per_iteration": 2.6712818145751953
    },
    {
      "auxiliary_loss_clip": 0.01189295,
      "auxiliary_loss_mlp": 0.01031961,
      "balance_loss_clip": 1.05843389,
      "balance_loss_mlp": 1.02254915,
      "epoch": 0.4509108398965911,
      "flos": 27636313570560.0,
      "grad_norm": 2.756570435767487,
      "language_loss": 0.72579312,
      "learning_rate": 2.4122161172684623e-06,
      "loss": 0.74800563,
      "num_input_tokens_seen": 80721675,
      "step": 3750,
      "time_per_iteration": 2.6885745525360107
    },
    {
      "auxiliary_loss_clip": 0.01187841,
      "auxiliary_loss_mlp": 0.01026627,
      "balance_loss_clip": 1.05655456,
      "balance_loss_mlp": 1.01771641,
      "epoch": 0.4510310827872302,
      "flos": 20995712640000.0,
      "grad_norm": 2.3388796990882605,
      "language_loss": 0.84094423,
      "learning_rate": 2.4114538394894216e-06,
      "loss": 0.86308897,
      "num_input_tokens_seen": 80739315,
      "step": 3751,
      "time_per_iteration": 2.6364426612854004
    },
    {
      "auxiliary_loss_clip": 0.01183833,
      "auxiliary_loss_mlp": 0.01031755,
      "balance_loss_clip": 1.0556109,
      "balance_loss_mlp": 1.02344656,
      "epoch": 0.4511513256778693,
      "flos": 16216684945920.0,
      "grad_norm": 19.25269882809617,
      "language_loss": 0.83056653,
      "learning_rate": 2.410691499293161e-06,
      "loss": 0.85272241,
      "num_input_tokens_seen": 80757470,
      "step": 3752,
      "time_per_iteration": 3.5348007678985596
    },
    {
      "auxiliary_loss_clip": 0.01183576,
      "auxiliary_loss_mlp": 0.01028906,
      "balance_loss_clip": 1.0551722,
      "balance_loss_mlp": 1.02043581,
      "epoch": 0.45127156856850836,
      "flos": 25186702780800.0,
      "grad_norm": 2.646160951940717,
      "language_loss": 0.73733819,
      "learning_rate": 2.409929096795326e-06,
      "loss": 0.75946295,
      "num_input_tokens_seen": 80777840,
      "step": 3753,
      "time_per_iteration": 2.6150457859039307
    },
    {
      "auxiliary_loss_clip": 0.0118674,
      "auxiliary_loss_mlp": 0.01029973,
      "balance_loss_clip": 1.05603945,
      "balance_loss_mlp": 1.0206207,
      "epoch": 0.4513918114591475,
      "flos": 20412523422720.0,
      "grad_norm": 2.39688344146793,
      "language_loss": 0.79249924,
      "learning_rate": 2.409166632111573e-06,
      "loss": 0.81466639,
      "num_input_tokens_seen": 80795975,
      "step": 3754,
      "time_per_iteration": 2.6635653972625732
    },
    {
      "auxiliary_loss_clip": 0.01192463,
      "auxiliary_loss_mlp": 0.01031584,
      "balance_loss_clip": 1.05917478,
      "balance_loss_mlp": 1.02173126,
      "epoch": 0.4515120543497866,
      "flos": 26648482665600.0,
      "grad_norm": 3.198198270811842,
      "language_loss": 0.80484402,
      "learning_rate": 2.4084041053575674e-06,
      "loss": 0.82708454,
      "num_input_tokens_seen": 80815395,
      "step": 3755,
      "time_per_iteration": 3.6023058891296387
    },
    {
      "auxiliary_loss_clip": 0.01188669,
      "auxiliary_loss_mlp": 0.01031551,
      "balance_loss_clip": 1.05786109,
      "balance_loss_mlp": 1.0226047,
      "epoch": 0.45163229724042564,
      "flos": 20595093275520.0,
      "grad_norm": 2.0818363459659923,
      "language_loss": 0.72462916,
      "learning_rate": 2.4076415166489834e-06,
      "loss": 0.74683142,
      "num_input_tokens_seen": 80834805,
      "step": 3756,
      "time_per_iteration": 2.6166553497314453
    },
    {
      "auxiliary_loss_clip": 0.01184501,
      "auxiliary_loss_mlp": 0.01032773,
      "balance_loss_clip": 1.05489516,
      "balance_loss_mlp": 1.02449942,
      "epoch": 0.45175254013106475,
      "flos": 21689004021120.0,
      "grad_norm": 4.474186291196652,
      "language_loss": 0.79075265,
      "learning_rate": 2.406878866101506e-06,
      "loss": 0.81292534,
      "num_input_tokens_seen": 80853770,
      "step": 3757,
      "time_per_iteration": 2.619966506958008
    },
    {
      "auxiliary_loss_clip": 0.0118715,
      "auxiliary_loss_mlp": 0.01032112,
      "balance_loss_clip": 1.05764747,
      "balance_loss_mlp": 1.02299881,
      "epoch": 0.45187278302170386,
      "flos": 18878850466560.0,
      "grad_norm": 2.6423978829055472,
      "language_loss": 0.78516394,
      "learning_rate": 2.4061161538308273e-06,
      "loss": 0.8073566,
      "num_input_tokens_seen": 80870615,
      "step": 3758,
      "time_per_iteration": 4.193307161331177
    },
    {
      "auxiliary_loss_clip": 0.01186274,
      "auxiliary_loss_mlp": 0.01022508,
      "balance_loss_clip": 1.05662835,
      "balance_loss_mlp": 1.01425862,
      "epoch": 0.4519930259123429,
      "flos": 18582479349120.0,
      "grad_norm": 2.1004770015007015,
      "language_loss": 0.88918734,
      "learning_rate": 2.4053533799526523e-06,
      "loss": 0.91127515,
      "num_input_tokens_seen": 80886335,
      "step": 3759,
      "time_per_iteration": 2.427422523498535
    },
    {
      "auxiliary_loss_clip": 0.01186894,
      "auxiliary_loss_mlp": 0.0102647,
      "balance_loss_clip": 1.05827415,
      "balance_loss_mlp": 1.01803565,
      "epoch": 0.452113268802982,
      "flos": 25192377129600.0,
      "grad_norm": 1.6758791024911333,
      "language_loss": 0.85996622,
      "learning_rate": 2.404590544582691e-06,
      "loss": 0.88209987,
      "num_input_tokens_seen": 80904570,
      "step": 3760,
      "time_per_iteration": 2.5620789527893066
    },
    {
      "auxiliary_loss_clip": 0.01184738,
      "auxiliary_loss_mlp": 0.01035249,
      "balance_loss_clip": 1.05639219,
      "balance_loss_mlp": 1.02594447,
      "epoch": 0.45223351169362114,
      "flos": 39378922312320.0,
      "grad_norm": 2.187964576319779,
      "language_loss": 0.81002069,
      "learning_rate": 2.403827647836666e-06,
      "loss": 0.83222055,
      "num_input_tokens_seen": 80925125,
      "step": 3761,
      "time_per_iteration": 2.729060649871826
    },
    {
      "auxiliary_loss_clip": 0.01185019,
      "auxiliary_loss_mlp": 0.01033277,
      "balance_loss_clip": 1.05513692,
      "balance_loss_mlp": 1.02427101,
      "epoch": 0.4523537545842602,
      "flos": 21582169994880.0,
      "grad_norm": 2.0236923154285944,
      "language_loss": 0.69317442,
      "learning_rate": 2.4030646898303075e-06,
      "loss": 0.71535742,
      "num_input_tokens_seen": 80946615,
      "step": 3762,
      "time_per_iteration": 2.618187427520752
    },
    {
      "auxiliary_loss_clip": 0.01185749,
      "auxiliary_loss_mlp": 0.0102324,
      "balance_loss_clip": 1.05568862,
      "balance_loss_mlp": 1.01437724,
      "epoch": 0.4524739974748993,
      "flos": 28439527547520.0,
      "grad_norm": 2.3740142223730585,
      "language_loss": 0.81901908,
      "learning_rate": 2.4023016706793566e-06,
      "loss": 0.84110892,
      "num_input_tokens_seen": 80966410,
      "step": 3763,
      "time_per_iteration": 2.8004660606384277
    },
    {
      "auxiliary_loss_clip": 0.01083817,
      "auxiliary_loss_mlp": 0.01005271,
      "balance_loss_clip": 1.01993299,
      "balance_loss_mlp": 1.00401974,
      "epoch": 0.4525942403655384,
      "flos": 61556492148480.0,
      "grad_norm": 0.7839003337342993,
      "language_loss": 0.56821311,
      "learning_rate": 2.401538590499561e-06,
      "loss": 0.589104,
      "num_input_tokens_seen": 81026865,
      "step": 3764,
      "time_per_iteration": 3.2423362731933594
    },
    {
      "auxiliary_loss_clip": 0.01184449,
      "auxiliary_loss_mlp": 0.0103178,
      "balance_loss_clip": 1.05505252,
      "balance_loss_mlp": 1.02273178,
      "epoch": 0.45271448325617747,
      "flos": 27529838680320.0,
      "grad_norm": 2.1262626577702357,
      "language_loss": 0.71734977,
      "learning_rate": 2.400775449406682e-06,
      "loss": 0.73951203,
      "num_input_tokens_seen": 81050060,
      "step": 3765,
      "time_per_iteration": 2.651470184326172
    },
    {
      "auxiliary_loss_clip": 0.0118164,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 1.05392635,
      "balance_loss_mlp": 1.01889324,
      "epoch": 0.4528347261468166,
      "flos": 22452608275200.0,
      "grad_norm": 1.8461591121942547,
      "language_loss": 0.72969091,
      "learning_rate": 2.400012247516485e-06,
      "loss": 0.75177968,
      "num_input_tokens_seen": 81070625,
      "step": 3766,
      "time_per_iteration": 2.602030038833618
    },
    {
      "auxiliary_loss_clip": 0.01185971,
      "auxiliary_loss_mlp": 0.0103002,
      "balance_loss_clip": 1.05682206,
      "balance_loss_mlp": 1.02125776,
      "epoch": 0.45295496903745563,
      "flos": 21103875469440.0,
      "grad_norm": 5.879318536763092,
      "language_loss": 0.90179133,
      "learning_rate": 2.3992489849447484e-06,
      "loss": 0.92395127,
      "num_input_tokens_seen": 81089080,
      "step": 3767,
      "time_per_iteration": 2.5443403720855713
    },
    {
      "auxiliary_loss_clip": 0.01187797,
      "auxiliary_loss_mlp": 0.01028638,
      "balance_loss_clip": 1.05721378,
      "balance_loss_mlp": 1.01993561,
      "epoch": 0.45307521192809475,
      "flos": 23221168606080.0,
      "grad_norm": 1.9537497088296922,
      "language_loss": 0.78787386,
      "learning_rate": 2.3984856618072584e-06,
      "loss": 0.81003821,
      "num_input_tokens_seen": 81109115,
      "step": 3768,
      "time_per_iteration": 2.6029293537139893
    },
    {
      "auxiliary_loss_clip": 0.01187355,
      "auxiliary_loss_mlp": 0.01026375,
      "balance_loss_clip": 1.05894935,
      "balance_loss_mlp": 1.01839447,
      "epoch": 0.45319545481873386,
      "flos": 15560094286080.0,
      "grad_norm": 2.7035977754921556,
      "language_loss": 0.74135435,
      "learning_rate": 2.3977222782198098e-06,
      "loss": 0.76349157,
      "num_input_tokens_seen": 81127750,
      "step": 3769,
      "time_per_iteration": 2.5885207653045654
    },
    {
      "auxiliary_loss_clip": 0.01188126,
      "auxiliary_loss_mlp": 0.01029372,
      "balance_loss_clip": 1.05783772,
      "balance_loss_mlp": 1.02016354,
      "epoch": 0.4533156977093729,
      "flos": 21944759834880.0,
      "grad_norm": 2.700266986551084,
      "language_loss": 0.75325006,
      "learning_rate": 2.3969588342982077e-06,
      "loss": 0.77542508,
      "num_input_tokens_seen": 81147125,
      "step": 3770,
      "time_per_iteration": 2.6296584606170654
    },
    {
      "auxiliary_loss_clip": 0.01184388,
      "auxiliary_loss_mlp": 0.01029042,
      "balance_loss_clip": 1.05651808,
      "balance_loss_mlp": 1.01997018,
      "epoch": 0.453435940600012,
      "flos": 24242180699520.0,
      "grad_norm": 1.6996060507265072,
      "language_loss": 0.72571802,
      "learning_rate": 2.396195330158267e-06,
      "loss": 0.74785239,
      "num_input_tokens_seen": 81167015,
      "step": 3771,
      "time_per_iteration": 2.6920924186706543
    },
    {
      "auxiliary_loss_clip": 0.01186303,
      "auxiliary_loss_mlp": 0.01028034,
      "balance_loss_clip": 1.05633545,
      "balance_loss_mlp": 1.01858616,
      "epoch": 0.45355618349065113,
      "flos": 23440367352960.0,
      "grad_norm": 1.990740745291259,
      "language_loss": 0.796341,
      "learning_rate": 2.3954317659158094e-06,
      "loss": 0.81848431,
      "num_input_tokens_seen": 81187350,
      "step": 3772,
      "time_per_iteration": 2.636967658996582
    },
    {
      "auxiliary_loss_clip": 0.01083266,
      "auxiliary_loss_mlp": 0.01003323,
      "balance_loss_clip": 1.01923537,
      "balance_loss_mlp": 1.00210714,
      "epoch": 0.4536764263812902,
      "flos": 66903161448960.0,
      "grad_norm": 0.9044821550839641,
      "language_loss": 0.56884915,
      "learning_rate": 2.394668141686667e-06,
      "loss": 0.589715,
      "num_input_tokens_seen": 81249315,
      "step": 3773,
      "time_per_iteration": 3.2159852981567383
    },
    {
      "auxiliary_loss_clip": 0.01185211,
      "auxiliary_loss_mlp": 0.01028392,
      "balance_loss_clip": 1.05711925,
      "balance_loss_mlp": 1.0202446,
      "epoch": 0.4537966692719293,
      "flos": 42739766254080.0,
      "grad_norm": 4.048722454186319,
      "language_loss": 0.69594032,
      "learning_rate": 2.3939044575866813e-06,
      "loss": 0.71807635,
      "num_input_tokens_seen": 81272065,
      "step": 3774,
      "time_per_iteration": 2.7532386779785156
    },
    {
      "auxiliary_loss_clip": 0.01185855,
      "auxiliary_loss_mlp": 0.01024178,
      "balance_loss_clip": 1.0569365,
      "balance_loss_mlp": 1.01570201,
      "epoch": 0.4539169121625684,
      "flos": 35549480517120.0,
      "grad_norm": 5.7208093337074075,
      "language_loss": 0.75193834,
      "learning_rate": 2.3931407137317024e-06,
      "loss": 0.77403867,
      "num_input_tokens_seen": 81292220,
      "step": 3775,
      "time_per_iteration": 2.752424716949463
    },
    {
      "auxiliary_loss_clip": 0.01185138,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.05630612,
      "balance_loss_mlp": 1.02176809,
      "epoch": 0.45403715505320746,
      "flos": 18514716341760.0,
      "grad_norm": 1.7526250522665372,
      "language_loss": 0.84866965,
      "learning_rate": 2.3923769102375907e-06,
      "loss": 0.87082016,
      "num_input_tokens_seen": 81311085,
      "step": 3776,
      "time_per_iteration": 2.607253313064575
    },
    {
      "auxiliary_loss_clip": 0.01189343,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.05798614,
      "balance_loss_mlp": 1.02010024,
      "epoch": 0.4541573979438466,
      "flos": 25045825639680.0,
      "grad_norm": 2.3417753547939424,
      "language_loss": 0.78573775,
      "learning_rate": 2.391613047220213e-06,
      "loss": 0.80791962,
      "num_input_tokens_seen": 81330985,
      "step": 3777,
      "time_per_iteration": 2.7021498680114746
    },
    {
      "auxiliary_loss_clip": 0.01184125,
      "auxiliary_loss_mlp": 0.01025697,
      "balance_loss_clip": 1.0554949,
      "balance_loss_mlp": 1.01684594,
      "epoch": 0.4542776408344857,
      "flos": 18332397884160.0,
      "grad_norm": 1.9896749702604406,
      "language_loss": 0.78647059,
      "learning_rate": 2.390849124795447e-06,
      "loss": 0.80856884,
      "num_input_tokens_seen": 81346985,
      "step": 3778,
      "time_per_iteration": 2.5572571754455566
    },
    {
      "auxiliary_loss_clip": 0.01188619,
      "auxiliary_loss_mlp": 0.01024456,
      "balance_loss_clip": 1.05757284,
      "balance_loss_mlp": 1.016541,
      "epoch": 0.45439788372512474,
      "flos": 20701173116160.0,
      "grad_norm": 2.2377131131126937,
      "language_loss": 0.84525096,
      "learning_rate": 2.3900851430791804e-06,
      "loss": 0.86738175,
      "num_input_tokens_seen": 81365005,
      "step": 3779,
      "time_per_iteration": 3.570164203643799
    },
    {
      "auxiliary_loss_clip": 0.01188228,
      "auxiliary_loss_mlp": 0.0103306,
      "balance_loss_clip": 1.05617523,
      "balance_loss_mlp": 1.02314186,
      "epoch": 0.45451812661576385,
      "flos": 22309432663680.0,
      "grad_norm": 2.047033221385213,
      "language_loss": 0.84604752,
      "learning_rate": 2.389321102187307e-06,
      "loss": 0.86826044,
      "num_input_tokens_seen": 81383785,
      "step": 3780,
      "time_per_iteration": 2.6390411853790283
    },
    {
      "auxiliary_loss_clip": 0.01184419,
      "auxiliary_loss_mlp": 0.01033219,
      "balance_loss_clip": 1.05527437,
      "balance_loss_mlp": 1.02448702,
      "epoch": 0.4546383695064029,
      "flos": 21763303303680.0,
      "grad_norm": 1.894540965029981,
      "language_loss": 0.81567711,
      "learning_rate": 2.3885570022357326e-06,
      "loss": 0.83785343,
      "num_input_tokens_seen": 81402915,
      "step": 3781,
      "time_per_iteration": 3.514907121658325
    },
    {
      "auxiliary_loss_clip": 0.01083,
      "auxiliary_loss_mlp": 0.01000692,
      "balance_loss_clip": 1.0188055,
      "balance_loss_mlp": 0.99944019,
      "epoch": 0.454758612397042,
      "flos": 64242755694720.0,
      "grad_norm": 0.7990977300440862,
      "language_loss": 0.60885704,
      "learning_rate": 2.38779284334037e-06,
      "loss": 0.62969387,
      "num_input_tokens_seen": 81467890,
      "step": 3782,
      "time_per_iteration": 3.2125084400177
    },
    {
      "auxiliary_loss_clip": 0.01187211,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 1.05756307,
      "balance_loss_mlp": 1.02010155,
      "epoch": 0.4548788552876811,
      "flos": 27304175485440.0,
      "grad_norm": 2.0443206711831254,
      "language_loss": 0.7886945,
      "learning_rate": 2.387028625617141e-06,
      "loss": 0.81084973,
      "num_input_tokens_seen": 81487105,
      "step": 3783,
      "time_per_iteration": 3.6256258487701416
    },
    {
      "auxiliary_loss_clip": 0.01182398,
      "auxiliary_loss_mlp": 0.01020901,
      "balance_loss_clip": 1.05381334,
      "balance_loss_mlp": 1.01251507,
      "epoch": 0.4549990981783202,
      "flos": 22857142222080.0,
      "grad_norm": 2.201880334104194,
      "language_loss": 0.84628272,
      "learning_rate": 2.3862643491819766e-06,
      "loss": 0.8683157,
      "num_input_tokens_seen": 81505670,
      "step": 3784,
      "time_per_iteration": 3.5211455821990967
    },
    {
      "auxiliary_loss_clip": 0.01182286,
      "auxiliary_loss_mlp": 0.01031235,
      "balance_loss_clip": 1.05546689,
      "balance_loss_mlp": 1.02291989,
      "epoch": 0.4551193410689593,
      "flos": 23258587599360.0,
      "grad_norm": 2.095416681796531,
      "language_loss": 0.8456614,
      "learning_rate": 2.3855000141508186e-06,
      "loss": 0.86779654,
      "num_input_tokens_seen": 81525825,
      "step": 3785,
      "time_per_iteration": 2.658689498901367
    },
    {
      "auxiliary_loss_clip": 0.01185524,
      "auxiliary_loss_mlp": 0.01032418,
      "balance_loss_clip": 1.05582762,
      "balance_loss_mlp": 1.02393007,
      "epoch": 0.4552395839595984,
      "flos": 20777519473920.0,
      "grad_norm": 2.6816228241723126,
      "language_loss": 0.83794618,
      "learning_rate": 2.3847356206396143e-06,
      "loss": 0.86012554,
      "num_input_tokens_seen": 81543135,
      "step": 3786,
      "time_per_iteration": 2.6127047538757324
    },
    {
      "auxiliary_loss_clip": 0.01185143,
      "auxiliary_loss_mlp": 0.01029834,
      "balance_loss_clip": 1.05658591,
      "balance_loss_mlp": 1.02137637,
      "epoch": 0.45535982685023746,
      "flos": 23257510191360.0,
      "grad_norm": 1.5086305518819687,
      "language_loss": 0.78613615,
      "learning_rate": 2.3839711687643227e-06,
      "loss": 0.80828595,
      "num_input_tokens_seen": 81564360,
      "step": 3787,
      "time_per_iteration": 2.62813138961792
    },
    {
      "auxiliary_loss_clip": 0.01188613,
      "auxiliary_loss_mlp": 0.01027621,
      "balance_loss_clip": 1.05658221,
      "balance_loss_mlp": 1.0181613,
      "epoch": 0.45548006974087657,
      "flos": 19646117907840.0,
      "grad_norm": 2.7957529709448137,
      "language_loss": 0.74026132,
      "learning_rate": 2.38320665864091e-06,
      "loss": 0.76242363,
      "num_input_tokens_seen": 81583710,
      "step": 3788,
      "time_per_iteration": 2.681004285812378
    },
    {
      "auxiliary_loss_clip": 0.01182762,
      "auxiliary_loss_mlp": 0.01024097,
      "balance_loss_clip": 1.0518744,
      "balance_loss_mlp": 1.01513815,
      "epoch": 0.4556003126315157,
      "flos": 20047778766720.0,
      "grad_norm": 2.2924235546314318,
      "language_loss": 0.81716442,
      "learning_rate": 2.3824420903853516e-06,
      "loss": 0.83923304,
      "num_input_tokens_seen": 81602175,
      "step": 3789,
      "time_per_iteration": 2.6909658908843994
    },
    {
      "auxiliary_loss_clip": 0.01186551,
      "auxiliary_loss_mlp": 0.0103021,
      "balance_loss_clip": 1.05856156,
      "balance_loss_mlp": 1.02113187,
      "epoch": 0.45572055552215474,
      "flos": 22959738443520.0,
      "grad_norm": 2.6294174580817353,
      "language_loss": 0.81923288,
      "learning_rate": 2.3816774641136324e-06,
      "loss": 0.8414005,
      "num_input_tokens_seen": 81619430,
      "step": 3790,
      "time_per_iteration": 2.6788723468780518
    },
    {
      "auxiliary_loss_clip": 0.01183905,
      "auxiliary_loss_mlp": 0.01027239,
      "balance_loss_clip": 1.0561018,
      "balance_loss_mlp": 1.01897788,
      "epoch": 0.45584079841279385,
      "flos": 33109925535360.0,
      "grad_norm": 1.9291084704954067,
      "language_loss": 0.71562463,
      "learning_rate": 2.380912779941745e-06,
      "loss": 0.73773611,
      "num_input_tokens_seen": 81642550,
      "step": 3791,
      "time_per_iteration": 2.7375528812408447
    },
    {
      "auxiliary_loss_clip": 0.01185837,
      "auxiliary_loss_mlp": 0.01022422,
      "balance_loss_clip": 1.05327511,
      "balance_loss_mlp": 1.01385689,
      "epoch": 0.45596104130343296,
      "flos": 27272179445760.0,
      "grad_norm": 2.6705334238852165,
      "language_loss": 0.83430362,
      "learning_rate": 2.3801480379856918e-06,
      "loss": 0.85638618,
      "num_input_tokens_seen": 81664260,
      "step": 3792,
      "time_per_iteration": 2.699954032897949
    },
    {
      "auxiliary_loss_clip": 0.01186425,
      "auxiliary_loss_mlp": 0.01028315,
      "balance_loss_clip": 1.05701232,
      "balance_loss_mlp": 1.02022314,
      "epoch": 0.456081284194072,
      "flos": 21579799697280.0,
      "grad_norm": 1.910957054609566,
      "language_loss": 0.83840203,
      "learning_rate": 2.379383238361484e-06,
      "loss": 0.86054933,
      "num_input_tokens_seen": 81683620,
      "step": 3793,
      "time_per_iteration": 2.6069884300231934
    },
    {
      "auxiliary_loss_clip": 0.01182935,
      "auxiliary_loss_mlp": 0.01026676,
      "balance_loss_clip": 1.05435741,
      "balance_loss_mlp": 1.01856387,
      "epoch": 0.4562015270847111,
      "flos": 35918822113920.0,
      "grad_norm": 1.9980837625827685,
      "language_loss": 0.79816419,
      "learning_rate": 2.3786183811851407e-06,
      "loss": 0.82026029,
      "num_input_tokens_seen": 81704325,
      "step": 3794,
      "time_per_iteration": 2.7599477767944336
    },
    {
      "auxiliary_loss_clip": 0.01186468,
      "auxiliary_loss_mlp": 0.01033609,
      "balance_loss_clip": 1.05628896,
      "balance_loss_mlp": 1.024966,
      "epoch": 0.45632176997535023,
      "flos": 13589783602560.0,
      "grad_norm": 2.021870573692774,
      "language_loss": 0.80149579,
      "learning_rate": 2.3778534665726892e-06,
      "loss": 0.82369655,
      "num_input_tokens_seen": 81721155,
      "step": 3795,
      "time_per_iteration": 2.629343032836914
    },
    {
      "auxiliary_loss_clip": 0.01183631,
      "auxiliary_loss_mlp": 0.01025035,
      "balance_loss_clip": 1.05718529,
      "balance_loss_mlp": 1.01714349,
      "epoch": 0.4564420128659893,
      "flos": 32635401937920.0,
      "grad_norm": 3.657635335266539,
      "language_loss": 0.72782457,
      "learning_rate": 2.377088494640168e-06,
      "loss": 0.74991119,
      "num_input_tokens_seen": 81742905,
      "step": 3796,
      "time_per_iteration": 2.768181800842285
    },
    {
      "auxiliary_loss_clip": 0.01184779,
      "auxiliary_loss_mlp": 0.01029104,
      "balance_loss_clip": 1.05648792,
      "balance_loss_mlp": 1.02096236,
      "epoch": 0.4565622557566284,
      "flos": 20377690208640.0,
      "grad_norm": 2.358458375411147,
      "language_loss": 0.78244728,
      "learning_rate": 2.3763234655036216e-06,
      "loss": 0.80458611,
      "num_input_tokens_seen": 81762105,
      "step": 3797,
      "time_per_iteration": 2.6025710105895996
    },
    {
      "auxiliary_loss_clip": 0.0118315,
      "auxiliary_loss_mlp": 0.01028077,
      "balance_loss_clip": 1.05485725,
      "balance_loss_mlp": 1.01945853,
      "epoch": 0.45668249864726745,
      "flos": 25374372364800.0,
      "grad_norm": 4.13488080907592,
      "language_loss": 0.8676464,
      "learning_rate": 2.3755583792791046e-06,
      "loss": 0.88975871,
      "num_input_tokens_seen": 81781975,
      "step": 3798,
      "time_per_iteration": 2.7054545879364014
    },
    {
      "auxiliary_loss_clip": 0.01184187,
      "auxiliary_loss_mlp": 0.0102929,
      "balance_loss_clip": 1.05435061,
      "balance_loss_mlp": 1.02093399,
      "epoch": 0.45680274153790656,
      "flos": 15559806977280.0,
      "grad_norm": 2.040563985228427,
      "language_loss": 0.74846369,
      "learning_rate": 2.3747932360826803e-06,
      "loss": 0.77059847,
      "num_input_tokens_seen": 81798905,
      "step": 3799,
      "time_per_iteration": 2.5751657485961914
    },
    {
      "auxiliary_loss_clip": 0.01185587,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.05532598,
      "balance_loss_mlp": 1.01815927,
      "epoch": 0.4569229844285457,
      "flos": 19792884879360.0,
      "grad_norm": 2.008057332795775,
      "language_loss": 0.82296491,
      "learning_rate": 2.3740280360304205e-06,
      "loss": 0.84508526,
      "num_input_tokens_seen": 81816630,
      "step": 3800,
      "time_per_iteration": 2.660635471343994
    },
    {
      "auxiliary_loss_clip": 0.01185135,
      "auxiliary_loss_mlp": 0.01025071,
      "balance_loss_clip": 1.05610049,
      "balance_loss_mlp": 1.01643968,
      "epoch": 0.45704322731918473,
      "flos": 24093941270400.0,
      "grad_norm": 3.5793602368708295,
      "language_loss": 0.68027377,
      "learning_rate": 2.3732627792384038e-06,
      "loss": 0.70237577,
      "num_input_tokens_seen": 81837700,
      "step": 3801,
      "time_per_iteration": 2.6954586505889893
    },
    {
      "auxiliary_loss_clip": 0.01183545,
      "auxiliary_loss_mlp": 0.01022636,
      "balance_loss_clip": 1.05364108,
      "balance_loss_mlp": 1.01494694,
      "epoch": 0.45716347020982384,
      "flos": 31317803245440.0,
      "grad_norm": 3.849918817904848,
      "language_loss": 0.75627124,
      "learning_rate": 2.3724974658227207e-06,
      "loss": 0.77833301,
      "num_input_tokens_seen": 81858490,
      "step": 3802,
      "time_per_iteration": 2.7206530570983887
    },
    {
      "auxiliary_loss_clip": 0.01186316,
      "auxiliary_loss_mlp": 0.01035816,
      "balance_loss_clip": 1.05635905,
      "balance_loss_mlp": 1.02728701,
      "epoch": 0.45728371310046295,
      "flos": 26501392471680.0,
      "grad_norm": 2.4861245353016788,
      "language_loss": 0.70877016,
      "learning_rate": 2.3717320958994687e-06,
      "loss": 0.73099148,
      "num_input_tokens_seen": 81876050,
      "step": 3803,
      "time_per_iteration": 2.7195446491241455
    },
    {
      "auxiliary_loss_clip": 0.01183138,
      "auxiliary_loss_mlp": 0.01025834,
      "balance_loss_clip": 1.05467761,
      "balance_loss_mlp": 1.01743007,
      "epoch": 0.457403955991102,
      "flos": 17929408222080.0,
      "grad_norm": 2.5284053093869794,
      "language_loss": 0.70472062,
      "learning_rate": 2.3709666695847534e-06,
      "loss": 0.72681034,
      "num_input_tokens_seen": 81894230,
      "step": 3804,
      "time_per_iteration": 3.5367939472198486
    },
    {
      "auxiliary_loss_clip": 0.01182832,
      "auxiliary_loss_mlp": 0.0102861,
      "balance_loss_clip": 1.05594671,
      "balance_loss_mlp": 1.02062881,
      "epoch": 0.4575241988817411,
      "flos": 42230660837760.0,
      "grad_norm": 1.7236569798448955,
      "language_loss": 0.70312852,
      "learning_rate": 2.370201186994689e-06,
      "loss": 0.72524297,
      "num_input_tokens_seen": 81917915,
      "step": 3805,
      "time_per_iteration": 2.857787609100342
    },
    {
      "auxiliary_loss_clip": 0.01185505,
      "auxiliary_loss_mlp": 0.01028218,
      "balance_loss_clip": 1.0571841,
      "balance_loss_mlp": 1.01993334,
      "epoch": 0.45764444177238023,
      "flos": 30117309868800.0,
      "grad_norm": 1.9866174605273008,
      "language_loss": 0.69932365,
      "learning_rate": 2.369435648245399e-06,
      "loss": 0.72146094,
      "num_input_tokens_seen": 81938130,
      "step": 3806,
      "time_per_iteration": 2.6585750579833984
    },
    {
      "auxiliary_loss_clip": 0.01185465,
      "auxiliary_loss_mlp": 0.01029272,
      "balance_loss_clip": 1.05634594,
      "balance_loss_mlp": 1.02076054,
      "epoch": 0.4577646846630193,
      "flos": 24060293205120.0,
      "grad_norm": 1.9478967504338256,
      "language_loss": 0.8547678,
      "learning_rate": 2.368670053453015e-06,
      "loss": 0.87691516,
      "num_input_tokens_seen": 81959820,
      "step": 3807,
      "time_per_iteration": 2.667463541030884
    },
    {
      "auxiliary_loss_clip": 0.01189419,
      "auxiliary_loss_mlp": 0.01030701,
      "balance_loss_clip": 1.05962443,
      "balance_loss_mlp": 1.02238584,
      "epoch": 0.4578849275536584,
      "flos": 17418578952960.0,
      "grad_norm": 2.8087573125327836,
      "language_loss": 0.74634159,
      "learning_rate": 2.3679044027336757e-06,
      "loss": 0.76854277,
      "num_input_tokens_seen": 81975710,
      "step": 3808,
      "time_per_iteration": 3.5652499198913574
    },
    {
      "auxiliary_loss_clip": 0.01189054,
      "auxiliary_loss_mlp": 0.01025414,
      "balance_loss_clip": 1.05819428,
      "balance_loss_mlp": 1.01662278,
      "epoch": 0.4580051704442975,
      "flos": 13510169107200.0,
      "grad_norm": 2.617653102596918,
      "language_loss": 0.6911236,
      "learning_rate": 2.3671386962035326e-06,
      "loss": 0.71326828,
      "num_input_tokens_seen": 81993180,
      "step": 3809,
      "time_per_iteration": 2.552274703979492
    },
    {
      "auxiliary_loss_clip": 0.01186398,
      "auxiliary_loss_mlp": 0.01029961,
      "balance_loss_clip": 1.05595112,
      "balance_loss_mlp": 1.02094245,
      "epoch": 0.45812541333493656,
      "flos": 18037606965120.0,
      "grad_norm": 2.387606893272986,
      "language_loss": 0.6853537,
      "learning_rate": 2.3663729339787405e-06,
      "loss": 0.70751727,
      "num_input_tokens_seen": 82010115,
      "step": 3810,
      "time_per_iteration": 3.5390639305114746
    },
    {
      "auxiliary_loss_clip": 0.01186456,
      "auxiliary_loss_mlp": 0.01033101,
      "balance_loss_clip": 1.056458,
      "balance_loss_mlp": 1.0241189,
      "epoch": 0.45824565622557567,
      "flos": 20222196232320.0,
      "grad_norm": 3.3252068911345054,
      "language_loss": 0.73723817,
      "learning_rate": 2.365607116175466e-06,
      "loss": 0.75943375,
      "num_input_tokens_seen": 82025540,
      "step": 3811,
      "time_per_iteration": 3.5768861770629883
    },
    {
      "auxiliary_loss_clip": 0.01182797,
      "auxiliary_loss_mlp": 0.01026455,
      "balance_loss_clip": 1.05540371,
      "balance_loss_mlp": 1.01784801,
      "epoch": 0.4583658991162148,
      "flos": 19864885691520.0,
      "grad_norm": 7.054239062987705,
      "language_loss": 0.66891676,
      "learning_rate": 2.3648412429098825e-06,
      "loss": 0.69100928,
      "num_input_tokens_seen": 82043890,
      "step": 3812,
      "time_per_iteration": 2.639930486679077
    },
    {
      "auxiliary_loss_clip": 0.01189692,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.0571661,
      "balance_loss_mlp": 1.02108359,
      "epoch": 0.45848614200685384,
      "flos": 21029935322880.0,
      "grad_norm": 2.0195895495244702,
      "language_loss": 0.81616211,
      "learning_rate": 2.364075314298172e-06,
      "loss": 0.83836305,
      "num_input_tokens_seen": 82061345,
      "step": 3813,
      "time_per_iteration": 2.631974458694458
    },
    {
      "auxiliary_loss_clip": 0.01184233,
      "auxiliary_loss_mlp": 0.01030736,
      "balance_loss_clip": 1.05726957,
      "balance_loss_mlp": 1.02206302,
      "epoch": 0.45860638489749295,
      "flos": 21069293650560.0,
      "grad_norm": 1.861175869633583,
      "language_loss": 0.70178342,
      "learning_rate": 2.3633093304565267e-06,
      "loss": 0.7239331,
      "num_input_tokens_seen": 82080400,
      "step": 3814,
      "time_per_iteration": 2.6281747817993164
    },
    {
      "auxiliary_loss_clip": 0.01187944,
      "auxiliary_loss_mlp": 0.01026532,
      "balance_loss_clip": 1.05770421,
      "balance_loss_mlp": 1.01839626,
      "epoch": 0.458726627788132,
      "flos": 26833889692800.0,
      "grad_norm": 2.064266621791038,
      "language_loss": 0.62789339,
      "learning_rate": 2.3625432915011443e-06,
      "loss": 0.65003812,
      "num_input_tokens_seen": 82102310,
      "step": 3815,
      "time_per_iteration": 2.719245195388794
    },
    {
      "auxiliary_loss_clip": 0.01185206,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 1.05618989,
      "balance_loss_mlp": 1.01790893,
      "epoch": 0.4588468706787711,
      "flos": 24097927680000.0,
      "grad_norm": 2.1648426445499522,
      "language_loss": 0.65348256,
      "learning_rate": 2.3617771975482334e-06,
      "loss": 0.67559785,
      "num_input_tokens_seen": 82121140,
      "step": 3816,
      "time_per_iteration": 2.6158108711242676
    },
    {
      "auxiliary_loss_clip": 0.01184912,
      "auxiliary_loss_mlp": 0.01037127,
      "balance_loss_clip": 1.0566318,
      "balance_loss_mlp": 1.02846014,
      "epoch": 0.4589671135694102,
      "flos": 17889331622400.0,
      "grad_norm": 1.5500795885128738,
      "language_loss": 0.74258959,
      "learning_rate": 2.3610110487140083e-06,
      "loss": 0.76480997,
      "num_input_tokens_seen": 82139575,
      "step": 3817,
      "time_per_iteration": 2.6741931438446045
    },
    {
      "auxiliary_loss_clip": 0.0118192,
      "auxiliary_loss_mlp": 0.01022784,
      "balance_loss_clip": 1.0540396,
      "balance_loss_mlp": 1.01462996,
      "epoch": 0.4590873564600493,
      "flos": 25626967781760.0,
      "grad_norm": 2.3362366825075145,
      "language_loss": 0.80583537,
      "learning_rate": 2.360244845114695e-06,
      "loss": 0.82788241,
      "num_input_tokens_seen": 82159195,
      "step": 3818,
      "time_per_iteration": 2.66005277633667
    },
    {
      "auxiliary_loss_clip": 0.01184285,
      "auxiliary_loss_mlp": 0.01021101,
      "balance_loss_clip": 1.05468655,
      "balance_loss_mlp": 1.01281595,
      "epoch": 0.4592075993506884,
      "flos": 18514788168960.0,
      "grad_norm": 2.6857405498426696,
      "language_loss": 0.68469715,
      "learning_rate": 2.3594785868665245e-06,
      "loss": 0.70675099,
      "num_input_tokens_seen": 82175500,
      "step": 3819,
      "time_per_iteration": 2.667421579360962
    },
    {
      "auxiliary_loss_clip": 0.01187392,
      "auxiliary_loss_mlp": 0.01024291,
      "balance_loss_clip": 1.057145,
      "balance_loss_mlp": 1.01574993,
      "epoch": 0.4593278422413275,
      "flos": 20631111638400.0,
      "grad_norm": 2.470500898471037,
      "language_loss": 0.80489999,
      "learning_rate": 2.3587122740857386e-06,
      "loss": 0.82701683,
      "num_input_tokens_seen": 82192600,
      "step": 3820,
      "time_per_iteration": 2.6353096961975098
    },
    {
      "auxiliary_loss_clip": 0.01179963,
      "auxiliary_loss_mlp": 0.01024357,
      "balance_loss_clip": 1.05343735,
      "balance_loss_mlp": 1.01667142,
      "epoch": 0.45944808513196655,
      "flos": 21358517961600.0,
      "grad_norm": 1.768074354043535,
      "language_loss": 0.78171003,
      "learning_rate": 2.357945906888586e-06,
      "loss": 0.8037532,
      "num_input_tokens_seen": 82212040,
      "step": 3821,
      "time_per_iteration": 2.6713461875915527
    },
    {
      "auxiliary_loss_clip": 0.01188164,
      "auxiliary_loss_mlp": 0.01030284,
      "balance_loss_clip": 1.05603123,
      "balance_loss_mlp": 1.02098536,
      "epoch": 0.45956832802260567,
      "flos": 21427789340160.0,
      "grad_norm": 2.47833313158153,
      "language_loss": 0.79628742,
      "learning_rate": 2.357179485391324e-06,
      "loss": 0.81847191,
      "num_input_tokens_seen": 82229895,
      "step": 3822,
      "time_per_iteration": 2.5897226333618164
    },
    {
      "auxiliary_loss_clip": 0.01179988,
      "auxiliary_loss_mlp": 0.01024539,
      "balance_loss_clip": 1.0541774,
      "balance_loss_mlp": 1.01677823,
      "epoch": 0.4596885709132448,
      "flos": 22382654538240.0,
      "grad_norm": 2.0378229820456757,
      "language_loss": 0.85568964,
      "learning_rate": 2.3564130097102173e-06,
      "loss": 0.8777349,
      "num_input_tokens_seen": 82249550,
      "step": 3823,
      "time_per_iteration": 2.7145400047302246
    },
    {
      "auxiliary_loss_clip": 0.01182397,
      "auxiliary_loss_mlp": 0.0102709,
      "balance_loss_clip": 1.05501342,
      "balance_loss_mlp": 1.01869202,
      "epoch": 0.45980881380388383,
      "flos": 28981957806720.0,
      "grad_norm": 2.419009324955025,
      "language_loss": 0.74962485,
      "learning_rate": 2.355646479961541e-06,
      "loss": 0.77171981,
      "num_input_tokens_seen": 82268860,
      "step": 3824,
      "time_per_iteration": 2.643979072570801
    },
    {
      "auxiliary_loss_clip": 0.01184037,
      "auxiliary_loss_mlp": 0.01028113,
      "balance_loss_clip": 1.05457163,
      "balance_loss_mlp": 1.01979852,
      "epoch": 0.45992905669452294,
      "flos": 33396599980800.0,
      "grad_norm": 3.626329915607056,
      "language_loss": 0.7154386,
      "learning_rate": 2.354879896261576e-06,
      "loss": 0.73756009,
      "num_input_tokens_seen": 82289070,
      "step": 3825,
      "time_per_iteration": 2.746083974838257
    },
    {
      "auxiliary_loss_clip": 0.01184855,
      "auxiliary_loss_mlp": 0.01028382,
      "balance_loss_clip": 1.0547061,
      "balance_loss_mlp": 1.01967955,
      "epoch": 0.46004929958516205,
      "flos": 36318184502400.0,
      "grad_norm": 3.636105369804241,
      "language_loss": 0.56551206,
      "learning_rate": 2.3541132587266133e-06,
      "loss": 0.5876444,
      "num_input_tokens_seen": 82311790,
      "step": 3826,
      "time_per_iteration": 2.78776216506958
    },
    {
      "auxiliary_loss_clip": 0.01187305,
      "auxiliary_loss_mlp": 0.01028186,
      "balance_loss_clip": 1.0568819,
      "balance_loss_mlp": 1.01982975,
      "epoch": 0.4601695424758011,
      "flos": 17238451224960.0,
      "grad_norm": 2.167013759772325,
      "language_loss": 0.68978512,
      "learning_rate": 2.3533465674729515e-06,
      "loss": 0.71194005,
      "num_input_tokens_seen": 82329020,
      "step": 3827,
      "time_per_iteration": 2.5843660831451416
    },
    {
      "auxiliary_loss_clip": 0.01184667,
      "auxiliary_loss_mlp": 0.01027672,
      "balance_loss_clip": 1.05575705,
      "balance_loss_mlp": 1.01839757,
      "epoch": 0.4602897853664402,
      "flos": 15888425529600.0,
      "grad_norm": 7.335528174050928,
      "language_loss": 0.72738445,
      "learning_rate": 2.352579822616895e-06,
      "loss": 0.7495079,
      "num_input_tokens_seen": 82346455,
      "step": 3828,
      "time_per_iteration": 2.63712477684021
    },
    {
      "auxiliary_loss_clip": 0.01183415,
      "auxiliary_loss_mlp": 0.01025475,
      "balance_loss_clip": 1.05586386,
      "balance_loss_mlp": 1.01735711,
      "epoch": 0.4604100282570793,
      "flos": 25412617370880.0,
      "grad_norm": 2.1135682809053336,
      "language_loss": 0.77470714,
      "learning_rate": 2.351813024274761e-06,
      "loss": 0.79679602,
      "num_input_tokens_seen": 82367810,
      "step": 3829,
      "time_per_iteration": 2.6236610412597656
    },
    {
      "auxiliary_loss_clip": 0.01185167,
      "auxiliary_loss_mlp": 0.01029555,
      "balance_loss_clip": 1.05669689,
      "balance_loss_mlp": 1.0209837,
      "epoch": 0.4605302711477184,
      "flos": 27630711048960.0,
      "grad_norm": 1.8592451081467765,
      "language_loss": 0.73456103,
      "learning_rate": 2.3510461725628693e-06,
      "loss": 0.75670826,
      "num_input_tokens_seen": 82388275,
      "step": 3830,
      "time_per_iteration": 2.7149908542633057
    },
    {
      "auxiliary_loss_clip": 0.01183074,
      "auxiliary_loss_mlp": 0.01027087,
      "balance_loss_clip": 1.05554795,
      "balance_loss_mlp": 1.01860559,
      "epoch": 0.4606505140383575,
      "flos": 23839657914240.0,
      "grad_norm": 2.091540206938377,
      "language_loss": 0.71150208,
      "learning_rate": 2.350279267597554e-06,
      "loss": 0.73360366,
      "num_input_tokens_seen": 82408915,
      "step": 3831,
      "time_per_iteration": 3.546818733215332
    },
    {
      "auxiliary_loss_clip": 0.01185375,
      "auxiliary_loss_mlp": 0.01025135,
      "balance_loss_clip": 1.05529892,
      "balance_loss_mlp": 1.0161283,
      "epoch": 0.46077075692899655,
      "flos": 16107013745280.0,
      "grad_norm": 2.4296345658369076,
      "language_loss": 0.82645434,
      "learning_rate": 2.3495123094951515e-06,
      "loss": 0.8485595,
      "num_input_tokens_seen": 82427260,
      "step": 3832,
      "time_per_iteration": 2.632294178009033
    },
    {
      "auxiliary_loss_clip": 0.01183199,
      "auxiliary_loss_mlp": 0.01029319,
      "balance_loss_clip": 1.05497742,
      "balance_loss_mlp": 1.02039647,
      "epoch": 0.46089099981963566,
      "flos": 48798147634560.0,
      "grad_norm": 2.2310041735213333,
      "language_loss": 0.76263863,
      "learning_rate": 2.34874529837201e-06,
      "loss": 0.78476381,
      "num_input_tokens_seen": 82450805,
      "step": 3833,
      "time_per_iteration": 2.8856754302978516
    },
    {
      "auxiliary_loss_clip": 0.01181846,
      "auxiliary_loss_mlp": 0.01022958,
      "balance_loss_clip": 1.05447948,
      "balance_loss_mlp": 1.0144105,
      "epoch": 0.46101124271027477,
      "flos": 19099234362240.0,
      "grad_norm": 2.0129893485612658,
      "language_loss": 0.78827876,
      "learning_rate": 2.347978234344483e-06,
      "loss": 0.81032681,
      "num_input_tokens_seen": 82467010,
      "step": 3834,
      "time_per_iteration": 2.5897929668426514
    },
    {
      "auxiliary_loss_clip": 0.01188418,
      "auxiliary_loss_mlp": 0.01031718,
      "balance_loss_clip": 1.05718827,
      "balance_loss_mlp": 1.02230644,
      "epoch": 0.4611314856009138,
      "flos": 39347931853440.0,
      "grad_norm": 2.0002097866132877,
      "language_loss": 0.69043201,
      "learning_rate": 2.347211117528935e-06,
      "loss": 0.71263337,
      "num_input_tokens_seen": 82489310,
      "step": 3835,
      "time_per_iteration": 3.666682004928589
    },
    {
      "auxiliary_loss_clip": 0.01187223,
      "auxiliary_loss_mlp": 0.01031931,
      "balance_loss_clip": 1.05903196,
      "balance_loss_mlp": 1.02288592,
      "epoch": 0.46125172849155294,
      "flos": 20810772489600.0,
      "grad_norm": 2.6644482276362877,
      "language_loss": 0.72070861,
      "learning_rate": 2.3464439480417374e-06,
      "loss": 0.74290019,
      "num_input_tokens_seen": 82508830,
      "step": 3836,
      "time_per_iteration": 2.606269359588623
    },
    {
      "auxiliary_loss_clip": 0.01186059,
      "auxiliary_loss_mlp": 0.01030478,
      "balance_loss_clip": 1.05578148,
      "balance_loss_mlp": 1.02086401,
      "epoch": 0.46137197138219205,
      "flos": 17930808852480.0,
      "grad_norm": 3.0751921905587065,
      "language_loss": 0.7776655,
      "learning_rate": 2.3456767259992676e-06,
      "loss": 0.79983091,
      "num_input_tokens_seen": 82526475,
      "step": 3837,
      "time_per_iteration": 3.505676746368408
    },
    {
      "auxiliary_loss_clip": 0.01184375,
      "auxiliary_loss_mlp": 0.01028418,
      "balance_loss_clip": 1.05385303,
      "balance_loss_mlp": 1.0195365,
      "epoch": 0.4614922142728311,
      "flos": 16836610798080.0,
      "grad_norm": 2.2325369651821276,
      "language_loss": 0.88516784,
      "learning_rate": 2.3449094515179135e-06,
      "loss": 0.9072957,
      "num_input_tokens_seen": 82543935,
      "step": 3838,
      "time_per_iteration": 3.5258257389068604
    },
    {
      "auxiliary_loss_clip": 0.01184238,
      "auxiliary_loss_mlp": 0.01027194,
      "balance_loss_clip": 1.05522299,
      "balance_loss_mlp": 1.01847363,
      "epoch": 0.4616124571634702,
      "flos": 26614906427520.0,
      "grad_norm": 1.6726725455217806,
      "language_loss": 0.81940866,
      "learning_rate": 2.34414212471407e-06,
      "loss": 0.84152305,
      "num_input_tokens_seen": 82563730,
      "step": 3839,
      "time_per_iteration": 2.6309075355529785
    },
    {
      "auxiliary_loss_clip": 0.01186981,
      "auxiliary_loss_mlp": 0.01027226,
      "balance_loss_clip": 1.05524063,
      "balance_loss_mlp": 1.01799333,
      "epoch": 0.4617327000541093,
      "flos": 20340127560960.0,
      "grad_norm": 1.9171441485272953,
      "language_loss": 0.73071229,
      "learning_rate": 2.3433747457041394e-06,
      "loss": 0.75285435,
      "num_input_tokens_seen": 82582435,
      "step": 3840,
      "time_per_iteration": 2.6700448989868164
    },
    {
      "auxiliary_loss_clip": 0.01186826,
      "auxiliary_loss_mlp": 0.01026817,
      "balance_loss_clip": 1.05632091,
      "balance_loss_mlp": 1.01769161,
      "epoch": 0.4618529429447484,
      "flos": 29570749545600.0,
      "grad_norm": 3.7588630718386473,
      "language_loss": 0.85228539,
      "learning_rate": 2.342607314604533e-06,
      "loss": 0.87442183,
      "num_input_tokens_seen": 82602185,
      "step": 3841,
      "time_per_iteration": 2.682589530944824
    },
    {
      "auxiliary_loss_clip": 0.01185044,
      "auxiliary_loss_mlp": 0.01035078,
      "balance_loss_clip": 1.05555022,
      "balance_loss_mlp": 1.0254519,
      "epoch": 0.4619731858353875,
      "flos": 19787030962560.0,
      "grad_norm": 38.21182192727919,
      "language_loss": 0.84275234,
      "learning_rate": 2.3418398315316694e-06,
      "loss": 0.86495358,
      "num_input_tokens_seen": 82620005,
      "step": 3842,
      "time_per_iteration": 2.634704113006592
    },
    {
      "auxiliary_loss_clip": 0.0118321,
      "auxiliary_loss_mlp": 0.01033702,
      "balance_loss_clip": 1.05540073,
      "balance_loss_mlp": 1.02521992,
      "epoch": 0.4620934287260266,
      "flos": 18951138587520.0,
      "grad_norm": 2.4764570627055047,
      "language_loss": 0.78415698,
      "learning_rate": 2.3410722966019755e-06,
      "loss": 0.80632609,
      "num_input_tokens_seen": 82635120,
      "step": 3843,
      "time_per_iteration": 2.5685696601867676
    },
    {
      "auxiliary_loss_clip": 0.01182946,
      "auxiliary_loss_mlp": 0.01031707,
      "balance_loss_clip": 1.0545069,
      "balance_loss_mlp": 1.0226891,
      "epoch": 0.46221367161666566,
      "flos": 37341674634240.0,
      "grad_norm": 2.132634175485031,
      "language_loss": 0.65534836,
      "learning_rate": 2.3403047099318848e-06,
      "loss": 0.67749488,
      "num_input_tokens_seen": 82659190,
      "step": 3844,
      "time_per_iteration": 2.7618045806884766
    },
    {
      "auxiliary_loss_clip": 0.01183617,
      "auxiliary_loss_mlp": 0.01034646,
      "balance_loss_clip": 1.05411947,
      "balance_loss_mlp": 1.02578902,
      "epoch": 0.46233391450730477,
      "flos": 14428549065600.0,
      "grad_norm": 2.282809883746521,
      "language_loss": 0.74956489,
      "learning_rate": 2.3395370716378405e-06,
      "loss": 0.77174759,
      "num_input_tokens_seen": 82676635,
      "step": 3845,
      "time_per_iteration": 2.577341079711914
    },
    {
      "auxiliary_loss_clip": 0.01185521,
      "auxiliary_loss_mlp": 0.01029689,
      "balance_loss_clip": 1.05552745,
      "balance_loss_mlp": 1.02182722,
      "epoch": 0.4624541573979438,
      "flos": 22493044010880.0,
      "grad_norm": 2.231607612927241,
      "language_loss": 0.72523791,
      "learning_rate": 2.338769381836292e-06,
      "loss": 0.74738997,
      "num_input_tokens_seen": 82696245,
      "step": 3846,
      "time_per_iteration": 2.7537319660186768
    },
    {
      "auxiliary_loss_clip": 0.01183663,
      "auxiliary_loss_mlp": 0.01030554,
      "balance_loss_clip": 1.05378747,
      "balance_loss_mlp": 1.02163076,
      "epoch": 0.46257440028858293,
      "flos": 14465070218880.0,
      "grad_norm": 2.641500876625124,
      "language_loss": 0.73304725,
      "learning_rate": 2.3380016406436984e-06,
      "loss": 0.75518942,
      "num_input_tokens_seen": 82713725,
      "step": 3847,
      "time_per_iteration": 2.62556791305542
    },
    {
      "auxiliary_loss_clip": 0.01187868,
      "auxiliary_loss_mlp": 0.0102725,
      "balance_loss_clip": 1.0577451,
      "balance_loss_mlp": 1.01851165,
      "epoch": 0.46269464317922204,
      "flos": 23332204523520.0,
      "grad_norm": 2.2448648483272167,
      "language_loss": 0.81719345,
      "learning_rate": 2.337233848176524e-06,
      "loss": 0.83934462,
      "num_input_tokens_seen": 82731495,
      "step": 3848,
      "time_per_iteration": 2.6309854984283447
    },
    {
      "auxiliary_loss_clip": 0.0118659,
      "auxiliary_loss_mlp": 0.01029093,
      "balance_loss_clip": 1.05736291,
      "balance_loss_mlp": 1.01948452,
      "epoch": 0.4628148860698611,
      "flos": 18552027594240.0,
      "grad_norm": 4.294989657791737,
      "language_loss": 0.83647549,
      "learning_rate": 2.3364660045512435e-06,
      "loss": 0.85863233,
      "num_input_tokens_seen": 82750255,
      "step": 3849,
      "time_per_iteration": 2.666165351867676
    },
    {
      "auxiliary_loss_clip": 0.01085272,
      "auxiliary_loss_mlp": 0.01000992,
      "balance_loss_clip": 1.02143085,
      "balance_loss_mlp": 0.99960911,
      "epoch": 0.4629351289605002,
      "flos": 70667569670400.0,
      "grad_norm": 0.7409943502292967,
      "language_loss": 0.58163208,
      "learning_rate": 2.335698109884337e-06,
      "loss": 0.60249472,
      "num_input_tokens_seen": 82815460,
      "step": 3850,
      "time_per_iteration": 3.3828535079956055
    },
    {
      "auxiliary_loss_clip": 0.01085242,
      "auxiliary_loss_mlp": 0.01001177,
      "balance_loss_clip": 1.02163315,
      "balance_loss_mlp": 0.99981254,
      "epoch": 0.4630553718511393,
      "flos": 59687200465920.0,
      "grad_norm": 0.8937684428992114,
      "language_loss": 0.59871328,
      "learning_rate": 2.334930164292294e-06,
      "loss": 0.61957741,
      "num_input_tokens_seen": 82878010,
      "step": 3851,
      "time_per_iteration": 3.343456745147705
    },
    {
      "auxiliary_loss_clip": 0.01185981,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.05667925,
      "balance_loss_mlp": 1.01840365,
      "epoch": 0.4631756147417784,
      "flos": 15960605909760.0,
      "grad_norm": 1.9658928911440072,
      "language_loss": 0.80086207,
      "learning_rate": 2.334162167891612e-06,
      "loss": 0.82298607,
      "num_input_tokens_seen": 82895275,
      "step": 3852,
      "time_per_iteration": 2.6368157863616943
    },
    {
      "auxiliary_loss_clip": 0.01186804,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.0570147,
      "balance_loss_mlp": 1.01878357,
      "epoch": 0.4632958576324175,
      "flos": 16472907636480.0,
      "grad_norm": 3.168358946391235,
      "language_loss": 0.75343931,
      "learning_rate": 2.333394120798795e-06,
      "loss": 0.77558327,
      "num_input_tokens_seen": 82914010,
      "step": 3853,
      "time_per_iteration": 2.66949725151062
    },
    {
      "auxiliary_loss_clip": 0.0118234,
      "auxiliary_loss_mlp": 0.0102333,
      "balance_loss_clip": 1.05254912,
      "balance_loss_mlp": 1.01463938,
      "epoch": 0.4634161005230566,
      "flos": 22346492520960.0,
      "grad_norm": 3.6319013425398063,
      "language_loss": 0.72145164,
      "learning_rate": 2.3326260231303545e-06,
      "loss": 0.74350834,
      "num_input_tokens_seen": 82932610,
      "step": 3854,
      "time_per_iteration": 2.671882152557373
    },
    {
      "auxiliary_loss_clip": 0.01182028,
      "auxiliary_loss_mlp": 0.01024159,
      "balance_loss_clip": 1.05615449,
      "balance_loss_mlp": 1.01668453,
      "epoch": 0.46353634341369565,
      "flos": 15742233175680.0,
      "grad_norm": 2.0849281424515747,
      "language_loss": 0.86798263,
      "learning_rate": 2.331857875002811e-06,
      "loss": 0.89004445,
      "num_input_tokens_seen": 82951210,
      "step": 3855,
      "time_per_iteration": 2.6266000270843506
    },
    {
      "auxiliary_loss_clip": 0.01187501,
      "auxiliary_loss_mlp": 0.01031562,
      "balance_loss_clip": 1.06060803,
      "balance_loss_mlp": 1.02287149,
      "epoch": 0.46365658630433476,
      "flos": 28329820433280.0,
      "grad_norm": 1.6852313679782362,
      "language_loss": 0.76327741,
      "learning_rate": 2.3310896765326916e-06,
      "loss": 0.7854681,
      "num_input_tokens_seen": 82972210,
      "step": 3856,
      "time_per_iteration": 2.7101900577545166
    },
    {
      "auxiliary_loss_clip": 0.01187197,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.05702329,
      "balance_loss_mlp": 1.01866138,
      "epoch": 0.46377682919497387,
      "flos": 24608074590720.0,
      "grad_norm": 2.0852227265367587,
      "language_loss": 0.84121865,
      "learning_rate": 2.330321427836531e-06,
      "loss": 0.86336517,
      "num_input_tokens_seen": 82994080,
      "step": 3857,
      "time_per_iteration": 3.624746084213257
    },
    {
      "auxiliary_loss_clip": 0.01186617,
      "auxiliary_loss_mlp": 0.01024281,
      "balance_loss_clip": 1.05811167,
      "balance_loss_mlp": 1.01611865,
      "epoch": 0.4638970720856129,
      "flos": 19060953442560.0,
      "grad_norm": 2.1866580990366202,
      "language_loss": 0.82605505,
      "learning_rate": 2.3295531290308733e-06,
      "loss": 0.84816402,
      "num_input_tokens_seen": 83012230,
      "step": 3858,
      "time_per_iteration": 2.685763120651245
    },
    {
      "auxiliary_loss_clip": 0.0119176,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.05923116,
      "balance_loss_mlp": 1.01937366,
      "epoch": 0.46401731497625204,
      "flos": 18471012468480.0,
      "grad_norm": 2.7026179456204815,
      "language_loss": 0.75857687,
      "learning_rate": 2.3287847802322678e-06,
      "loss": 0.78078043,
      "num_input_tokens_seen": 83027800,
      "step": 3859,
      "time_per_iteration": 2.625837564468384
    },
    {
      "auxiliary_loss_clip": 0.01189335,
      "auxiliary_loss_mlp": 0.01037803,
      "balance_loss_clip": 1.05711615,
      "balance_loss_mlp": 1.02776551,
      "epoch": 0.4641375578668911,
      "flos": 26067053214720.0,
      "grad_norm": 2.1019801469780472,
      "language_loss": 0.84261233,
      "learning_rate": 2.3280163815572723e-06,
      "loss": 0.86488372,
      "num_input_tokens_seen": 83048395,
      "step": 3860,
      "time_per_iteration": 2.7524302005767822
    },
    {
      "auxiliary_loss_clip": 0.01184998,
      "auxiliary_loss_mlp": 0.01025274,
      "balance_loss_clip": 1.05580842,
      "balance_loss_mlp": 1.01725101,
      "epoch": 0.4642578007575302,
      "flos": 19570382081280.0,
      "grad_norm": 5.52800556033923,
      "language_loss": 0.77114463,
      "learning_rate": 2.3272479331224522e-06,
      "loss": 0.79324734,
      "num_input_tokens_seen": 83065825,
      "step": 3861,
      "time_per_iteration": 3.551731824874878
    },
    {
      "auxiliary_loss_clip": 0.01187222,
      "auxiliary_loss_mlp": 0.01031344,
      "balance_loss_clip": 1.05607986,
      "balance_loss_mlp": 1.02279067,
      "epoch": 0.4643780436481693,
      "flos": 28186249772160.0,
      "grad_norm": 3.051107300669057,
      "language_loss": 0.7772702,
      "learning_rate": 2.3264794350443817e-06,
      "loss": 0.79945582,
      "num_input_tokens_seen": 83087920,
      "step": 3862,
      "time_per_iteration": 2.6837308406829834
    },
    {
      "auxiliary_loss_clip": 0.01184737,
      "auxiliary_loss_mlp": 0.01024446,
      "balance_loss_clip": 1.05542231,
      "balance_loss_mlp": 1.01566672,
      "epoch": 0.46449828653880837,
      "flos": 25375270204800.0,
      "grad_norm": 1.9289959558710164,
      "language_loss": 0.78425711,
      "learning_rate": 2.3257108874396396e-06,
      "loss": 0.80634892,
      "num_input_tokens_seen": 83109015,
      "step": 3863,
      "time_per_iteration": 2.690980911254883
    },
    {
      "auxiliary_loss_clip": 0.01182995,
      "auxiliary_loss_mlp": 0.01030866,
      "balance_loss_clip": 1.05634725,
      "balance_loss_mlp": 1.02277422,
      "epoch": 0.4646185294294475,
      "flos": 16034330574720.0,
      "grad_norm": 2.016009736950604,
      "language_loss": 0.73425531,
      "learning_rate": 2.3249422904248152e-06,
      "loss": 0.75639391,
      "num_input_tokens_seen": 83127450,
      "step": 3864,
      "time_per_iteration": 4.466308832168579
    },
    {
      "auxiliary_loss_clip": 0.01183982,
      "auxiliary_loss_mlp": 0.01025927,
      "balance_loss_clip": 1.05661154,
      "balance_loss_mlp": 1.01722479,
      "epoch": 0.4647387723200866,
      "flos": 26363101109760.0,
      "grad_norm": 1.5451175091296732,
      "language_loss": 0.87064105,
      "learning_rate": 2.324173644116504e-06,
      "loss": 0.89274019,
      "num_input_tokens_seen": 83150300,
      "step": 3865,
      "time_per_iteration": 2.756955623626709
    },
    {
      "auxiliary_loss_clip": 0.01184179,
      "auxiliary_loss_mlp": 0.01033449,
      "balance_loss_clip": 1.05664229,
      "balance_loss_mlp": 1.02470529,
      "epoch": 0.46485901521072565,
      "flos": 27160209774720.0,
      "grad_norm": 1.9205149293134471,
      "language_loss": 0.81931627,
      "learning_rate": 2.3234049486313087e-06,
      "loss": 0.84149265,
      "num_input_tokens_seen": 83171750,
      "step": 3866,
      "time_per_iteration": 2.7220816612243652
    },
    {
      "auxiliary_loss_clip": 0.01181927,
      "auxiliary_loss_mlp": 0.01025036,
      "balance_loss_clip": 1.0558238,
      "balance_loss_mlp": 1.01734114,
      "epoch": 0.46497925810136476,
      "flos": 24279851088000.0,
      "grad_norm": 1.9335443327833317,
      "language_loss": 0.75906837,
      "learning_rate": 2.322636204085839e-06,
      "loss": 0.78113794,
      "num_input_tokens_seen": 83191820,
      "step": 3867,
      "time_per_iteration": 2.6490683555603027
    },
    {
      "auxiliary_loss_clip": 0.01182862,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.05498672,
      "balance_loss_mlp": 1.01924586,
      "epoch": 0.46509950099200387,
      "flos": 16253134272000.0,
      "grad_norm": 2.4729783950957467,
      "language_loss": 0.78868616,
      "learning_rate": 2.3218674105967143e-06,
      "loss": 0.81078529,
      "num_input_tokens_seen": 83210085,
      "step": 3868,
      "time_per_iteration": 2.719346046447754
    },
    {
      "auxiliary_loss_clip": 0.01183872,
      "auxiliary_loss_mlp": 0.01026653,
      "balance_loss_clip": 1.05623031,
      "balance_loss_mlp": 1.0183444,
      "epoch": 0.4652197438826429,
      "flos": 23442270773760.0,
      "grad_norm": 1.6705578028446597,
      "language_loss": 0.83437586,
      "learning_rate": 2.3210985682805593e-06,
      "loss": 0.85648119,
      "num_input_tokens_seen": 83231865,
      "step": 3869,
      "time_per_iteration": 2.6892306804656982
    },
    {
      "auxiliary_loss_clip": 0.01188036,
      "auxiliary_loss_mlp": 0.01027988,
      "balance_loss_clip": 1.05937541,
      "balance_loss_mlp": 1.01867199,
      "epoch": 0.46533998677328203,
      "flos": 16216397637120.0,
      "grad_norm": 3.123567792535282,
      "language_loss": 0.68273449,
      "learning_rate": 2.320329677254007e-06,
      "loss": 0.70489472,
      "num_input_tokens_seen": 83249195,
      "step": 3870,
      "time_per_iteration": 2.6649320125579834
    },
    {
      "auxiliary_loss_clip": 0.01185241,
      "auxiliary_loss_mlp": 0.0103024,
      "balance_loss_clip": 1.05660391,
      "balance_loss_mlp": 1.02179384,
      "epoch": 0.46546022966392114,
      "flos": 21141869080320.0,
      "grad_norm": 2.812216569588495,
      "language_loss": 0.72857201,
      "learning_rate": 2.319560737633697e-06,
      "loss": 0.75072682,
      "num_input_tokens_seen": 83267915,
      "step": 3871,
      "time_per_iteration": 2.614287853240967
    },
    {
      "auxiliary_loss_clip": 0.01189845,
      "auxiliary_loss_mlp": 0.01029282,
      "balance_loss_clip": 1.05679083,
      "balance_loss_mlp": 1.01942921,
      "epoch": 0.4655804725545602,
      "flos": 41171942442240.0,
      "grad_norm": 3.5418298304311846,
      "language_loss": 0.67655444,
      "learning_rate": 2.3187917495362775e-06,
      "loss": 0.69874573,
      "num_input_tokens_seen": 83292325,
      "step": 3872,
      "time_per_iteration": 2.798100233078003
    },
    {
      "auxiliary_loss_clip": 0.01185763,
      "auxiliary_loss_mlp": 0.01039254,
      "balance_loss_clip": 1.05838001,
      "balance_loss_mlp": 1.03128457,
      "epoch": 0.4657007154451993,
      "flos": 19570956698880.0,
      "grad_norm": 2.473794241180278,
      "language_loss": 0.77015722,
      "learning_rate": 2.318022713078403e-06,
      "loss": 0.79240739,
      "num_input_tokens_seen": 83306905,
      "step": 3873,
      "time_per_iteration": 2.6048057079315186
    },
    {
      "auxiliary_loss_clip": 0.01184191,
      "auxiliary_loss_mlp": 0.01023605,
      "balance_loss_clip": 1.05626178,
      "balance_loss_mlp": 1.01515877,
      "epoch": 0.4658209583358384,
      "flos": 15517826956800.0,
      "grad_norm": 2.9728855824500804,
      "language_loss": 0.85323596,
      "learning_rate": 2.3172536283767354e-06,
      "loss": 0.875314,
      "num_input_tokens_seen": 83320665,
      "step": 3874,
      "time_per_iteration": 2.6589136123657227
    },
    {
      "auxiliary_loss_clip": 0.01190985,
      "auxiliary_loss_mlp": 0.01034225,
      "balance_loss_clip": 1.06309819,
      "balance_loss_mlp": 1.0249927,
      "epoch": 0.4659412012264775,
      "flos": 14903180403840.0,
      "grad_norm": 2.033017799714049,
      "language_loss": 0.80789411,
      "learning_rate": 2.3164844955479447e-06,
      "loss": 0.83014619,
      "num_input_tokens_seen": 83336475,
      "step": 3875,
      "time_per_iteration": 2.57712984085083
    },
    {
      "auxiliary_loss_clip": 0.01187777,
      "auxiliary_loss_mlp": 0.01023309,
      "balance_loss_clip": 1.05908036,
      "balance_loss_mlp": 1.01473737,
      "epoch": 0.4660614441171166,
      "flos": 24425612478720.0,
      "grad_norm": 2.3158223101276687,
      "language_loss": 0.70472097,
      "learning_rate": 2.3157153147087082e-06,
      "loss": 0.72683179,
      "num_input_tokens_seen": 83358365,
      "step": 3876,
      "time_per_iteration": 2.7257018089294434
    },
    {
      "auxiliary_loss_clip": 0.01187952,
      "auxiliary_loss_mlp": 0.01027564,
      "balance_loss_clip": 1.0596633,
      "balance_loss_mlp": 1.0192908,
      "epoch": 0.46618168700775564,
      "flos": 22091095843200.0,
      "grad_norm": 1.9562981736832121,
      "language_loss": 0.83296943,
      "learning_rate": 2.314946085975709e-06,
      "loss": 0.85512459,
      "num_input_tokens_seen": 83377345,
      "step": 3877,
      "time_per_iteration": 2.689816951751709
    },
    {
      "auxiliary_loss_clip": 0.01185675,
      "auxiliary_loss_mlp": 0.01025279,
      "balance_loss_clip": 1.06014395,
      "balance_loss_mlp": 1.01742291,
      "epoch": 0.46630192989839475,
      "flos": 26176975810560.0,
      "grad_norm": 1.860990066623377,
      "language_loss": 0.82529879,
      "learning_rate": 2.3141768094656393e-06,
      "loss": 0.84740829,
      "num_input_tokens_seen": 83395920,
      "step": 3878,
      "time_per_iteration": 2.679969072341919
    },
    {
      "auxiliary_loss_clip": 0.0118648,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.05958903,
      "balance_loss_mlp": 1.01813054,
      "epoch": 0.46642217278903386,
      "flos": 11509622150400.0,
      "grad_norm": 18.248894659425535,
      "language_loss": 0.83153778,
      "learning_rate": 2.3134074852951966e-06,
      "loss": 0.85366678,
      "num_input_tokens_seen": 83412510,
      "step": 3879,
      "time_per_iteration": 2.6374595165252686
    },
    {
      "auxiliary_loss_clip": 0.01188991,
      "auxiliary_loss_mlp": 0.0102606,
      "balance_loss_clip": 1.06256926,
      "balance_loss_mlp": 1.01776612,
      "epoch": 0.4665424156796729,
      "flos": 32306819299200.0,
      "grad_norm": 3.5789182101936077,
      "language_loss": 0.77948952,
      "learning_rate": 2.312638113581088e-06,
      "loss": 0.80164003,
      "num_input_tokens_seen": 83432995,
      "step": 3880,
      "time_per_iteration": 2.740518093109131
    },
    {
      "auxiliary_loss_clip": 0.01185429,
      "auxiliary_loss_mlp": 0.01024618,
      "balance_loss_clip": 1.05729759,
      "balance_loss_mlp": 1.0165658,
      "epoch": 0.46666265857031203,
      "flos": 18436179254400.0,
      "grad_norm": 3.0099385824882474,
      "language_loss": 0.78681225,
      "learning_rate": 2.311868694440027e-06,
      "loss": 0.80891275,
      "num_input_tokens_seen": 83447415,
      "step": 3881,
      "time_per_iteration": 2.6278998851776123
    },
    {
      "auxiliary_loss_clip": 0.01090634,
      "auxiliary_loss_mlp": 0.0100474,
      "balance_loss_clip": 1.02750492,
      "balance_loss_mlp": 1.00329196,
      "epoch": 0.46678290146095114,
      "flos": 68438989221120.0,
      "grad_norm": 0.7319902626547302,
      "language_loss": 0.62465352,
      "learning_rate": 2.3110992279887323e-06,
      "loss": 0.64560729,
      "num_input_tokens_seen": 83519340,
      "step": 3882,
      "time_per_iteration": 3.292335033416748
    },
    {
      "auxiliary_loss_clip": 0.01186705,
      "auxiliary_loss_mlp": 0.01028432,
      "balance_loss_clip": 1.05823827,
      "balance_loss_mlp": 1.01893711,
      "epoch": 0.4669031443515902,
      "flos": 17712507945600.0,
      "grad_norm": 2.1582230224654064,
      "language_loss": 0.84232962,
      "learning_rate": 2.310329714343932e-06,
      "loss": 0.86448097,
      "num_input_tokens_seen": 83535490,
      "step": 3883,
      "time_per_iteration": 2.845539093017578
    },
    {
      "auxiliary_loss_clip": 0.0118683,
      "auxiliary_loss_mlp": 0.0102499,
      "balance_loss_clip": 1.06011415,
      "balance_loss_mlp": 1.01693749,
      "epoch": 0.4670233872422293,
      "flos": 23947748916480.0,
      "grad_norm": 2.0480542632751395,
      "language_loss": 0.82115805,
      "learning_rate": 2.309560153622361e-06,
      "loss": 0.84327626,
      "num_input_tokens_seen": 83552400,
      "step": 3884,
      "time_per_iteration": 3.624058485031128
    },
    {
      "auxiliary_loss_clip": 0.01186978,
      "auxiliary_loss_mlp": 0.01024159,
      "balance_loss_clip": 1.05917001,
      "balance_loss_mlp": 1.01548672,
      "epoch": 0.4671436301328684,
      "flos": 28111268131200.0,
      "grad_norm": 2.120657472160983,
      "language_loss": 0.74464345,
      "learning_rate": 2.3087905459407602e-06,
      "loss": 0.76675481,
      "num_input_tokens_seen": 83571340,
      "step": 3885,
      "time_per_iteration": 2.703113079071045
    },
    {
      "auxiliary_loss_clip": 0.01090171,
      "auxiliary_loss_mlp": 0.01000695,
      "balance_loss_clip": 1.0270927,
      "balance_loss_mlp": 0.99935955,
      "epoch": 0.46726387302350747,
      "flos": 69369684566400.0,
      "grad_norm": 0.7942900298130943,
      "language_loss": 0.62823689,
      "learning_rate": 2.3080208914158795e-06,
      "loss": 0.64914554,
      "num_input_tokens_seen": 83634340,
      "step": 3886,
      "time_per_iteration": 3.225168228149414
    },
    {
      "auxiliary_loss_clip": 0.01189397,
      "auxiliary_loss_mlp": 0.01023597,
      "balance_loss_clip": 1.06189513,
      "balance_loss_mlp": 1.01424479,
      "epoch": 0.4673841159141466,
      "flos": 25519666878720.0,
      "grad_norm": 2.240643213143422,
      "language_loss": 0.72454393,
      "learning_rate": 2.3072511901644753e-06,
      "loss": 0.74667394,
      "num_input_tokens_seen": 83653410,
      "step": 3887,
      "time_per_iteration": 2.641084671020508
    },
    {
      "auxiliary_loss_clip": 0.01184049,
      "auxiliary_loss_mlp": 0.01024219,
      "balance_loss_clip": 1.05928206,
      "balance_loss_mlp": 1.01661968,
      "epoch": 0.4675043588047857,
      "flos": 24499265316480.0,
      "grad_norm": 5.07659703609059,
      "language_loss": 0.8113786,
      "learning_rate": 2.306481442303309e-06,
      "loss": 0.83346134,
      "num_input_tokens_seen": 83672985,
      "step": 3888,
      "time_per_iteration": 3.5863780975341797
    },
    {
      "auxiliary_loss_clip": 0.01187007,
      "auxiliary_loss_mlp": 0.01028659,
      "balance_loss_clip": 1.05759454,
      "balance_loss_mlp": 1.020082,
      "epoch": 0.46762460169542475,
      "flos": 20960771685120.0,
      "grad_norm": 2.1418211072864146,
      "language_loss": 0.73420674,
      "learning_rate": 2.3057116479491515e-06,
      "loss": 0.75636339,
      "num_input_tokens_seen": 83692395,
      "step": 3889,
      "time_per_iteration": 2.686478853225708
    },
    {
      "auxiliary_loss_clip": 0.01187155,
      "auxiliary_loss_mlp": 0.01026087,
      "balance_loss_clip": 1.05941844,
      "balance_loss_mlp": 1.01804626,
      "epoch": 0.46774484458606386,
      "flos": 19171666137600.0,
      "grad_norm": 4.074298595264454,
      "language_loss": 0.76276362,
      "learning_rate": 2.30494180721878e-06,
      "loss": 0.78489602,
      "num_input_tokens_seen": 83709735,
      "step": 3890,
      "time_per_iteration": 4.5088722705841064
    },
    {
      "auxiliary_loss_clip": 0.01183232,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.05836749,
      "balance_loss_mlp": 1.01766562,
      "epoch": 0.4678650874767029,
      "flos": 17967689141760.0,
      "grad_norm": 2.0444265324410504,
      "language_loss": 0.89659113,
      "learning_rate": 2.3041719202289794e-06,
      "loss": 0.9186781,
      "num_input_tokens_seen": 83725910,
      "step": 3891,
      "time_per_iteration": 2.649590492248535
    },
    {
      "auxiliary_loss_clip": 0.01186133,
      "auxiliary_loss_mlp": 0.01026153,
      "balance_loss_clip": 1.06117344,
      "balance_loss_mlp": 1.01834428,
      "epoch": 0.467985330367342,
      "flos": 21360816432000.0,
      "grad_norm": 2.565500629924067,
      "language_loss": 0.802652,
      "learning_rate": 2.30340198709654e-06,
      "loss": 0.82477486,
      "num_input_tokens_seen": 83745745,
      "step": 3892,
      "time_per_iteration": 2.7221758365631104
    },
    {
      "auxiliary_loss_clip": 0.01188123,
      "auxiliary_loss_mlp": 0.010271,
      "balance_loss_clip": 1.06116962,
      "balance_loss_mlp": 1.01882076,
      "epoch": 0.46810557325798113,
      "flos": 20521835487360.0,
      "grad_norm": 2.093131948185471,
      "language_loss": 0.74361348,
      "learning_rate": 2.3026320079382605e-06,
      "loss": 0.76576573,
      "num_input_tokens_seen": 83762680,
      "step": 3893,
      "time_per_iteration": 2.6106061935424805
    },
    {
      "auxiliary_loss_clip": 0.0118546,
      "auxiliary_loss_mlp": 0.01025351,
      "balance_loss_clip": 1.05947757,
      "balance_loss_mlp": 1.01806176,
      "epoch": 0.4682258161486202,
      "flos": 30117848572800.0,
      "grad_norm": 2.4915238423931965,
      "language_loss": 0.76532316,
      "learning_rate": 2.3018619828709454e-06,
      "loss": 0.78743136,
      "num_input_tokens_seen": 83784220,
      "step": 3894,
      "time_per_iteration": 2.716184616088867
    },
    {
      "auxiliary_loss_clip": 0.01187407,
      "auxiliary_loss_mlp": 0.01029556,
      "balance_loss_clip": 1.06050682,
      "balance_loss_mlp": 1.0216465,
      "epoch": 0.4683460590392593,
      "flos": 25293357239040.0,
      "grad_norm": 4.334440419592175,
      "language_loss": 0.82318145,
      "learning_rate": 2.3010919120114084e-06,
      "loss": 0.8453511,
      "num_input_tokens_seen": 83800750,
      "step": 3895,
      "time_per_iteration": 2.6082825660705566
    },
    {
      "auxiliary_loss_clip": 0.01185496,
      "auxiliary_loss_mlp": 0.01026262,
      "balance_loss_clip": 1.05685759,
      "balance_loss_mlp": 1.01751184,
      "epoch": 0.4684663019298984,
      "flos": 15368330551680.0,
      "grad_norm": 2.5507941408582417,
      "language_loss": 0.65797502,
      "learning_rate": 2.3003217954764672e-06,
      "loss": 0.68009269,
      "num_input_tokens_seen": 83815455,
      "step": 3896,
      "time_per_iteration": 2.6137373447418213
    },
    {
      "auxiliary_loss_clip": 0.01185995,
      "auxiliary_loss_mlp": 0.0103151,
      "balance_loss_clip": 1.05757928,
      "balance_loss_mlp": 1.0229094,
      "epoch": 0.46858654482053747,
      "flos": 27778842737280.0,
      "grad_norm": 2.9061206994059874,
      "language_loss": 0.79632002,
      "learning_rate": 2.299551633382949e-06,
      "loss": 0.81849509,
      "num_input_tokens_seen": 83835765,
      "step": 3897,
      "time_per_iteration": 2.6602375507354736
    },
    {
      "auxiliary_loss_clip": 0.01184094,
      "auxiliary_loss_mlp": 0.01029754,
      "balance_loss_clip": 1.05775166,
      "balance_loss_mlp": 1.02119458,
      "epoch": 0.4687067877111766,
      "flos": 18040623707520.0,
      "grad_norm": 1.9470813515442256,
      "language_loss": 0.85632145,
      "learning_rate": 2.2987814258476854e-06,
      "loss": 0.87845993,
      "num_input_tokens_seen": 83853565,
      "step": 3898,
      "time_per_iteration": 2.658123254776001
    },
    {
      "auxiliary_loss_clip": 0.01186736,
      "auxiliary_loss_mlp": 0.01027718,
      "balance_loss_clip": 1.05578589,
      "balance_loss_mlp": 1.01893198,
      "epoch": 0.4688270306018157,
      "flos": 16977380198400.0,
      "grad_norm": 27.635008325284197,
      "language_loss": 0.67595536,
      "learning_rate": 2.2980111729875177e-06,
      "loss": 0.69809991,
      "num_input_tokens_seen": 83869815,
      "step": 3899,
      "time_per_iteration": 2.588899612426758
    },
    {
      "auxiliary_loss_clip": 0.0118682,
      "auxiliary_loss_mlp": 0.01028142,
      "balance_loss_clip": 1.05900681,
      "balance_loss_mlp": 1.01979792,
      "epoch": 0.46894727349245474,
      "flos": 17821640442240.0,
      "grad_norm": 3.132200522878495,
      "language_loss": 0.82778496,
      "learning_rate": 2.2972408749192917e-06,
      "loss": 0.84993452,
      "num_input_tokens_seen": 83887545,
      "step": 3900,
      "time_per_iteration": 2.6768553256988525
    },
    {
      "auxiliary_loss_clip": 0.01182785,
      "auxiliary_loss_mlp": 0.01027801,
      "balance_loss_clip": 1.05648506,
      "balance_loss_mlp": 1.01980782,
      "epoch": 0.46906751638309385,
      "flos": 21471349559040.0,
      "grad_norm": 2.003467816540006,
      "language_loss": 0.66901147,
      "learning_rate": 2.296470531759861e-06,
      "loss": 0.69111729,
      "num_input_tokens_seen": 83905645,
      "step": 3901,
      "time_per_iteration": 2.726855754852295
    },
    {
      "auxiliary_loss_clip": 0.01186355,
      "auxiliary_loss_mlp": 0.01023437,
      "balance_loss_clip": 1.05703437,
      "balance_loss_mlp": 1.01475286,
      "epoch": 0.46918775927373296,
      "flos": 20337829090560.0,
      "grad_norm": 3.398965756981119,
      "language_loss": 0.79774493,
      "learning_rate": 2.2957001436260866e-06,
      "loss": 0.81984282,
      "num_input_tokens_seen": 83922705,
      "step": 3902,
      "time_per_iteration": 2.681013822555542
    },
    {
      "auxiliary_loss_clip": 0.01186624,
      "auxiliary_loss_mlp": 0.01030142,
      "balance_loss_clip": 1.05944312,
      "balance_loss_mlp": 1.02164853,
      "epoch": 0.469308002164372,
      "flos": 18403249461120.0,
      "grad_norm": 2.313536078913106,
      "language_loss": 0.73100209,
      "learning_rate": 2.294929710634836e-06,
      "loss": 0.75316978,
      "num_input_tokens_seen": 83940795,
      "step": 3903,
      "time_per_iteration": 2.6411118507385254
    },
    {
      "auxiliary_loss_clip": 0.01183597,
      "auxiliary_loss_mlp": 0.01026191,
      "balance_loss_clip": 1.05551004,
      "balance_loss_mlp": 1.01774466,
      "epoch": 0.46942824505501113,
      "flos": 37962067363200.0,
      "grad_norm": 2.4106901861193095,
      "language_loss": 0.61172283,
      "learning_rate": 2.2941592329029823e-06,
      "loss": 0.63382077,
      "num_input_tokens_seen": 83961900,
      "step": 3904,
      "time_per_iteration": 2.7562646865844727
    },
    {
      "auxiliary_loss_clip": 0.011885,
      "auxiliary_loss_mlp": 0.01032977,
      "balance_loss_clip": 1.06019402,
      "balance_loss_mlp": 1.0241679,
      "epoch": 0.46954848794565024,
      "flos": 21872507627520.0,
      "grad_norm": 1.7504614778972332,
      "language_loss": 0.79249156,
      "learning_rate": 2.2933887105474067e-06,
      "loss": 0.81470633,
      "num_input_tokens_seen": 83980075,
      "step": 3905,
      "time_per_iteration": 2.644881248474121
    },
    {
      "auxiliary_loss_clip": 0.01182502,
      "auxiliary_loss_mlp": 0.01023513,
      "balance_loss_clip": 1.05781817,
      "balance_loss_mlp": 1.01596081,
      "epoch": 0.4696687308362893,
      "flos": 22016545165440.0,
      "grad_norm": 1.9368226027855522,
      "language_loss": 0.8130489,
      "learning_rate": 2.2926181436849974e-06,
      "loss": 0.83510906,
      "num_input_tokens_seen": 83999430,
      "step": 3906,
      "time_per_iteration": 2.6672046184539795
    },
    {
      "auxiliary_loss_clip": 0.01185061,
      "auxiliary_loss_mlp": 0.01029331,
      "balance_loss_clip": 1.05790973,
      "balance_loss_mlp": 1.020998,
      "epoch": 0.4697889737269284,
      "flos": 21613663244160.0,
      "grad_norm": 1.993449287392347,
      "language_loss": 0.7259109,
      "learning_rate": 2.2918475324326478e-06,
      "loss": 0.74805486,
      "num_input_tokens_seen": 84019150,
      "step": 3907,
      "time_per_iteration": 2.687225818634033
    },
    {
      "auxiliary_loss_clip": 0.01189078,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.05847871,
      "balance_loss_mlp": 1.01847672,
      "epoch": 0.46990921661756746,
      "flos": 25228323665280.0,
      "grad_norm": 5.324409400552014,
      "language_loss": 0.91031408,
      "learning_rate": 2.2910768769072603e-06,
      "loss": 0.93247509,
      "num_input_tokens_seen": 84037930,
      "step": 3908,
      "time_per_iteration": 2.703235626220703
    },
    {
      "auxiliary_loss_clip": 0.0118582,
      "auxiliary_loss_mlp": 0.01026545,
      "balance_loss_clip": 1.05906129,
      "balance_loss_mlp": 1.0179379,
      "epoch": 0.47002945950820657,
      "flos": 13844031045120.0,
      "grad_norm": 2.749319757035466,
      "language_loss": 0.76060975,
      "learning_rate": 2.2903061772257417e-06,
      "loss": 0.78273332,
      "num_input_tokens_seen": 84055915,
      "step": 3909,
      "time_per_iteration": 2.635817289352417
    },
    {
      "auxiliary_loss_clip": 0.01184773,
      "auxiliary_loss_mlp": 0.01023525,
      "balance_loss_clip": 1.05711317,
      "balance_loss_mlp": 1.01518631,
      "epoch": 0.4701497023988457,
      "flos": 26247001374720.0,
      "grad_norm": 1.5148716937603992,
      "language_loss": 0.78978056,
      "learning_rate": 2.289535433505007e-06,
      "loss": 0.81186354,
      "num_input_tokens_seen": 84077270,
      "step": 3910,
      "time_per_iteration": 3.5844063758850098
    },
    {
      "auxiliary_loss_clip": 0.01186692,
      "auxiliary_loss_mlp": 0.01021831,
      "balance_loss_clip": 1.05921125,
      "balance_loss_mlp": 1.01327229,
      "epoch": 0.47026994528948474,
      "flos": 25629517647360.0,
      "grad_norm": 1.9997457212818461,
      "language_loss": 0.64023244,
      "learning_rate": 2.2887646458619767e-06,
      "loss": 0.66231769,
      "num_input_tokens_seen": 84098635,
      "step": 3911,
      "time_per_iteration": 2.6955785751342773
    },
    {
      "auxiliary_loss_clip": 0.01188732,
      "auxiliary_loss_mlp": 0.01032307,
      "balance_loss_clip": 1.05906904,
      "balance_loss_mlp": 1.02339017,
      "epoch": 0.47039018818012385,
      "flos": 20554406144640.0,
      "grad_norm": 1.946158130436256,
      "language_loss": 0.76600975,
      "learning_rate": 2.2879938144135797e-06,
      "loss": 0.78822017,
      "num_input_tokens_seen": 84114740,
      "step": 3912,
      "time_per_iteration": 2.633754253387451
    },
    {
      "auxiliary_loss_clip": 0.01186233,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.05886078,
      "balance_loss_mlp": 1.0171051,
      "epoch": 0.47051043107076296,
      "flos": 21577249831680.0,
      "grad_norm": 2.5312212722463294,
      "language_loss": 0.74740285,
      "learning_rate": 2.2872229392767496e-06,
      "loss": 0.76951754,
      "num_input_tokens_seen": 84134845,
      "step": 3913,
      "time_per_iteration": 2.666649103164673
    },
    {
      "auxiliary_loss_clip": 0.01187496,
      "auxiliary_loss_mlp": 0.01024137,
      "balance_loss_clip": 1.05883622,
      "balance_loss_mlp": 1.01619744,
      "epoch": 0.470630673961402,
      "flos": 18953185662720.0,
      "grad_norm": 1.800754954253852,
      "language_loss": 0.74637765,
      "learning_rate": 2.286452020568428e-06,
      "loss": 0.76849395,
      "num_input_tokens_seen": 84152920,
      "step": 3914,
      "time_per_iteration": 3.532149076461792
    },
    {
      "auxiliary_loss_clip": 0.01187881,
      "auxiliary_loss_mlp": 0.01032713,
      "balance_loss_clip": 1.05655169,
      "balance_loss_mlp": 1.02312839,
      "epoch": 0.4707509168520411,
      "flos": 19938969492480.0,
      "grad_norm": 1.9213885653815534,
      "language_loss": 0.73405766,
      "learning_rate": 2.2856810584055637e-06,
      "loss": 0.75626361,
      "num_input_tokens_seen": 84170455,
      "step": 3915,
      "time_per_iteration": 2.665111541748047
    },
    {
      "auxiliary_loss_clip": 0.01182685,
      "auxiliary_loss_mlp": 0.0102338,
      "balance_loss_clip": 1.05572736,
      "balance_loss_mlp": 1.01497602,
      "epoch": 0.47087115974268023,
      "flos": 40118754741120.0,
      "grad_norm": 3.7815172537240658,
      "language_loss": 0.67864996,
      "learning_rate": 2.2849100529051085e-06,
      "loss": 0.70071065,
      "num_input_tokens_seen": 84197390,
      "step": 3916,
      "time_per_iteration": 3.7276105880737305
    },
    {
      "auxiliary_loss_clip": 0.01181894,
      "auxiliary_loss_mlp": 0.01027653,
      "balance_loss_clip": 1.05596304,
      "balance_loss_mlp": 1.0199697,
      "epoch": 0.4709914026333193,
      "flos": 13552723745280.0,
      "grad_norm": 2.7787760083522293,
      "language_loss": 0.79871482,
      "learning_rate": 2.284139004184026e-06,
      "loss": 0.82081032,
      "num_input_tokens_seen": 84214620,
      "step": 3917,
      "time_per_iteration": 3.628908634185791
    },
    {
      "auxiliary_loss_clip": 0.01184677,
      "auxiliary_loss_mlp": 0.01029129,
      "balance_loss_clip": 1.05727792,
      "balance_loss_mlp": 1.02039123,
      "epoch": 0.4711116455239584,
      "flos": 19974628719360.0,
      "grad_norm": 2.2039601833749374,
      "language_loss": 0.74132347,
      "learning_rate": 2.2833679123592814e-06,
      "loss": 0.76346147,
      "num_input_tokens_seen": 84231880,
      "step": 3918,
      "time_per_iteration": 2.610663652420044
    },
    {
      "auxiliary_loss_clip": 0.01187595,
      "auxiliary_loss_mlp": 0.01025858,
      "balance_loss_clip": 1.05838251,
      "balance_loss_mlp": 1.01721489,
      "epoch": 0.4712318884145975,
      "flos": 32124824064000.0,
      "grad_norm": 1.9017462447552127,
      "language_loss": 0.63218856,
      "learning_rate": 2.2825967775478508e-06,
      "loss": 0.6543231,
      "num_input_tokens_seen": 84252980,
      "step": 3919,
      "time_per_iteration": 2.7160539627075195
    },
    {
      "auxiliary_loss_clip": 0.01184793,
      "auxiliary_loss_mlp": 0.01030984,
      "balance_loss_clip": 1.05538332,
      "balance_loss_mlp": 1.02287185,
      "epoch": 0.47135213130523657,
      "flos": 20047850593920.0,
      "grad_norm": 2.411258131128413,
      "language_loss": 0.83592665,
      "learning_rate": 2.2818255998667135e-06,
      "loss": 0.85808432,
      "num_input_tokens_seen": 84271490,
      "step": 3920,
      "time_per_iteration": 2.6379940509796143
    },
    {
      "auxiliary_loss_clip": 0.01186794,
      "auxiliary_loss_mlp": 0.01023568,
      "balance_loss_clip": 1.05898714,
      "balance_loss_mlp": 1.01498461,
      "epoch": 0.4714723741958757,
      "flos": 19426990988160.0,
      "grad_norm": 1.6423418860390464,
      "language_loss": 0.79026842,
      "learning_rate": 2.2810543794328566e-06,
      "loss": 0.81237209,
      "num_input_tokens_seen": 84290525,
      "step": 3921,
      "time_per_iteration": 2.668510675430298
    },
    {
      "auxiliary_loss_clip": 0.01186111,
      "auxiliary_loss_mlp": 0.01028839,
      "balance_loss_clip": 1.05711508,
      "balance_loss_mlp": 1.0199039,
      "epoch": 0.4715926170865148,
      "flos": 20373883367040.0,
      "grad_norm": 1.8281504228992638,
      "language_loss": 0.82167011,
      "learning_rate": 2.2802831163632735e-06,
      "loss": 0.84381962,
      "num_input_tokens_seen": 84309245,
      "step": 3922,
      "time_per_iteration": 2.5981857776641846
    },
    {
      "auxiliary_loss_clip": 0.01186105,
      "auxiliary_loss_mlp": 0.01029544,
      "balance_loss_clip": 1.05526114,
      "balance_loss_mlp": 1.02028728,
      "epoch": 0.47171285997715384,
      "flos": 22672884430080.0,
      "grad_norm": 3.9860055965415078,
      "language_loss": 0.74577874,
      "learning_rate": 2.279511810774965e-06,
      "loss": 0.76793528,
      "num_input_tokens_seen": 84330775,
      "step": 3923,
      "time_per_iteration": 2.6797964572906494
    },
    {
      "auxiliary_loss_clip": 0.01187982,
      "auxiliary_loss_mlp": 0.01025527,
      "balance_loss_clip": 1.05725634,
      "balance_loss_mlp": 1.01672363,
      "epoch": 0.47183310286779295,
      "flos": 21105419754240.0,
      "grad_norm": 2.491539404529969,
      "language_loss": 0.71391445,
      "learning_rate": 2.2787404627849364e-06,
      "loss": 0.73604953,
      "num_input_tokens_seen": 84349985,
      "step": 3924,
      "time_per_iteration": 2.6417903900146484
    },
    {
      "auxiliary_loss_clip": 0.01185631,
      "auxiliary_loss_mlp": 0.01023868,
      "balance_loss_clip": 1.05704212,
      "balance_loss_mlp": 1.0155586,
      "epoch": 0.471953345758432,
      "flos": 21726566668800.0,
      "grad_norm": 1.943890587312507,
      "language_loss": 0.79024291,
      "learning_rate": 2.277969072510202e-06,
      "loss": 0.81233788,
      "num_input_tokens_seen": 84368965,
      "step": 3925,
      "time_per_iteration": 2.611760377883911
    },
    {
      "auxiliary_loss_clip": 0.01185852,
      "auxiliary_loss_mlp": 0.010302,
      "balance_loss_clip": 1.05710602,
      "balance_loss_mlp": 1.02231169,
      "epoch": 0.4720735886490711,
      "flos": 19861078849920.0,
      "grad_norm": 1.8686867994971503,
      "language_loss": 0.81564313,
      "learning_rate": 2.2771976400677803e-06,
      "loss": 0.83780366,
      "num_input_tokens_seen": 84387795,
      "step": 3926,
      "time_per_iteration": 2.6052913665771484
    },
    {
      "auxiliary_loss_clip": 0.01183601,
      "auxiliary_loss_mlp": 0.01030022,
      "balance_loss_clip": 1.05701804,
      "balance_loss_mlp": 1.02130771,
      "epoch": 0.47219383153971023,
      "flos": 19171809792000.0,
      "grad_norm": 2.03150670724495,
      "language_loss": 0.78954113,
      "learning_rate": 2.2764261655746965e-06,
      "loss": 0.81167734,
      "num_input_tokens_seen": 84405290,
      "step": 3927,
      "time_per_iteration": 2.6517419815063477
    },
    {
      "auxiliary_loss_clip": 0.01188405,
      "auxiliary_loss_mlp": 0.01029948,
      "balance_loss_clip": 1.05841768,
      "balance_loss_mlp": 1.02121031,
      "epoch": 0.4723140744303493,
      "flos": 23224005780480.0,
      "grad_norm": 2.302342761342159,
      "language_loss": 0.75904405,
      "learning_rate": 2.2756546491479832e-06,
      "loss": 0.78122753,
      "num_input_tokens_seen": 84426205,
      "step": 3928,
      "time_per_iteration": 2.6265108585357666
    },
    {
      "auxiliary_loss_clip": 0.01186606,
      "auxiliary_loss_mlp": 0.01026755,
      "balance_loss_clip": 1.05550981,
      "balance_loss_mlp": 1.0176475,
      "epoch": 0.4724343173209884,
      "flos": 18223265387520.0,
      "grad_norm": 4.281724468392857,
      "language_loss": 0.80381858,
      "learning_rate": 2.274883090904679e-06,
      "loss": 0.82595217,
      "num_input_tokens_seen": 84443970,
      "step": 3929,
      "time_per_iteration": 2.650343894958496
    },
    {
      "auxiliary_loss_clip": 0.01191972,
      "auxiliary_loss_mlp": 0.01028334,
      "balance_loss_clip": 1.06041527,
      "balance_loss_mlp": 1.01995933,
      "epoch": 0.4725545602116275,
      "flos": 21251037490560.0,
      "grad_norm": 2.5116122642909318,
      "language_loss": 0.67554218,
      "learning_rate": 2.2741114909618283e-06,
      "loss": 0.6977452,
      "num_input_tokens_seen": 84459865,
      "step": 3930,
      "time_per_iteration": 2.618145227432251
    },
    {
      "auxiliary_loss_clip": 0.01184676,
      "auxiliary_loss_mlp": 0.01027441,
      "balance_loss_clip": 1.05645585,
      "balance_loss_mlp": 1.0191263,
      "epoch": 0.47267480310226656,
      "flos": 21434002392960.0,
      "grad_norm": 3.0276957989501283,
      "language_loss": 0.7184757,
      "learning_rate": 2.2733398494364828e-06,
      "loss": 0.74059689,
      "num_input_tokens_seen": 84479110,
      "step": 3931,
      "time_per_iteration": 2.613779306411743
    },
    {
      "auxiliary_loss_clip": 0.01186988,
      "auxiliary_loss_mlp": 0.01031317,
      "balance_loss_clip": 1.05759621,
      "balance_loss_mlp": 1.02276945,
      "epoch": 0.47279504599290567,
      "flos": 18770508069120.0,
      "grad_norm": 2.0976580926273245,
      "language_loss": 0.84228742,
      "learning_rate": 2.272568166445699e-06,
      "loss": 0.86447048,
      "num_input_tokens_seen": 84497675,
      "step": 3932,
      "time_per_iteration": 2.61492919921875
    },
    {
      "auxiliary_loss_clip": 0.01184828,
      "auxiliary_loss_mlp": 0.01031751,
      "balance_loss_clip": 1.05624199,
      "balance_loss_mlp": 1.02323985,
      "epoch": 0.4729152888835448,
      "flos": 21105742976640.0,
      "grad_norm": 1.9470607968402764,
      "language_loss": 0.64440393,
      "learning_rate": 2.271796442106541e-06,
      "loss": 0.66656971,
      "num_input_tokens_seen": 84517030,
      "step": 3933,
      "time_per_iteration": 2.650425672531128
    },
    {
      "auxiliary_loss_clip": 0.01083402,
      "auxiliary_loss_mlp": 0.01005712,
      "balance_loss_clip": 1.01959348,
      "balance_loss_mlp": 1.00405526,
      "epoch": 0.47303553177418384,
      "flos": 70201877840640.0,
      "grad_norm": 0.8662542714446778,
      "language_loss": 0.56518626,
      "learning_rate": 2.271024676536079e-06,
      "loss": 0.58607745,
      "num_input_tokens_seen": 84577290,
      "step": 3934,
      "time_per_iteration": 3.156177043914795
    },
    {
      "auxiliary_loss_clip": 0.01190692,
      "auxiliary_loss_mlp": 0.01029202,
      "balance_loss_clip": 1.05948997,
      "balance_loss_mlp": 1.01928997,
      "epoch": 0.47315577466482295,
      "flos": 22455122227200.0,
      "grad_norm": 2.1423217610582075,
      "language_loss": 0.7333883,
      "learning_rate": 2.2702528698513894e-06,
      "loss": 0.75558728,
      "num_input_tokens_seen": 84598415,
      "step": 3935,
      "time_per_iteration": 2.714967966079712
    },
    {
      "auxiliary_loss_clip": 0.01190023,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.0579499,
      "balance_loss_mlp": 1.01835334,
      "epoch": 0.47327601755546206,
      "flos": 24352857480960.0,
      "grad_norm": 3.281297367983088,
      "language_loss": 0.78614646,
      "learning_rate": 2.269481022169554e-06,
      "loss": 0.80831993,
      "num_input_tokens_seen": 84617010,
      "step": 3936,
      "time_per_iteration": 3.6083803176879883
    },
    {
      "auxiliary_loss_clip": 0.01187819,
      "auxiliary_loss_mlp": 0.01028978,
      "balance_loss_clip": 1.05567431,
      "balance_loss_mlp": 1.01946557,
      "epoch": 0.4733962604461011,
      "flos": 22926772736640.0,
      "grad_norm": 2.8665061992090117,
      "language_loss": 0.80278277,
      "learning_rate": 2.2687091336076614e-06,
      "loss": 0.82495081,
      "num_input_tokens_seen": 84636350,
      "step": 3937,
      "time_per_iteration": 2.656090497970581
    },
    {
      "auxiliary_loss_clip": 0.01189563,
      "auxiliary_loss_mlp": 0.01035071,
      "balance_loss_clip": 1.05921483,
      "balance_loss_mlp": 1.02507544,
      "epoch": 0.4735165033367402,
      "flos": 18327369980160.0,
      "grad_norm": 2.63311873101283,
      "language_loss": 0.79697102,
      "learning_rate": 2.267937204282807e-06,
      "loss": 0.81921744,
      "num_input_tokens_seen": 84653490,
      "step": 3938,
      "time_per_iteration": 2.61560320854187
    },
    {
      "auxiliary_loss_clip": 0.01194058,
      "auxiliary_loss_mlp": 0.01036612,
      "balance_loss_clip": 1.0601573,
      "balance_loss_mlp": 1.0258292,
      "epoch": 0.4736367462273793,
      "flos": 23037018554880.0,
      "grad_norm": 3.0154262042567033,
      "language_loss": 0.78567457,
      "learning_rate": 2.2671652343120926e-06,
      "loss": 0.80798125,
      "num_input_tokens_seen": 84673965,
      "step": 3939,
      "time_per_iteration": 2.6818623542785645
    },
    {
      "auxiliary_loss_clip": 0.01186963,
      "auxiliary_loss_mlp": 0.01027317,
      "balance_loss_clip": 1.05861545,
      "balance_loss_mlp": 1.01845956,
      "epoch": 0.4737569891180184,
      "flos": 25374336451200.0,
      "grad_norm": 1.9198869213378624,
      "language_loss": 0.80384159,
      "learning_rate": 2.2663932238126236e-06,
      "loss": 0.82598442,
      "num_input_tokens_seen": 84692525,
      "step": 3940,
      "time_per_iteration": 3.6073551177978516
    },
    {
      "auxiliary_loss_clip": 0.0118915,
      "auxiliary_loss_mlp": 0.01025029,
      "balance_loss_clip": 1.0580734,
      "balance_loss_mlp": 1.01636815,
      "epoch": 0.4738772320086575,
      "flos": 25849326925440.0,
      "grad_norm": 4.024388237919752,
      "language_loss": 0.80267429,
      "learning_rate": 2.265621172901515e-06,
      "loss": 0.82481611,
      "num_input_tokens_seen": 84715640,
      "step": 3941,
      "time_per_iteration": 2.730938673019409
    },
    {
      "auxiliary_loss_clip": 0.01196554,
      "auxiliary_loss_mlp": 0.01034028,
      "balance_loss_clip": 1.06464076,
      "balance_loss_mlp": 1.02537966,
      "epoch": 0.47399747489929656,
      "flos": 27564420499200.0,
      "grad_norm": 3.443416833561979,
      "language_loss": 0.71475452,
      "learning_rate": 2.2648490816958854e-06,
      "loss": 0.73706031,
      "num_input_tokens_seen": 84736635,
      "step": 3942,
      "time_per_iteration": 2.6998445987701416
    },
    {
      "auxiliary_loss_clip": 0.0119213,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.05771565,
      "balance_loss_mlp": 1.02138448,
      "epoch": 0.47411771778993567,
      "flos": 24863650836480.0,
      "grad_norm": 2.8317387615602696,
      "language_loss": 0.72937566,
      "learning_rate": 2.264076950312861e-06,
      "loss": 0.75160253,
      "num_input_tokens_seen": 84755445,
      "step": 3943,
      "time_per_iteration": 3.6060688495635986
    },
    {
      "auxiliary_loss_clip": 0.01190652,
      "auxiliary_loss_mlp": 0.01026163,
      "balance_loss_clip": 1.05821264,
      "balance_loss_mlp": 1.01675093,
      "epoch": 0.4742379606805748,
      "flos": 22748009725440.0,
      "grad_norm": 2.4186380829760585,
      "language_loss": 0.82800746,
      "learning_rate": 2.2633047788695727e-06,
      "loss": 0.85017556,
      "num_input_tokens_seen": 84775750,
      "step": 3944,
      "time_per_iteration": 3.645123243331909
    },
    {
      "auxiliary_loss_clip": 0.01188036,
      "auxiliary_loss_mlp": 0.01022456,
      "balance_loss_clip": 1.05947006,
      "balance_loss_mlp": 1.01411104,
      "epoch": 0.47435820357121383,
      "flos": 19681130689920.0,
      "grad_norm": 2.4715257586477652,
      "language_loss": 0.64040112,
      "learning_rate": 2.262532567483159e-06,
      "loss": 0.66250604,
      "num_input_tokens_seen": 84794310,
      "step": 3945,
      "time_per_iteration": 2.7052242755889893
    },
    {
      "auxiliary_loss_clip": 0.01194408,
      "auxiliary_loss_mlp": 0.01030154,
      "balance_loss_clip": 1.06224453,
      "balance_loss_mlp": 1.02063489,
      "epoch": 0.47447844646185294,
      "flos": 25228718714880.0,
      "grad_norm": 2.3631703690778045,
      "language_loss": 0.80106002,
      "learning_rate": 2.2617603162707635e-06,
      "loss": 0.82330561,
      "num_input_tokens_seen": 84814720,
      "step": 3946,
      "time_per_iteration": 2.72707462310791
    },
    {
      "auxiliary_loss_clip": 0.01187565,
      "auxiliary_loss_mlp": 0.01026702,
      "balance_loss_clip": 1.05775023,
      "balance_loss_mlp": 1.01701069,
      "epoch": 0.47459868935249205,
      "flos": 24570619683840.0,
      "grad_norm": 1.809562440232819,
      "language_loss": 0.82276434,
      "learning_rate": 2.2609880253495363e-06,
      "loss": 0.84490705,
      "num_input_tokens_seen": 84834355,
      "step": 3947,
      "time_per_iteration": 2.6869428157806396
    },
    {
      "auxiliary_loss_clip": 0.01193342,
      "auxiliary_loss_mlp": 0.01029627,
      "balance_loss_clip": 1.06084454,
      "balance_loss_mlp": 1.02041173,
      "epoch": 0.4747189322431311,
      "flos": 20558500295040.0,
      "grad_norm": 2.0306798588544583,
      "language_loss": 0.86695325,
      "learning_rate": 2.260215694836633e-06,
      "loss": 0.88918293,
      "num_input_tokens_seen": 84853530,
      "step": 3948,
      "time_per_iteration": 2.6473584175109863
    },
    {
      "auxiliary_loss_clip": 0.01189118,
      "auxiliary_loss_mlp": 0.01029866,
      "balance_loss_clip": 1.05914295,
      "balance_loss_mlp": 1.02033508,
      "epoch": 0.4748391751337702,
      "flos": 25995231970560.0,
      "grad_norm": 3.3242297123640023,
      "language_loss": 0.65102816,
      "learning_rate": 2.2594433248492157e-06,
      "loss": 0.67321801,
      "num_input_tokens_seen": 84872505,
      "step": 3949,
      "time_per_iteration": 2.77193284034729
    },
    {
      "auxiliary_loss_clip": 0.01193642,
      "auxiliary_loss_mlp": 0.01027781,
      "balance_loss_clip": 1.06059062,
      "balance_loss_mlp": 1.01876915,
      "epoch": 0.47495941802440933,
      "flos": 22821052032000.0,
      "grad_norm": 2.056566799434504,
      "language_loss": 0.80138695,
      "learning_rate": 2.2586709155044527e-06,
      "loss": 0.82360113,
      "num_input_tokens_seen": 84893105,
      "step": 3950,
      "time_per_iteration": 2.6628775596618652
    },
    {
      "auxiliary_loss_clip": 0.01191446,
      "auxiliary_loss_mlp": 0.01028132,
      "balance_loss_clip": 1.06086218,
      "balance_loss_mlp": 1.01885772,
      "epoch": 0.4750796609150484,
      "flos": 27891782075520.0,
      "grad_norm": 2.018597387438815,
      "language_loss": 0.7585091,
      "learning_rate": 2.2578984669195167e-06,
      "loss": 0.78070486,
      "num_input_tokens_seen": 84914070,
      "step": 3951,
      "time_per_iteration": 2.7200589179992676
    },
    {
      "auxiliary_loss_clip": 0.01184882,
      "auxiliary_loss_mlp": 0.01028093,
      "balance_loss_clip": 1.05667043,
      "balance_loss_mlp": 1.01985002,
      "epoch": 0.4751999038056875,
      "flos": 35660085471360.0,
      "grad_norm": 2.49765127466361,
      "language_loss": 0.6789639,
      "learning_rate": 2.2571259792115887e-06,
      "loss": 0.70109367,
      "num_input_tokens_seen": 84935290,
      "step": 3952,
      "time_per_iteration": 2.7285945415496826
    },
    {
      "auxiliary_loss_clip": 0.01186753,
      "auxiliary_loss_mlp": 0.01026681,
      "balance_loss_clip": 1.05809152,
      "balance_loss_mlp": 1.01821077,
      "epoch": 0.4753201466963266,
      "flos": 22090880361600.0,
      "grad_norm": 1.7905546499984255,
      "language_loss": 0.79037905,
      "learning_rate": 2.2563534524978544e-06,
      "loss": 0.81251335,
      "num_input_tokens_seen": 84952760,
      "step": 3953,
      "time_per_iteration": 2.645125150680542
    },
    {
      "auxiliary_loss_clip": 0.01187524,
      "auxiliary_loss_mlp": 0.01026521,
      "balance_loss_clip": 1.05913591,
      "balance_loss_mlp": 1.01829565,
      "epoch": 0.47544038958696566,
      "flos": 30190854965760.0,
      "grad_norm": 2.045322164481933,
      "language_loss": 0.70363438,
      "learning_rate": 2.2555808868955052e-06,
      "loss": 0.72577477,
      "num_input_tokens_seen": 84974890,
      "step": 3954,
      "time_per_iteration": 2.6836750507354736
    },
    {
      "auxiliary_loss_clip": 0.01190982,
      "auxiliary_loss_mlp": 0.01035044,
      "balance_loss_clip": 1.0576272,
      "balance_loss_mlp": 1.02509046,
      "epoch": 0.47556063247760477,
      "flos": 23472219738240.0,
      "grad_norm": 2.7191373335216147,
      "language_loss": 0.74120867,
      "learning_rate": 2.254808282521738e-06,
      "loss": 0.76346892,
      "num_input_tokens_seen": 84993640,
      "step": 3955,
      "time_per_iteration": 2.637258529663086
    },
    {
      "auxiliary_loss_clip": 0.01186865,
      "auxiliary_loss_mlp": 0.01029079,
      "balance_loss_clip": 1.05744386,
      "balance_loss_mlp": 1.01965523,
      "epoch": 0.4756808753682438,
      "flos": 25155209531520.0,
      "grad_norm": 4.525834442067659,
      "language_loss": 0.81155121,
      "learning_rate": 2.2540356394937573e-06,
      "loss": 0.83371067,
      "num_input_tokens_seen": 85012340,
      "step": 3956,
      "time_per_iteration": 2.656137228012085
    },
    {
      "auxiliary_loss_clip": 0.01191831,
      "auxiliary_loss_mlp": 0.01033629,
      "balance_loss_clip": 1.05866635,
      "balance_loss_mlp": 1.0242238,
      "epoch": 0.47580111825888294,
      "flos": 15669729573120.0,
      "grad_norm": 3.0184687109309065,
      "language_loss": 0.84122944,
      "learning_rate": 2.253262957928772e-06,
      "loss": 0.86348403,
      "num_input_tokens_seen": 85029225,
      "step": 3957,
      "time_per_iteration": 2.6732804775238037
    },
    {
      "auxiliary_loss_clip": 0.01188472,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.05713844,
      "balance_loss_mlp": 1.02056789,
      "epoch": 0.47592136114952205,
      "flos": 17636556637440.0,
      "grad_norm": 1.9647490025601158,
      "language_loss": 0.721264,
      "learning_rate": 2.2524902379439976e-06,
      "loss": 0.74344623,
      "num_input_tokens_seen": 85047895,
      "step": 3958,
      "time_per_iteration": 2.5661988258361816
    },
    {
      "auxiliary_loss_clip": 0.01084957,
      "auxiliary_loss_mlp": 0.01001815,
      "balance_loss_clip": 1.0206753,
      "balance_loss_mlp": 1.00013375,
      "epoch": 0.4760416040401611,
      "flos": 61417159292160.0,
      "grad_norm": 0.8425623322859146,
      "language_loss": 0.63667774,
      "learning_rate": 2.251717479656655e-06,
      "loss": 0.65754551,
      "num_input_tokens_seen": 85112690,
      "step": 3959,
      "time_per_iteration": 3.235837697982788
    },
    {
      "auxiliary_loss_clip": 0.01191883,
      "auxiliary_loss_mlp": 0.01035396,
      "balance_loss_clip": 1.05960011,
      "balance_loss_mlp": 1.02569842,
      "epoch": 0.4761618469308002,
      "flos": 18405871153920.0,
      "grad_norm": 2.018930635622746,
      "language_loss": 0.76354051,
      "learning_rate": 2.2509446831839704e-06,
      "loss": 0.78581333,
      "num_input_tokens_seen": 85132130,
      "step": 3960,
      "time_per_iteration": 2.7827553749084473
    },
    {
      "auxiliary_loss_clip": 0.01190381,
      "auxiliary_loss_mlp": 0.01030243,
      "balance_loss_clip": 1.05834699,
      "balance_loss_mlp": 1.02156484,
      "epoch": 0.4762820898214393,
      "flos": 18040911016320.0,
      "grad_norm": 2.560476751875917,
      "language_loss": 0.82371294,
      "learning_rate": 2.250171848643177e-06,
      "loss": 0.84591925,
      "num_input_tokens_seen": 85149420,
      "step": 3961,
      "time_per_iteration": 2.6397573947906494
    },
    {
      "auxiliary_loss_clip": 0.01186943,
      "auxiliary_loss_mlp": 0.01026415,
      "balance_loss_clip": 1.05853486,
      "balance_loss_mlp": 1.01770139,
      "epoch": 0.4764023327120784,
      "flos": 19318253541120.0,
      "grad_norm": 2.365076115138586,
      "language_loss": 0.85795808,
      "learning_rate": 2.249398976151513e-06,
      "loss": 0.88009167,
      "num_input_tokens_seen": 85166970,
      "step": 3962,
      "time_per_iteration": 2.691297769546509
    },
    {
      "auxiliary_loss_clip": 0.01186275,
      "auxiliary_loss_mlp": 0.01029383,
      "balance_loss_clip": 1.05681837,
      "balance_loss_mlp": 1.02041221,
      "epoch": 0.4765225756027175,
      "flos": 22747255539840.0,
      "grad_norm": 2.803812557308349,
      "language_loss": 0.78315246,
      "learning_rate": 2.248626065826223e-06,
      "loss": 0.80530906,
      "num_input_tokens_seen": 85185175,
      "step": 3963,
      "time_per_iteration": 3.5597305297851562
    },
    {
      "auxiliary_loss_clip": 0.01085228,
      "auxiliary_loss_mlp": 0.01000655,
      "balance_loss_clip": 1.02098775,
      "balance_loss_mlp": 0.99914151,
      "epoch": 0.4766428184933566,
      "flos": 65933392106880.0,
      "grad_norm": 0.7588808625244641,
      "language_loss": 0.62501788,
      "learning_rate": 2.2478531177845564e-06,
      "loss": 0.64587671,
      "num_input_tokens_seen": 85246170,
      "step": 3964,
      "time_per_iteration": 3.200334310531616
    },
    {
      "auxiliary_loss_clip": 0.01190973,
      "auxiliary_loss_mlp": 0.01023956,
      "balance_loss_clip": 1.0601542,
      "balance_loss_mlp": 1.01521778,
      "epoch": 0.47676306138399566,
      "flos": 24136495908480.0,
      "grad_norm": 3.538147705445958,
      "language_loss": 0.8491528,
      "learning_rate": 2.247080132143769e-06,
      "loss": 0.87130201,
      "num_input_tokens_seen": 85268525,
      "step": 3965,
      "time_per_iteration": 2.664539337158203
    },
    {
      "auxiliary_loss_clip": 0.01188088,
      "auxiliary_loss_mlp": 0.01025006,
      "balance_loss_clip": 1.0558964,
      "balance_loss_mlp": 1.01559424,
      "epoch": 0.47688330427463477,
      "flos": 12604322995200.0,
      "grad_norm": 3.3318146744566586,
      "language_loss": 0.68736994,
      "learning_rate": 2.246307109021121e-06,
      "loss": 0.70950091,
      "num_input_tokens_seen": 85285930,
      "step": 3966,
      "time_per_iteration": 2.617917776107788
    },
    {
      "auxiliary_loss_clip": 0.01186198,
      "auxiliary_loss_mlp": 0.01028429,
      "balance_loss_clip": 1.05744827,
      "balance_loss_mlp": 1.01974416,
      "epoch": 0.4770035471652739,
      "flos": 21390585828480.0,
      "grad_norm": 2.0580748711708443,
      "language_loss": 0.82337648,
      "learning_rate": 2.2455340485338817e-06,
      "loss": 0.84552276,
      "num_input_tokens_seen": 85303565,
      "step": 3967,
      "time_per_iteration": 3.5754032135009766
    },
    {
      "auxiliary_loss_clip": 0.01186355,
      "auxiliary_loss_mlp": 0.0102718,
      "balance_loss_clip": 1.05698013,
      "balance_loss_mlp": 1.01858497,
      "epoch": 0.47712379005591293,
      "flos": 25156251025920.0,
      "grad_norm": 1.7752792149038015,
      "language_loss": 0.67707455,
      "learning_rate": 2.244760950799322e-06,
      "loss": 0.69920993,
      "num_input_tokens_seen": 85321835,
      "step": 3968,
      "time_per_iteration": 2.6854090690612793
    },
    {
      "auxiliary_loss_clip": 0.01185682,
      "auxiliary_loss_mlp": 0.01030165,
      "balance_loss_clip": 1.05889273,
      "balance_loss_mlp": 1.02064025,
      "epoch": 0.47724403294655204,
      "flos": 22054323294720.0,
      "grad_norm": 2.608029709820921,
      "language_loss": 0.72318804,
      "learning_rate": 2.2439878159347203e-06,
      "loss": 0.74534649,
      "num_input_tokens_seen": 85341260,
      "step": 3969,
      "time_per_iteration": 3.5624561309814453
    },
    {
      "auxiliary_loss_clip": 0.01084934,
      "auxiliary_loss_mlp": 0.01002447,
      "balance_loss_clip": 1.0207516,
      "balance_loss_mlp": 1.00095713,
      "epoch": 0.4773642758371911,
      "flos": 70229387658240.0,
      "grad_norm": 0.7267370418195354,
      "language_loss": 0.55233908,
      "learning_rate": 2.2432146440573616e-06,
      "loss": 0.57321298,
      "num_input_tokens_seen": 85407220,
      "step": 3970,
      "time_per_iteration": 4.300107479095459
    },
    {
      "auxiliary_loss_clip": 0.01184992,
      "auxiliary_loss_mlp": 0.01028517,
      "balance_loss_clip": 1.05668283,
      "balance_loss_mlp": 1.01960564,
      "epoch": 0.4774845187278302,
      "flos": 23548602009600.0,
      "grad_norm": 1.9687922798588304,
      "language_loss": 0.66664404,
      "learning_rate": 2.242441435284534e-06,
      "loss": 0.68877912,
      "num_input_tokens_seen": 85426095,
      "step": 3971,
      "time_per_iteration": 2.6146340370178223
    },
    {
      "auxiliary_loss_clip": 0.01191955,
      "auxiliary_loss_mlp": 0.01031599,
      "balance_loss_clip": 1.05968428,
      "balance_loss_mlp": 1.02219939,
      "epoch": 0.4776047616184693,
      "flos": 23075371301760.0,
      "grad_norm": 2.2806037949427203,
      "language_loss": 0.8532151,
      "learning_rate": 2.2416681897335337e-06,
      "loss": 0.87545061,
      "num_input_tokens_seen": 85444245,
      "step": 3972,
      "time_per_iteration": 2.659275770187378
    },
    {
      "auxiliary_loss_clip": 0.0118985,
      "auxiliary_loss_mlp": 0.0103158,
      "balance_loss_clip": 1.05723548,
      "balance_loss_mlp": 1.02177489,
      "epoch": 0.4777250045091084,
      "flos": 31898119374720.0,
      "grad_norm": 2.2061772521123015,
      "language_loss": 0.67156672,
      "learning_rate": 2.240894907521661e-06,
      "loss": 0.69378102,
      "num_input_tokens_seen": 85463325,
      "step": 3973,
      "time_per_iteration": 2.729320526123047
    },
    {
      "auxiliary_loss_clip": 0.01189604,
      "auxiliary_loss_mlp": 0.01024744,
      "balance_loss_clip": 1.05922842,
      "balance_loss_mlp": 1.01620293,
      "epoch": 0.4778452473997475,
      "flos": 24278163148800.0,
      "grad_norm": 2.9425051380169496,
      "language_loss": 0.63810933,
      "learning_rate": 2.240121588766223e-06,
      "loss": 0.66025281,
      "num_input_tokens_seen": 85483375,
      "step": 3974,
      "time_per_iteration": 2.64578914642334
    },
    {
      "auxiliary_loss_clip": 0.01185979,
      "auxiliary_loss_mlp": 0.01027831,
      "balance_loss_clip": 1.05767262,
      "balance_loss_mlp": 1.02026749,
      "epoch": 0.4779654902903866,
      "flos": 31575031516800.0,
      "grad_norm": 2.1225760142190926,
      "language_loss": 0.71114391,
      "learning_rate": 2.239348233584531e-06,
      "loss": 0.73328203,
      "num_input_tokens_seen": 85504230,
      "step": 3975,
      "time_per_iteration": 2.7439990043640137
    },
    {
      "auxiliary_loss_clip": 0.01187179,
      "auxiliary_loss_mlp": 0.01027691,
      "balance_loss_clip": 1.05674815,
      "balance_loss_mlp": 1.01919127,
      "epoch": 0.47808573318102565,
      "flos": 19500428344320.0,
      "grad_norm": 1.9505223581085938,
      "language_loss": 0.80448788,
      "learning_rate": 2.2385748420939013e-06,
      "loss": 0.82663655,
      "num_input_tokens_seen": 85523425,
      "step": 3976,
      "time_per_iteration": 2.6149723529815674
    },
    {
      "auxiliary_loss_clip": 0.01189338,
      "auxiliary_loss_mlp": 0.01034698,
      "balance_loss_clip": 1.06253433,
      "balance_loss_mlp": 1.02637756,
      "epoch": 0.47820597607166476,
      "flos": 22601135013120.0,
      "grad_norm": 2.123326187183704,
      "language_loss": 0.72021252,
      "learning_rate": 2.2378014144116583e-06,
      "loss": 0.74245286,
      "num_input_tokens_seen": 85542235,
      "step": 3977,
      "time_per_iteration": 2.6372952461242676
    },
    {
      "auxiliary_loss_clip": 0.01191772,
      "auxiliary_loss_mlp": 0.01023712,
      "balance_loss_clip": 1.05974841,
      "balance_loss_mlp": 1.01494968,
      "epoch": 0.4783262189623039,
      "flos": 23003011353600.0,
      "grad_norm": 1.7771179322368813,
      "language_loss": 0.80004466,
      "learning_rate": 2.23702795065513e-06,
      "loss": 0.82219946,
      "num_input_tokens_seen": 85561815,
      "step": 3978,
      "time_per_iteration": 2.6723334789276123
    },
    {
      "auxiliary_loss_clip": 0.01083922,
      "auxiliary_loss_mlp": 0.01003819,
      "balance_loss_clip": 1.01988935,
      "balance_loss_mlp": 1.00237679,
      "epoch": 0.47844646185294293,
      "flos": 49772801226240.0,
      "grad_norm": 0.9830923920537323,
      "language_loss": 0.67447799,
      "learning_rate": 2.2362544509416493e-06,
      "loss": 0.69535542,
      "num_input_tokens_seen": 85613930,
      "step": 3979,
      "time_per_iteration": 3.094454526901245
    },
    {
      "auxiliary_loss_clip": 0.01184859,
      "auxiliary_loss_mlp": 0.01026682,
      "balance_loss_clip": 1.05655229,
      "balance_loss_mlp": 1.01800919,
      "epoch": 0.47856670474358204,
      "flos": 20229558520320.0,
      "grad_norm": 2.6713210960395415,
      "language_loss": 0.82498133,
      "learning_rate": 2.2354809153885572e-06,
      "loss": 0.84709668,
      "num_input_tokens_seen": 85631000,
      "step": 3980,
      "time_per_iteration": 2.590894937515259
    },
    {
      "auxiliary_loss_clip": 0.01187117,
      "auxiliary_loss_mlp": 0.01028852,
      "balance_loss_clip": 1.05626535,
      "balance_loss_mlp": 1.02041841,
      "epoch": 0.47868694763422115,
      "flos": 20990936131200.0,
      "grad_norm": 2.146860178733411,
      "language_loss": 0.83006138,
      "learning_rate": 2.234707344113197e-06,
      "loss": 0.85222107,
      "num_input_tokens_seen": 85649095,
      "step": 3981,
      "time_per_iteration": 2.711381673812866
    },
    {
      "auxiliary_loss_clip": 0.01185458,
      "auxiliary_loss_mlp": 0.01024672,
      "balance_loss_clip": 1.05760324,
      "balance_loss_mlp": 1.0165298,
      "epoch": 0.4788071905248602,
      "flos": 19026551191680.0,
      "grad_norm": 7.171352620484166,
      "language_loss": 0.77539623,
      "learning_rate": 2.233933737232919e-06,
      "loss": 0.79749751,
      "num_input_tokens_seen": 85666875,
      "step": 3982,
      "time_per_iteration": 2.581099271774292
    },
    {
      "auxiliary_loss_clip": 0.01188126,
      "auxiliary_loss_mlp": 0.01028407,
      "balance_loss_clip": 1.05896628,
      "balance_loss_mlp": 1.01965737,
      "epoch": 0.4789274334154993,
      "flos": 23002221254400.0,
      "grad_norm": 2.613341709184141,
      "language_loss": 0.78264385,
      "learning_rate": 2.2331600948650793e-06,
      "loss": 0.80480921,
      "num_input_tokens_seen": 85687020,
      "step": 3983,
      "time_per_iteration": 2.6688036918640137
    },
    {
      "auxiliary_loss_clip": 0.01190515,
      "auxiliary_loss_mlp": 0.01023716,
      "balance_loss_clip": 1.05773151,
      "balance_loss_mlp": 1.01468611,
      "epoch": 0.4790476763061384,
      "flos": 23075586783360.0,
      "grad_norm": 2.292691757729396,
      "language_loss": 0.8013649,
      "learning_rate": 2.2323864171270386e-06,
      "loss": 0.82350719,
      "num_input_tokens_seen": 85708290,
      "step": 3984,
      "time_per_iteration": 2.632028341293335
    },
    {
      "auxiliary_loss_clip": 0.01190747,
      "auxiliary_loss_mlp": 0.0103034,
      "balance_loss_clip": 1.05691719,
      "balance_loss_mlp": 1.02099407,
      "epoch": 0.4791679191967775,
      "flos": 21179288073600.0,
      "grad_norm": 2.126680915637496,
      "language_loss": 0.73020613,
      "learning_rate": 2.231612704136164e-06,
      "loss": 0.75241697,
      "num_input_tokens_seen": 85728660,
      "step": 3985,
      "time_per_iteration": 2.6773085594177246
    },
    {
      "auxiliary_loss_clip": 0.01187837,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.05609465,
      "balance_loss_mlp": 1.01751959,
      "epoch": 0.4792881620874166,
      "flos": 22301495758080.0,
      "grad_norm": 8.188103043888415,
      "language_loss": 0.75043321,
      "learning_rate": 2.2308389560098253e-06,
      "loss": 0.77257413,
      "num_input_tokens_seen": 85745035,
      "step": 3986,
      "time_per_iteration": 2.5765841007232666
    },
    {
      "auxiliary_loss_clip": 0.01188437,
      "auxiliary_loss_mlp": 0.01024503,
      "balance_loss_clip": 1.05725086,
      "balance_loss_mlp": 1.01557422,
      "epoch": 0.47940840497805565,
      "flos": 17420877423360.0,
      "grad_norm": 4.991704776323061,
      "language_loss": 0.77147877,
      "learning_rate": 2.2300651728654008e-06,
      "loss": 0.79360819,
      "num_input_tokens_seen": 85760295,
      "step": 3987,
      "time_per_iteration": 2.593380928039551
    },
    {
      "auxiliary_loss_clip": 0.01084159,
      "auxiliary_loss_mlp": 0.01002562,
      "balance_loss_clip": 1.02030706,
      "balance_loss_mlp": 1.00109601,
      "epoch": 0.47952864786869476,
      "flos": 65358175708800.0,
      "grad_norm": 0.7331141782182288,
      "language_loss": 0.6018821,
      "learning_rate": 2.229291354820272e-06,
      "loss": 0.62274933,
      "num_input_tokens_seen": 85821305,
      "step": 3988,
      "time_per_iteration": 3.211642265319824
    },
    {
      "auxiliary_loss_clip": 0.0118978,
      "auxiliary_loss_mlp": 0.01030686,
      "balance_loss_clip": 1.05794764,
      "balance_loss_mlp": 1.02131593,
      "epoch": 0.47964889075933387,
      "flos": 16799802336000.0,
      "grad_norm": 2.464356859895801,
      "language_loss": 0.75707299,
      "learning_rate": 2.228517501991828e-06,
      "loss": 0.77927762,
      "num_input_tokens_seen": 85840105,
      "step": 3989,
      "time_per_iteration": 2.58622407913208
    },
    {
      "auxiliary_loss_clip": 0.0108366,
      "auxiliary_loss_mlp": 0.010025,
      "balance_loss_clip": 1.01993656,
      "balance_loss_mlp": 1.00117636,
      "epoch": 0.4797691336499729,
      "flos": 70079244808320.0,
      "grad_norm": 0.8136692719690081,
      "language_loss": 0.61075425,
      "learning_rate": 2.22774361449746e-06,
      "loss": 0.63161582,
      "num_input_tokens_seen": 85896585,
      "step": 3990,
      "time_per_iteration": 4.191905498504639
    },
    {
      "auxiliary_loss_clip": 0.01185679,
      "auxiliary_loss_mlp": 0.01033918,
      "balance_loss_clip": 1.05340767,
      "balance_loss_mlp": 1.02491117,
      "epoch": 0.47988937654061203,
      "flos": 18953329317120.0,
      "grad_norm": 2.49626870174472,
      "language_loss": 0.70565957,
      "learning_rate": 2.2269696924545668e-06,
      "loss": 0.72785556,
      "num_input_tokens_seen": 85914415,
      "step": 3991,
      "time_per_iteration": 2.6391384601593018
    },
    {
      "auxiliary_loss_clip": 0.01186117,
      "auxiliary_loss_mlp": 0.01031074,
      "balance_loss_clip": 1.05853724,
      "balance_loss_mlp": 1.02314651,
      "epoch": 0.48000961943125114,
      "flos": 14461981649280.0,
      "grad_norm": 6.058031825401001,
      "language_loss": 0.77951938,
      "learning_rate": 2.2261957359805523e-06,
      "loss": 0.80169129,
      "num_input_tokens_seen": 85931650,
      "step": 3992,
      "time_per_iteration": 2.655893325805664
    },
    {
      "auxiliary_loss_clip": 0.01189025,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.05883396,
      "balance_loss_mlp": 1.01897883,
      "epoch": 0.4801298623218902,
      "flos": 27051149105280.0,
      "grad_norm": 16.956194846180956,
      "language_loss": 0.73610687,
      "learning_rate": 2.225421745192823e-06,
      "loss": 0.75827664,
      "num_input_tokens_seen": 85951805,
      "step": 3993,
      "time_per_iteration": 3.5889034271240234
    },
    {
      "auxiliary_loss_clip": 0.01188123,
      "auxiliary_loss_mlp": 0.01027338,
      "balance_loss_clip": 1.05713785,
      "balance_loss_mlp": 1.01823044,
      "epoch": 0.4802501052125293,
      "flos": 26355236031360.0,
      "grad_norm": 3.017125103432875,
      "language_loss": 0.78372377,
      "learning_rate": 2.2246477202087955e-06,
      "loss": 0.8058784,
      "num_input_tokens_seen": 85972485,
      "step": 3994,
      "time_per_iteration": 2.6408095359802246
    },
    {
      "auxiliary_loss_clip": 0.01186735,
      "auxiliary_loss_mlp": 0.01024382,
      "balance_loss_clip": 1.05823922,
      "balance_loss_mlp": 1.01604927,
      "epoch": 0.4803703481031684,
      "flos": 20993916960000.0,
      "grad_norm": 5.754953984413755,
      "language_loss": 0.82933211,
      "learning_rate": 2.223873661145887e-06,
      "loss": 0.85144329,
      "num_input_tokens_seen": 85992540,
      "step": 3995,
      "time_per_iteration": 3.587092161178589
    },
    {
      "auxiliary_loss_clip": 0.01187621,
      "auxiliary_loss_mlp": 0.01023384,
      "balance_loss_clip": 1.05876291,
      "balance_loss_mlp": 1.01478958,
      "epoch": 0.4804905909938075,
      "flos": 20703722981760.0,
      "grad_norm": 13.763415864522425,
      "language_loss": 0.71359342,
      "learning_rate": 2.2230995681215226e-06,
      "loss": 0.73570347,
      "num_input_tokens_seen": 86012065,
      "step": 3996,
      "time_per_iteration": 3.5518548488616943
    },
    {
      "auxiliary_loss_clip": 0.01190986,
      "auxiliary_loss_mlp": 0.01024693,
      "balance_loss_clip": 1.05992246,
      "balance_loss_mlp": 1.01597881,
      "epoch": 0.4806108338844466,
      "flos": 16654831044480.0,
      "grad_norm": 2.2009903651528897,
      "language_loss": 0.78303385,
      "learning_rate": 2.2223254412531305e-06,
      "loss": 0.80519068,
      "num_input_tokens_seen": 86029435,
      "step": 3997,
      "time_per_iteration": 2.6129400730133057
    },
    {
      "auxiliary_loss_clip": 0.01181295,
      "auxiliary_loss_mlp": 0.01026882,
      "balance_loss_clip": 1.05424738,
      "balance_loss_mlp": 1.01880527,
      "epoch": 0.4807310767750857,
      "flos": 20011329440640.0,
      "grad_norm": 2.076449612069399,
      "language_loss": 0.8279894,
      "learning_rate": 2.221551280658146e-06,
      "loss": 0.85007119,
      "num_input_tokens_seen": 86048495,
      "step": 3998,
      "time_per_iteration": 2.6500725746154785
    },
    {
      "auxiliary_loss_clip": 0.01185664,
      "auxiliary_loss_mlp": 0.01026303,
      "balance_loss_clip": 1.05693901,
      "balance_loss_mlp": 1.01823235,
      "epoch": 0.48085131966572475,
      "flos": 23185257984000.0,
      "grad_norm": 1.7735759299983815,
      "language_loss": 0.74314547,
      "learning_rate": 2.2207770864540085e-06,
      "loss": 0.76526523,
      "num_input_tokens_seen": 86067470,
      "step": 3999,
      "time_per_iteration": 2.6480813026428223
    },
    {
      "auxiliary_loss_clip": 0.01186224,
      "auxiliary_loss_mlp": 0.01026243,
      "balance_loss_clip": 1.05734813,
      "balance_loss_mlp": 1.01790392,
      "epoch": 0.48097156255636386,
      "flos": 20558643949440.0,
      "grad_norm": 2.1116963576555845,
      "language_loss": 0.72944427,
      "learning_rate": 2.220002858758162e-06,
      "loss": 0.75156885,
      "num_input_tokens_seen": 86085460,
      "step": 4000,
      "time_per_iteration": 2.668778896331787
    },
    {
      "auxiliary_loss_clip": 0.01085579,
      "auxiliary_loss_mlp": 0.01001409,
      "balance_loss_clip": 1.02183199,
      "balance_loss_mlp": 1.00016356,
      "epoch": 0.481091805447003,
      "flos": 70511608817280.0,
      "grad_norm": 0.875932704882228,
      "language_loss": 0.60810924,
      "learning_rate": 2.2192285976880573e-06,
      "loss": 0.62897909,
      "num_input_tokens_seen": 86149715,
      "step": 4001,
      "time_per_iteration": 3.1883389949798584
    },
    {
      "auxiliary_loss_clip": 0.01185424,
      "auxiliary_loss_mlp": 0.01025251,
      "balance_loss_clip": 1.05792224,
      "balance_loss_mlp": 1.01701999,
      "epoch": 0.48121204833764203,
      "flos": 36428214839040.0,
      "grad_norm": 1.6317098834996446,
      "language_loss": 0.80601788,
      "learning_rate": 2.2184543033611485e-06,
      "loss": 0.82812464,
      "num_input_tokens_seen": 86170795,
      "step": 4002,
      "time_per_iteration": 2.758612632751465
    },
    {
      "auxiliary_loss_clip": 0.01185317,
      "auxiliary_loss_mlp": 0.01028503,
      "balance_loss_clip": 1.05633593,
      "balance_loss_mlp": 1.02063525,
      "epoch": 0.48133229122828114,
      "flos": 27490264871040.0,
      "grad_norm": 3.0244883816184696,
      "language_loss": 0.81505549,
      "learning_rate": 2.2176799758948957e-06,
      "loss": 0.83719373,
      "num_input_tokens_seen": 86190955,
      "step": 4003,
      "time_per_iteration": 2.7050817012786865
    },
    {
      "auxiliary_loss_clip": 0.01190338,
      "auxiliary_loss_mlp": 0.01031745,
      "balance_loss_clip": 1.05974841,
      "balance_loss_mlp": 1.02379417,
      "epoch": 0.4814525341189202,
      "flos": 43072802179200.0,
      "grad_norm": 2.029221853655018,
      "language_loss": 0.73272014,
      "learning_rate": 2.2169056154067635e-06,
      "loss": 0.75494099,
      "num_input_tokens_seen": 86214875,
      "step": 4004,
      "time_per_iteration": 2.794896364212036
    },
    {
      "auxiliary_loss_clip": 0.01187342,
      "auxiliary_loss_mlp": 0.01021912,
      "balance_loss_clip": 1.0566411,
      "balance_loss_mlp": 1.01362729,
      "epoch": 0.4815727770095593,
      "flos": 24236901400320.0,
      "grad_norm": 2.4332094210544684,
      "language_loss": 0.82227254,
      "learning_rate": 2.216131222014222e-06,
      "loss": 0.84436512,
      "num_input_tokens_seen": 86232950,
      "step": 4005,
      "time_per_iteration": 2.689333200454712
    },
    {
      "auxiliary_loss_clip": 0.01190408,
      "auxiliary_loss_mlp": 0.0102641,
      "balance_loss_clip": 1.06056309,
      "balance_loss_mlp": 1.01765418,
      "epoch": 0.4816930199001984,
      "flos": 18113630100480.0,
      "grad_norm": 2.12229758416466,
      "language_loss": 0.80218315,
      "learning_rate": 2.2153567958347455e-06,
      "loss": 0.82435131,
      "num_input_tokens_seen": 86249160,
      "step": 4006,
      "time_per_iteration": 2.568194627761841
    },
    {
      "auxiliary_loss_clip": 0.01187577,
      "auxiliary_loss_mlp": 0.01027923,
      "balance_loss_clip": 1.05780315,
      "balance_loss_mlp": 1.01929796,
      "epoch": 0.48181326279083747,
      "flos": 17274720983040.0,
      "grad_norm": 2.745065261992797,
      "language_loss": 0.797813,
      "learning_rate": 2.214582336985815e-06,
      "loss": 0.81996799,
      "num_input_tokens_seen": 86267060,
      "step": 4007,
      "time_per_iteration": 2.7573306560516357
    },
    {
      "auxiliary_loss_clip": 0.01190453,
      "auxiliary_loss_mlp": 0.01029585,
      "balance_loss_clip": 1.05970216,
      "balance_loss_mlp": 1.02090693,
      "epoch": 0.4819335056814766,
      "flos": 14903252231040.0,
      "grad_norm": 2.13906701139992,
      "language_loss": 0.65802795,
      "learning_rate": 2.2138078455849142e-06,
      "loss": 0.68022835,
      "num_input_tokens_seen": 86285055,
      "step": 4008,
      "time_per_iteration": 2.616729259490967
    },
    {
      "auxiliary_loss_clip": 0.01188785,
      "auxiliary_loss_mlp": 0.01026938,
      "balance_loss_clip": 1.05878949,
      "balance_loss_mlp": 1.01934481,
      "epoch": 0.4820537485721157,
      "flos": 19244888012160.0,
      "grad_norm": 2.0120301880070404,
      "language_loss": 0.78487074,
      "learning_rate": 2.2130333217495334e-06,
      "loss": 0.80702806,
      "num_input_tokens_seen": 86304225,
      "step": 4009,
      "time_per_iteration": 2.5956807136535645
    },
    {
      "auxiliary_loss_clip": 0.01186627,
      "auxiliary_loss_mlp": 0.01023759,
      "balance_loss_clip": 1.05560684,
      "balance_loss_mlp": 1.01488411,
      "epoch": 0.48217399146275475,
      "flos": 16033791870720.0,
      "grad_norm": 3.158578671587073,
      "language_loss": 0.67869216,
      "learning_rate": 2.2122587655971665e-06,
      "loss": 0.70079601,
      "num_input_tokens_seen": 86319170,
      "step": 4010,
      "time_per_iteration": 2.568146228790283
    },
    {
      "auxiliary_loss_clip": 0.01187437,
      "auxiliary_loss_mlp": 0.01026923,
      "balance_loss_clip": 1.05832183,
      "balance_loss_mlp": 1.01838779,
      "epoch": 0.48229423435339386,
      "flos": 24134197438080.0,
      "grad_norm": 3.9833363419763965,
      "language_loss": 0.64360058,
      "learning_rate": 2.211484177245314e-06,
      "loss": 0.66574419,
      "num_input_tokens_seen": 86338760,
      "step": 4011,
      "time_per_iteration": 2.6544229984283447
    },
    {
      "auxiliary_loss_clip": 0.01189444,
      "auxiliary_loss_mlp": 0.01023394,
      "balance_loss_clip": 1.05993342,
      "balance_loss_mlp": 1.01482296,
      "epoch": 0.48241447724403297,
      "flos": 23805435231360.0,
      "grad_norm": 2.251990920390241,
      "language_loss": 0.72347265,
      "learning_rate": 2.21070955681148e-06,
      "loss": 0.74560106,
      "num_input_tokens_seen": 86357865,
      "step": 4012,
      "time_per_iteration": 2.580921173095703
    },
    {
      "auxiliary_loss_clip": 0.01185805,
      "auxiliary_loss_mlp": 0.01031095,
      "balance_loss_clip": 1.05665636,
      "balance_loss_mlp": 1.02267325,
      "epoch": 0.482534720134672,
      "flos": 23110312256640.0,
      "grad_norm": 2.1759026746099437,
      "language_loss": 0.78084862,
      "learning_rate": 2.209934904413174e-06,
      "loss": 0.80301762,
      "num_input_tokens_seen": 86379470,
      "step": 4013,
      "time_per_iteration": 2.7381227016448975
    },
    {
      "auxiliary_loss_clip": 0.01188562,
      "auxiliary_loss_mlp": 0.01038001,
      "balance_loss_clip": 1.05710125,
      "balance_loss_mlp": 1.02928114,
      "epoch": 0.48265496302531113,
      "flos": 20923819568640.0,
      "grad_norm": 1.8398573932678024,
      "language_loss": 0.71599603,
      "learning_rate": 2.2091602201679095e-06,
      "loss": 0.7382617,
      "num_input_tokens_seen": 86399080,
      "step": 4014,
      "time_per_iteration": 2.643726110458374
    },
    {
      "auxiliary_loss_clip": 0.01186725,
      "auxiliary_loss_mlp": 0.0102628,
      "balance_loss_clip": 1.0579654,
      "balance_loss_mlp": 1.01754165,
      "epoch": 0.48277520591595025,
      "flos": 15231152511360.0,
      "grad_norm": 2.3812852877673034,
      "language_loss": 0.82843584,
      "learning_rate": 2.208385504193206e-06,
      "loss": 0.85056591,
      "num_input_tokens_seen": 86416580,
      "step": 4015,
      "time_per_iteration": 2.6643424034118652
    },
    {
      "auxiliary_loss_clip": 0.01186573,
      "auxiliary_loss_mlp": 0.01022087,
      "balance_loss_clip": 1.05572212,
      "balance_loss_mlp": 1.01395714,
      "epoch": 0.4828954488065893,
      "flos": 17858664385920.0,
      "grad_norm": 3.381066136945774,
      "language_loss": 0.81586087,
      "learning_rate": 2.2076107566065873e-06,
      "loss": 0.83794749,
      "num_input_tokens_seen": 86434365,
      "step": 4016,
      "time_per_iteration": 3.5978336334228516
    },
    {
      "auxiliary_loss_clip": 0.01188108,
      "auxiliary_loss_mlp": 0.01032728,
      "balance_loss_clip": 1.05975664,
      "balance_loss_mlp": 1.024544,
      "epoch": 0.4830156916972284,
      "flos": 32087405070720.0,
      "grad_norm": 4.562873726399259,
      "language_loss": 0.75711501,
      "learning_rate": 2.2068359775255816e-06,
      "loss": 0.77932334,
      "num_input_tokens_seen": 86452675,
      "step": 4017,
      "time_per_iteration": 2.702258825302124
    },
    {
      "auxiliary_loss_clip": 0.01186268,
      "auxiliary_loss_mlp": 0.0102631,
      "balance_loss_clip": 1.05540752,
      "balance_loss_mlp": 1.0174824,
      "epoch": 0.48313593458786747,
      "flos": 21871717528320.0,
      "grad_norm": 2.779401183377711,
      "language_loss": 0.78531301,
      "learning_rate": 2.206061167067723e-06,
      "loss": 0.80743873,
      "num_input_tokens_seen": 86470785,
      "step": 4018,
      "time_per_iteration": 2.66884708404541
    },
    {
      "auxiliary_loss_clip": 0.01186236,
      "auxiliary_loss_mlp": 0.01024438,
      "balance_loss_clip": 1.05566692,
      "balance_loss_mlp": 1.01556265,
      "epoch": 0.4832561774785066,
      "flos": 22601206840320.0,
      "grad_norm": 3.1951997432550834,
      "language_loss": 0.80178535,
      "learning_rate": 2.205286325350549e-06,
      "loss": 0.82389212,
      "num_input_tokens_seen": 86489850,
      "step": 4019,
      "time_per_iteration": 3.564640998840332
    },
    {
      "auxiliary_loss_clip": 0.01185678,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.05530405,
      "balance_loss_mlp": 1.01824498,
      "epoch": 0.4833764203691457,
      "flos": 13437342282240.0,
      "grad_norm": 2.2048186995827566,
      "language_loss": 0.7276656,
      "learning_rate": 2.204511452491603e-06,
      "loss": 0.74979293,
      "num_input_tokens_seen": 86506475,
      "step": 4020,
      "time_per_iteration": 2.6618504524230957
    },
    {
      "auxiliary_loss_clip": 0.01185638,
      "auxiliary_loss_mlp": 0.0102392,
      "balance_loss_clip": 1.05954385,
      "balance_loss_mlp": 1.01579618,
      "epoch": 0.48349666325978474,
      "flos": 44128036955520.0,
      "grad_norm": 1.6741431071749802,
      "language_loss": 0.75108749,
      "learning_rate": 2.2037365486084316e-06,
      "loss": 0.77318311,
      "num_input_tokens_seen": 86529715,
      "step": 4021,
      "time_per_iteration": 2.8597264289855957
    },
    {
      "auxiliary_loss_clip": 0.01186257,
      "auxiliary_loss_mlp": 0.01026062,
      "balance_loss_clip": 1.05578375,
      "balance_loss_mlp": 1.01816487,
      "epoch": 0.48361690615042385,
      "flos": 26028377245440.0,
      "grad_norm": 2.433640745211433,
      "language_loss": 0.77673411,
      "learning_rate": 2.2029616138185886e-06,
      "loss": 0.79885739,
      "num_input_tokens_seen": 86548715,
      "step": 4022,
      "time_per_iteration": 2.6440365314483643
    },
    {
      "auxiliary_loss_clip": 0.01186704,
      "auxiliary_loss_mlp": 0.01029637,
      "balance_loss_clip": 1.05682933,
      "balance_loss_mlp": 1.0214417,
      "epoch": 0.48373714904106296,
      "flos": 22273306560000.0,
      "grad_norm": 1.89290607314886,
      "language_loss": 0.82694238,
      "learning_rate": 2.202186648239629e-06,
      "loss": 0.84910583,
      "num_input_tokens_seen": 86568650,
      "step": 4023,
      "time_per_iteration": 3.656848430633545
    },
    {
      "auxiliary_loss_clip": 0.01185072,
      "auxiliary_loss_mlp": 0.01028893,
      "balance_loss_clip": 1.05602551,
      "balance_loss_mlp": 1.02002406,
      "epoch": 0.483857391931702,
      "flos": 28292293699200.0,
      "grad_norm": 2.220651167945844,
      "language_loss": 0.7164216,
      "learning_rate": 2.201411651989117e-06,
      "loss": 0.73856127,
      "num_input_tokens_seen": 86590630,
      "step": 4024,
      "time_per_iteration": 2.6526215076446533
    },
    {
      "auxiliary_loss_clip": 0.01184638,
      "auxiliary_loss_mlp": 0.01027062,
      "balance_loss_clip": 1.0556767,
      "balance_loss_mlp": 1.01874709,
      "epoch": 0.48397763482234113,
      "flos": 27418048577280.0,
      "grad_norm": 2.3131649184642016,
      "language_loss": 0.7800281,
      "learning_rate": 2.2006366251846167e-06,
      "loss": 0.802145,
      "num_input_tokens_seen": 86611270,
      "step": 4025,
      "time_per_iteration": 2.7666382789611816
    },
    {
      "auxiliary_loss_clip": 0.01188649,
      "auxiliary_loss_mlp": 0.01029216,
      "balance_loss_clip": 1.05943704,
      "balance_loss_mlp": 1.02081156,
      "epoch": 0.48409787771298024,
      "flos": 16797252470400.0,
      "grad_norm": 2.050343181869471,
      "language_loss": 0.75685388,
      "learning_rate": 2.1998615679436997e-06,
      "loss": 0.77903247,
      "num_input_tokens_seen": 86628810,
      "step": 4026,
      "time_per_iteration": 2.6557154655456543
    },
    {
      "auxiliary_loss_clip": 0.01188409,
      "auxiliary_loss_mlp": 0.01030167,
      "balance_loss_clip": 1.05726874,
      "balance_loss_mlp": 1.02156615,
      "epoch": 0.4842181206036193,
      "flos": 25083496028160.0,
      "grad_norm": 2.6959327483601188,
      "language_loss": 0.77458704,
      "learning_rate": 2.199086480383942e-06,
      "loss": 0.79677284,
      "num_input_tokens_seen": 86648185,
      "step": 4027,
      "time_per_iteration": 2.665717840194702
    },
    {
      "auxiliary_loss_clip": 0.01193321,
      "auxiliary_loss_mlp": 0.0103015,
      "balance_loss_clip": 1.05961514,
      "balance_loss_mlp": 1.01972497,
      "epoch": 0.4843383634942584,
      "flos": 30372311496960.0,
      "grad_norm": 3.538568678604959,
      "language_loss": 0.67807657,
      "learning_rate": 2.1983113626229234e-06,
      "loss": 0.70031124,
      "num_input_tokens_seen": 86667435,
      "step": 4028,
      "time_per_iteration": 2.7029876708984375
    },
    {
      "auxiliary_loss_clip": 0.01187147,
      "auxiliary_loss_mlp": 0.01029734,
      "balance_loss_clip": 1.05703866,
      "balance_loss_mlp": 1.02093697,
      "epoch": 0.4844586063848975,
      "flos": 20413564917120.0,
      "grad_norm": 2.514491009196959,
      "language_loss": 0.78526974,
      "learning_rate": 2.1975362147782293e-06,
      "loss": 0.80743861,
      "num_input_tokens_seen": 86686630,
      "step": 4029,
      "time_per_iteration": 2.6844277381896973
    },
    {
      "auxiliary_loss_clip": 0.01087554,
      "auxiliary_loss_mlp": 0.01005228,
      "balance_loss_clip": 1.02348709,
      "balance_loss_mlp": 1.00400031,
      "epoch": 0.48457884927553657,
      "flos": 70303722854400.0,
      "grad_norm": 0.7071016380119987,
      "language_loss": 0.54112822,
      "learning_rate": 2.196761036967448e-06,
      "loss": 0.56205601,
      "num_input_tokens_seen": 86754595,
      "step": 4030,
      "time_per_iteration": 3.2904744148254395
    },
    {
      "auxiliary_loss_clip": 0.01181654,
      "auxiliary_loss_mlp": 0.01026594,
      "balance_loss_clip": 1.05422115,
      "balance_loss_mlp": 1.0185231,
      "epoch": 0.4846990921661757,
      "flos": 19934516206080.0,
      "grad_norm": 2.070362911095609,
      "language_loss": 0.77662516,
      "learning_rate": 2.1959858293081743e-06,
      "loss": 0.7987076,
      "num_input_tokens_seen": 86773730,
      "step": 4031,
      "time_per_iteration": 2.715075969696045
    },
    {
      "auxiliary_loss_clip": 0.0119017,
      "auxiliary_loss_mlp": 0.0102944,
      "balance_loss_clip": 1.06000686,
      "balance_loss_mlp": 1.02020133,
      "epoch": 0.4848193350568148,
      "flos": 23075945919360.0,
      "grad_norm": 2.0053003261611915,
      "language_loss": 0.75684178,
      "learning_rate": 2.1952105919180056e-06,
      "loss": 0.77903783,
      "num_input_tokens_seen": 86792985,
      "step": 4032,
      "time_per_iteration": 2.6746628284454346
    },
    {
      "auxiliary_loss_clip": 0.01188576,
      "auxiliary_loss_mlp": 0.01033283,
      "balance_loss_clip": 1.05840802,
      "balance_loss_mlp": 1.02413368,
      "epoch": 0.48493957794745385,
      "flos": 22455481363200.0,
      "grad_norm": 3.0351287461661336,
      "language_loss": 0.67778158,
      "learning_rate": 2.1944353249145456e-06,
      "loss": 0.70000017,
      "num_input_tokens_seen": 86812095,
      "step": 4033,
      "time_per_iteration": 2.6823129653930664
    },
    {
      "auxiliary_loss_clip": 0.01187287,
      "auxiliary_loss_mlp": 0.01030261,
      "balance_loss_clip": 1.05890393,
      "balance_loss_mlp": 1.02184463,
      "epoch": 0.48505982083809296,
      "flos": 25046112948480.0,
      "grad_norm": 1.8189414744810732,
      "language_loss": 0.74603385,
      "learning_rate": 2.193660028415401e-06,
      "loss": 0.76820934,
      "num_input_tokens_seen": 86832875,
      "step": 4034,
      "time_per_iteration": 2.7068285942077637
    },
    {
      "auxiliary_loss_clip": 0.01187033,
      "auxiliary_loss_mlp": 0.01029511,
      "balance_loss_clip": 1.05665755,
      "balance_loss_mlp": 1.02139831,
      "epoch": 0.485180063728732,
      "flos": 26761386090240.0,
      "grad_norm": 2.3210883975307266,
      "language_loss": 0.81999797,
      "learning_rate": 2.1928847025381852e-06,
      "loss": 0.84216338,
      "num_input_tokens_seen": 86853480,
      "step": 4035,
      "time_per_iteration": 2.6366159915924072
    },
    {
      "auxiliary_loss_clip": 0.01186847,
      "auxiliary_loss_mlp": 0.01028643,
      "balance_loss_clip": 1.0536716,
      "balance_loss_mlp": 1.0199703,
      "epoch": 0.4853003066193711,
      "flos": 24059143969920.0,
      "grad_norm": 1.7902665254489598,
      "language_loss": 0.84052873,
      "learning_rate": 2.192109347400512e-06,
      "loss": 0.86268353,
      "num_input_tokens_seen": 86873695,
      "step": 4036,
      "time_per_iteration": 2.683666229248047
    },
    {
      "auxiliary_loss_clip": 0.01187834,
      "auxiliary_loss_mlp": 0.01031969,
      "balance_loss_clip": 1.05641985,
      "balance_loss_mlp": 1.02277184,
      "epoch": 0.48542054951001024,
      "flos": 23076376882560.0,
      "grad_norm": 1.8087067801115941,
      "language_loss": 0.78685516,
      "learning_rate": 2.191333963120004e-06,
      "loss": 0.80905318,
      "num_input_tokens_seen": 86892675,
      "step": 4037,
      "time_per_iteration": 2.6259403228759766
    },
    {
      "auxiliary_loss_clip": 0.01184497,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 1.05374503,
      "balance_loss_mlp": 1.02192688,
      "epoch": 0.4855407924006493,
      "flos": 25664889565440.0,
      "grad_norm": 2.8442435590507866,
      "language_loss": 0.70619869,
      "learning_rate": 2.190558549814286e-06,
      "loss": 0.7283439,
      "num_input_tokens_seen": 86912835,
      "step": 4038,
      "time_per_iteration": 2.672819137573242
    },
    {
      "auxiliary_loss_clip": 0.01188518,
      "auxiliary_loss_mlp": 0.01025616,
      "balance_loss_clip": 1.05831623,
      "balance_loss_mlp": 1.01677024,
      "epoch": 0.4856610352912884,
      "flos": 23987933256960.0,
      "grad_norm": 1.9928333119044384,
      "language_loss": 0.79769289,
      "learning_rate": 2.1897831076009872e-06,
      "loss": 0.81983423,
      "num_input_tokens_seen": 86932475,
      "step": 4039,
      "time_per_iteration": 2.6700658798217773
    },
    {
      "auxiliary_loss_clip": 0.01184485,
      "auxiliary_loss_mlp": 0.01031346,
      "balance_loss_clip": 1.05467391,
      "balance_loss_mlp": 1.0229001,
      "epoch": 0.4857812781819275,
      "flos": 24096814358400.0,
      "grad_norm": 1.9246625265505857,
      "language_loss": 0.80062425,
      "learning_rate": 2.1890076365977426e-06,
      "loss": 0.82278258,
      "num_input_tokens_seen": 86952300,
      "step": 4040,
      "time_per_iteration": 2.6370975971221924
    },
    {
      "auxiliary_loss_clip": 0.01085515,
      "auxiliary_loss_mlp": 0.0100224,
      "balance_loss_clip": 1.02109766,
      "balance_loss_mlp": 1.0010066,
      "epoch": 0.48590152107256657,
      "flos": 56266635185280.0,
      "grad_norm": 0.8531807568034662,
      "language_loss": 0.52779996,
      "learning_rate": 2.188232136922189e-06,
      "loss": 0.54867756,
      "num_input_tokens_seen": 87010420,
      "step": 4041,
      "time_per_iteration": 3.121689558029175
    },
    {
      "auxiliary_loss_clip": 0.01185186,
      "auxiliary_loss_mlp": 0.01033123,
      "balance_loss_clip": 1.05395281,
      "balance_loss_mlp": 1.02399123,
      "epoch": 0.4860217639632057,
      "flos": 20046988667520.0,
      "grad_norm": 7.2948038663553865,
      "language_loss": 0.75781465,
      "learning_rate": 2.187456608691971e-06,
      "loss": 0.77999771,
      "num_input_tokens_seen": 87029295,
      "step": 4042,
      "time_per_iteration": 2.578439474105835
    },
    {
      "auxiliary_loss_clip": 0.01186674,
      "auxiliary_loss_mlp": 0.0102989,
      "balance_loss_clip": 1.05815911,
      "balance_loss_mlp": 1.02141428,
      "epoch": 0.4861420068538448,
      "flos": 17822143232640.0,
      "grad_norm": 2.076830790512758,
      "language_loss": 0.8738606,
      "learning_rate": 2.1866810520247334e-06,
      "loss": 0.89602625,
      "num_input_tokens_seen": 87048165,
      "step": 4043,
      "time_per_iteration": 3.617093563079834
    },
    {
      "auxiliary_loss_clip": 0.0118757,
      "auxiliary_loss_mlp": 0.01031802,
      "balance_loss_clip": 1.05565989,
      "balance_loss_mlp": 1.02270603,
      "epoch": 0.48626224974448384,
      "flos": 26250125857920.0,
      "grad_norm": 2.0259068516794065,
      "language_loss": 0.64905035,
      "learning_rate": 2.185905467038129e-06,
      "loss": 0.67124408,
      "num_input_tokens_seen": 87067070,
      "step": 4044,
      "time_per_iteration": 2.6065828800201416
    },
    {
      "auxiliary_loss_clip": 0.01183214,
      "auxiliary_loss_mlp": 0.01025672,
      "balance_loss_clip": 1.05722511,
      "balance_loss_mlp": 1.01741099,
      "epoch": 0.48638249263512295,
      "flos": 22054502862720.0,
      "grad_norm": 2.073232102275352,
      "language_loss": 0.77763528,
      "learning_rate": 2.1851298538498127e-06,
      "loss": 0.79972416,
      "num_input_tokens_seen": 87086785,
      "step": 4045,
      "time_per_iteration": 3.589550018310547
    },
    {
      "auxiliary_loss_clip": 0.01191629,
      "auxiliary_loss_mlp": 0.01027214,
      "balance_loss_clip": 1.05937588,
      "balance_loss_mlp": 1.01767135,
      "epoch": 0.48650273552576206,
      "flos": 25119945354240.0,
      "grad_norm": 2.2661667346526495,
      "language_loss": 0.79772604,
      "learning_rate": 2.184354212577446e-06,
      "loss": 0.81991446,
      "num_input_tokens_seen": 87107090,
      "step": 4046,
      "time_per_iteration": 2.70937442779541
    },
    {
      "auxiliary_loss_clip": 0.01192138,
      "auxiliary_loss_mlp": 0.01033076,
      "balance_loss_clip": 1.05873775,
      "balance_loss_mlp": 1.02383113,
      "epoch": 0.4866229784164011,
      "flos": 17456931699840.0,
      "grad_norm": 3.370473158296385,
      "language_loss": 0.62590861,
      "learning_rate": 2.1835785433386907e-06,
      "loss": 0.64816076,
      "num_input_tokens_seen": 87125905,
      "step": 4047,
      "time_per_iteration": 2.6221606731414795
    },
    {
      "auxiliary_loss_clip": 0.01187839,
      "auxiliary_loss_mlp": 0.01033319,
      "balance_loss_clip": 1.05779123,
      "balance_loss_mlp": 1.02456856,
      "epoch": 0.48674322130704023,
      "flos": 23331127115520.0,
      "grad_norm": 2.3915133494512504,
      "language_loss": 0.65379399,
      "learning_rate": 2.182802846251216e-06,
      "loss": 0.6760056,
      "num_input_tokens_seen": 87146175,
      "step": 4048,
      "time_per_iteration": 2.6518054008483887
    },
    {
      "auxiliary_loss_clip": 0.01186472,
      "auxiliary_loss_mlp": 0.01027374,
      "balance_loss_clip": 1.05586505,
      "balance_loss_mlp": 1.01966119,
      "epoch": 0.4868634641976793,
      "flos": 28804344030720.0,
      "grad_norm": 2.2085150994694644,
      "language_loss": 0.72502625,
      "learning_rate": 2.182027121432696e-06,
      "loss": 0.74716467,
      "num_input_tokens_seen": 87166800,
      "step": 4049,
      "time_per_iteration": 2.6975464820861816
    },
    {
      "auxiliary_loss_clip": 0.01189729,
      "auxiliary_loss_mlp": 0.01030677,
      "balance_loss_clip": 1.05797362,
      "balance_loss_mlp": 1.02204597,
      "epoch": 0.4869837070883184,
      "flos": 19025976574080.0,
      "grad_norm": 2.15285645507139,
      "language_loss": 0.82297271,
      "learning_rate": 2.1812513690008054e-06,
      "loss": 0.84517676,
      "num_input_tokens_seen": 87185920,
      "step": 4050,
      "time_per_iteration": 4.454613924026489
    },
    {
      "auxiliary_loss_clip": 0.01188958,
      "auxiliary_loss_mlp": 0.01033488,
      "balance_loss_clip": 1.05682015,
      "balance_loss_mlp": 1.0243026,
      "epoch": 0.4871039499789575,
      "flos": 15121409483520.0,
      "grad_norm": 2.5576277509978054,
      "language_loss": 0.80094242,
      "learning_rate": 2.180475589073227e-06,
      "loss": 0.82316685,
      "num_input_tokens_seen": 87203620,
      "step": 4051,
      "time_per_iteration": 2.677687883377075
    },
    {
      "auxiliary_loss_clip": 0.0118545,
      "auxiliary_loss_mlp": 0.01027107,
      "balance_loss_clip": 1.05710888,
      "balance_loss_mlp": 1.01857769,
      "epoch": 0.48722419286959656,
      "flos": 26174066808960.0,
      "grad_norm": 1.686873528804224,
      "language_loss": 0.73711514,
      "learning_rate": 2.1796997817676456e-06,
      "loss": 0.75924075,
      "num_input_tokens_seen": 87224630,
      "step": 4052,
      "time_per_iteration": 2.639051914215088
    },
    {
      "auxiliary_loss_clip": 0.01184985,
      "auxiliary_loss_mlp": 0.01024374,
      "balance_loss_clip": 1.05852413,
      "balance_loss_mlp": 1.01635695,
      "epoch": 0.4873444357602357,
      "flos": 24026142349440.0,
      "grad_norm": 1.551533878143104,
      "language_loss": 0.67347801,
      "learning_rate": 2.1789239472017494e-06,
      "loss": 0.69557154,
      "num_input_tokens_seen": 87246280,
      "step": 4053,
      "time_per_iteration": 2.7554094791412354
    },
    {
      "auxiliary_loss_clip": 0.0118928,
      "auxiliary_loss_mlp": 0.0102623,
      "balance_loss_clip": 1.05927479,
      "balance_loss_mlp": 1.01777756,
      "epoch": 0.4874646786508748,
      "flos": 22820441500800.0,
      "grad_norm": 5.197796373326577,
      "language_loss": 0.73174256,
      "learning_rate": 2.1781480854932326e-06,
      "loss": 0.75389761,
      "num_input_tokens_seen": 87266045,
      "step": 4054,
      "time_per_iteration": 2.615952491760254
    },
    {
      "auxiliary_loss_clip": 0.01184985,
      "auxiliary_loss_mlp": 0.01022815,
      "balance_loss_clip": 1.05471206,
      "balance_loss_mlp": 1.01478016,
      "epoch": 0.48758492154151384,
      "flos": 21287594557440.0,
      "grad_norm": 2.5062273044940317,
      "language_loss": 0.79287595,
      "learning_rate": 2.1773721967597933e-06,
      "loss": 0.81495392,
      "num_input_tokens_seen": 87284495,
      "step": 4055,
      "time_per_iteration": 2.6215333938598633
    },
    {
      "auxiliary_loss_clip": 0.01085223,
      "auxiliary_loss_mlp": 0.01003396,
      "balance_loss_clip": 1.02130938,
      "balance_loss_mlp": 1.00206125,
      "epoch": 0.48770516443215295,
      "flos": 62244109180800.0,
      "grad_norm": 0.8798956717859163,
      "language_loss": 0.57416272,
      "learning_rate": 2.1765962811191322e-06,
      "loss": 0.5950489,
      "num_input_tokens_seen": 87338960,
      "step": 4056,
      "time_per_iteration": 3.0904526710510254
    },
    {
      "auxiliary_loss_clip": 0.01084858,
      "auxiliary_loss_mlp": 0.01000747,
      "balance_loss_clip": 1.02085185,
      "balance_loss_mlp": 0.99950087,
      "epoch": 0.48782540732279206,
      "flos": 66133451882880.0,
      "grad_norm": 0.8280202356048113,
      "language_loss": 0.620628,
      "learning_rate": 2.1758203386889566e-06,
      "loss": 0.64148402,
      "num_input_tokens_seen": 87401730,
      "step": 4057,
      "time_per_iteration": 3.2585976123809814
    },
    {
      "auxiliary_loss_clip": 0.01187131,
      "auxiliary_loss_mlp": 0.01026534,
      "balance_loss_clip": 1.05638862,
      "balance_loss_mlp": 1.01771295,
      "epoch": 0.4879456502134311,
      "flos": 14607922608000.0,
      "grad_norm": 2.4899767707828744,
      "language_loss": 0.84120107,
      "learning_rate": 2.1750443695869746e-06,
      "loss": 0.86333776,
      "num_input_tokens_seen": 87417300,
      "step": 4058,
      "time_per_iteration": 2.6266040802001953
    },
    {
      "auxiliary_loss_clip": 0.01186595,
      "auxiliary_loss_mlp": 0.01024686,
      "balance_loss_clip": 1.05417871,
      "balance_loss_mlp": 1.01649046,
      "epoch": 0.4880658931040702,
      "flos": 19500464257920.0,
      "grad_norm": 1.9680910803644829,
      "language_loss": 0.85636604,
      "learning_rate": 2.174268373930901e-06,
      "loss": 0.87847888,
      "num_input_tokens_seen": 87434815,
      "step": 4059,
      "time_per_iteration": 2.585766553878784
    },
    {
      "auxiliary_loss_clip": 0.01184284,
      "auxiliary_loss_mlp": 0.01025091,
      "balance_loss_clip": 1.05412769,
      "balance_loss_mlp": 1.01634073,
      "epoch": 0.48818613599470934,
      "flos": 16723060928640.0,
      "grad_norm": 6.470348377612592,
      "language_loss": 0.80369884,
      "learning_rate": 2.1734923518384537e-06,
      "loss": 0.82579255,
      "num_input_tokens_seen": 87451420,
      "step": 4060,
      "time_per_iteration": 2.6699626445770264
    },
    {
      "auxiliary_loss_clip": 0.01183662,
      "auxiliary_loss_mlp": 0.01025208,
      "balance_loss_clip": 1.05605578,
      "balance_loss_mlp": 1.0165416,
      "epoch": 0.4883063788853484,
      "flos": 26756932803840.0,
      "grad_norm": 2.0443937956147127,
      "language_loss": 0.82510841,
      "learning_rate": 2.1727163034273547e-06,
      "loss": 0.84719712,
      "num_input_tokens_seen": 87469585,
      "step": 4061,
      "time_per_iteration": 2.64505672454834
    },
    {
      "auxiliary_loss_clip": 0.01184071,
      "auxiliary_loss_mlp": 0.01028967,
      "balance_loss_clip": 1.05549848,
      "balance_loss_mlp": 1.01991951,
      "epoch": 0.4884266217759875,
      "flos": 16763388923520.0,
      "grad_norm": 2.766086451359099,
      "language_loss": 0.78759927,
      "learning_rate": 2.17194022881533e-06,
      "loss": 0.80972958,
      "num_input_tokens_seen": 87485675,
      "step": 4062,
      "time_per_iteration": 2.6282546520233154
    },
    {
      "auxiliary_loss_clip": 0.01189647,
      "auxiliary_loss_mlp": 0.01030617,
      "balance_loss_clip": 1.05704319,
      "balance_loss_mlp": 1.02127147,
      "epoch": 0.4885468646666266,
      "flos": 24207132003840.0,
      "grad_norm": 14.488586367655017,
      "language_loss": 0.67808914,
      "learning_rate": 2.1711641281201092e-06,
      "loss": 0.70029181,
      "num_input_tokens_seen": 87505605,
      "step": 4063,
      "time_per_iteration": 2.7650461196899414
    },
    {
      "auxiliary_loss_clip": 0.01183603,
      "auxiliary_loss_mlp": 0.01027443,
      "balance_loss_clip": 1.0559727,
      "balance_loss_mlp": 1.01909816,
      "epoch": 0.48866710755726567,
      "flos": 14610795696000.0,
      "grad_norm": 2.686623197557009,
      "language_loss": 0.79536974,
      "learning_rate": 2.1703880014594264e-06,
      "loss": 0.81748021,
      "num_input_tokens_seen": 87523195,
      "step": 4064,
      "time_per_iteration": 2.6746864318847656
    },
    {
      "auxiliary_loss_clip": 0.01183837,
      "auxiliary_loss_mlp": 0.01028919,
      "balance_loss_clip": 1.05638599,
      "balance_loss_mlp": 1.02094138,
      "epoch": 0.4887873504479048,
      "flos": 28804451771520.0,
      "grad_norm": 1.757669574673539,
      "language_loss": 0.73642182,
      "learning_rate": 2.1696118489510182e-06,
      "loss": 0.75854945,
      "num_input_tokens_seen": 87544125,
      "step": 4065,
      "time_per_iteration": 2.689431667327881
    },
    {
      "auxiliary_loss_clip": 0.01187247,
      "auxiliary_loss_mlp": 0.01027407,
      "balance_loss_clip": 1.05743718,
      "balance_loss_mlp": 1.01838255,
      "epoch": 0.48890759333854383,
      "flos": 22784387224320.0,
      "grad_norm": 2.611650209198992,
      "language_loss": 0.72594446,
      "learning_rate": 2.1688356707126286e-06,
      "loss": 0.74809098,
      "num_input_tokens_seen": 87563745,
      "step": 4066,
      "time_per_iteration": 2.671617031097412
    },
    {
      "auxiliary_loss_clip": 0.0119065,
      "auxiliary_loss_mlp": 0.01026811,
      "balance_loss_clip": 1.05886328,
      "balance_loss_mlp": 1.01804888,
      "epoch": 0.48902783622918294,
      "flos": 17786088956160.0,
      "grad_norm": 2.1453780561127807,
      "language_loss": 0.70413077,
      "learning_rate": 2.168059466862001e-06,
      "loss": 0.72630537,
      "num_input_tokens_seen": 87581895,
      "step": 4067,
      "time_per_iteration": 2.5909523963928223
    },
    {
      "auxiliary_loss_clip": 0.01182349,
      "auxiliary_loss_mlp": 0.01031545,
      "balance_loss_clip": 1.05202568,
      "balance_loss_mlp": 1.02311134,
      "epoch": 0.48914807911982205,
      "flos": 22310294590080.0,
      "grad_norm": 2.113215324508856,
      "language_loss": 0.81773734,
      "learning_rate": 2.167283237516887e-06,
      "loss": 0.83987629,
      "num_input_tokens_seen": 87600170,
      "step": 4068,
      "time_per_iteration": 2.686631441116333
    },
    {
      "auxiliary_loss_clip": 0.01184231,
      "auxiliary_loss_mlp": 0.01030092,
      "balance_loss_clip": 1.05436873,
      "balance_loss_mlp": 1.02136016,
      "epoch": 0.4892683220104611,
      "flos": 16363020954240.0,
      "grad_norm": 2.101948002769078,
      "language_loss": 0.74452949,
      "learning_rate": 2.1665069827950383e-06,
      "loss": 0.76667273,
      "num_input_tokens_seen": 87617455,
      "step": 4069,
      "time_per_iteration": 3.5427894592285156
    },
    {
      "auxiliary_loss_clip": 0.01183345,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.05622363,
      "balance_loss_mlp": 1.01812506,
      "epoch": 0.4893885649011002,
      "flos": 15739144606080.0,
      "grad_norm": 2.2298503335456603,
      "language_loss": 0.86897838,
      "learning_rate": 2.1657307028142126e-06,
      "loss": 0.89107144,
      "num_input_tokens_seen": 87634995,
      "step": 4070,
      "time_per_iteration": 2.616888999938965
    },
    {
      "auxiliary_loss_clip": 0.01187737,
      "auxiliary_loss_mlp": 0.01028511,
      "balance_loss_clip": 1.0567199,
      "balance_loss_mlp": 1.01985669,
      "epoch": 0.48950880779173933,
      "flos": 28581984887040.0,
      "grad_norm": 3.1730393520872155,
      "language_loss": 0.67712653,
      "learning_rate": 2.164954397692171e-06,
      "loss": 0.69928902,
      "num_input_tokens_seen": 87654420,
      "step": 4071,
      "time_per_iteration": 3.5669074058532715
    },
    {
      "auxiliary_loss_clip": 0.01084727,
      "auxiliary_loss_mlp": 0.01002795,
      "balance_loss_clip": 1.0209049,
      "balance_loss_mlp": 1.00153184,
      "epoch": 0.4896290506823784,
      "flos": 66186310746240.0,
      "grad_norm": 1.075771660412557,
      "language_loss": 0.77302486,
      "learning_rate": 2.164178067546678e-06,
      "loss": 0.79390007,
      "num_input_tokens_seen": 87713585,
      "step": 4072,
      "time_per_iteration": 3.2564756870269775
    },
    {
      "auxiliary_loss_clip": 0.0118124,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.0518676,
      "balance_loss_mlp": 1.0201416,
      "epoch": 0.4897492935730175,
      "flos": 12531065207040.0,
      "grad_norm": 1.976725997293038,
      "language_loss": 0.91481262,
      "learning_rate": 2.163401712495504e-06,
      "loss": 0.93691349,
      "num_input_tokens_seen": 87731280,
      "step": 4073,
      "time_per_iteration": 2.572968006134033
    },
    {
      "auxiliary_loss_clip": 0.01186408,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.05524445,
      "balance_loss_mlp": 1.01893044,
      "epoch": 0.4898695364636566,
      "flos": 23476816679040.0,
      "grad_norm": 2.059225323681628,
      "language_loss": 0.79305673,
      "learning_rate": 2.1626253326564194e-06,
      "loss": 0.81520033,
      "num_input_tokens_seen": 87750230,
      "step": 4074,
      "time_per_iteration": 2.7074480056762695
    },
    {
      "auxiliary_loss_clip": 0.01186154,
      "auxiliary_loss_mlp": 0.01030608,
      "balance_loss_clip": 1.05487227,
      "balance_loss_mlp": 1.0213455,
      "epoch": 0.48998977935429566,
      "flos": 27160209774720.0,
      "grad_norm": 1.7694719396028962,
      "language_loss": 0.77079177,
      "learning_rate": 2.161848928147201e-06,
      "loss": 0.79295939,
      "num_input_tokens_seen": 87770500,
      "step": 4075,
      "time_per_iteration": 3.6278598308563232
    },
    {
      "auxiliary_loss_clip": 0.01185454,
      "auxiliary_loss_mlp": 0.01029046,
      "balance_loss_clip": 1.05499268,
      "balance_loss_mlp": 1.01962209,
      "epoch": 0.4901100222449348,
      "flos": 20339588856960.0,
      "grad_norm": 1.975366864423833,
      "language_loss": 0.80808485,
      "learning_rate": 2.161072499085629e-06,
      "loss": 0.83022988,
      "num_input_tokens_seen": 87789495,
      "step": 4076,
      "time_per_iteration": 3.5910558700561523
    },
    {
      "auxiliary_loss_clip": 0.01186067,
      "auxiliary_loss_mlp": 0.01036186,
      "balance_loss_clip": 1.05768108,
      "balance_loss_mlp": 1.02806783,
      "epoch": 0.4902302651355739,
      "flos": 30446359384320.0,
      "grad_norm": 1.789098550585597,
      "language_loss": 0.83199137,
      "learning_rate": 2.160296045589487e-06,
      "loss": 0.85421389,
      "num_input_tokens_seen": 87812955,
      "step": 4077,
      "time_per_iteration": 2.893203020095825
    },
    {
      "auxiliary_loss_clip": 0.01186358,
      "auxiliary_loss_mlp": 0.010316,
      "balance_loss_clip": 1.05368543,
      "balance_loss_mlp": 1.02259958,
      "epoch": 0.49035050802621294,
      "flos": 19174180089600.0,
      "grad_norm": 2.6698907524642945,
      "language_loss": 0.70164466,
      "learning_rate": 2.159519567776562e-06,
      "loss": 0.7238242,
      "num_input_tokens_seen": 87832605,
      "step": 4078,
      "time_per_iteration": 2.595116376876831
    },
    {
      "auxiliary_loss_clip": 0.01187154,
      "auxiliary_loss_mlp": 0.0103222,
      "balance_loss_clip": 1.05457711,
      "balance_loss_mlp": 1.02306461,
      "epoch": 0.49047075091685205,
      "flos": 22228489365120.0,
      "grad_norm": 2.793808310476938,
      "language_loss": 0.70593703,
      "learning_rate": 2.1587430657646463e-06,
      "loss": 0.72813082,
      "num_input_tokens_seen": 87846040,
      "step": 4079,
      "time_per_iteration": 2.7271435260772705
    },
    {
      "auxiliary_loss_clip": 0.01184875,
      "auxiliary_loss_mlp": 0.01030474,
      "balance_loss_clip": 1.05497074,
      "balance_loss_mlp": 1.02161694,
      "epoch": 0.4905909938074911,
      "flos": 20156516213760.0,
      "grad_norm": 3.692731983794278,
      "language_loss": 0.77825236,
      "learning_rate": 2.157966539671533e-06,
      "loss": 0.80040586,
      "num_input_tokens_seen": 87865680,
      "step": 4080,
      "time_per_iteration": 2.623281717300415
    },
    {
      "auxiliary_loss_clip": 0.01181194,
      "auxiliary_loss_mlp": 0.01024057,
      "balance_loss_clip": 1.05343103,
      "balance_loss_mlp": 1.01575375,
      "epoch": 0.4907112366981302,
      "flos": 17202217380480.0,
      "grad_norm": 1.9643967507806939,
      "language_loss": 0.67172521,
      "learning_rate": 2.157189989615021e-06,
      "loss": 0.69377768,
      "num_input_tokens_seen": 87884270,
      "step": 4081,
      "time_per_iteration": 2.6085853576660156
    },
    {
      "auxiliary_loss_clip": 0.01189904,
      "auxiliary_loss_mlp": 0.01027633,
      "balance_loss_clip": 1.05679095,
      "balance_loss_mlp": 1.01813817,
      "epoch": 0.4908314795887693,
      "flos": 21688968107520.0,
      "grad_norm": 17.44391290500215,
      "language_loss": 0.7495091,
      "learning_rate": 2.156413415712913e-06,
      "loss": 0.77168441,
      "num_input_tokens_seen": 87906320,
      "step": 4082,
      "time_per_iteration": 2.625556468963623
    },
    {
      "auxiliary_loss_clip": 0.0118672,
      "auxiliary_loss_mlp": 0.01021715,
      "balance_loss_clip": 1.05540371,
      "balance_loss_mlp": 1.01303053,
      "epoch": 0.4909517224794084,
      "flos": 26213676531840.0,
      "grad_norm": 2.6536226089041284,
      "language_loss": 0.784096,
      "learning_rate": 2.155636818083014e-06,
      "loss": 0.80618036,
      "num_input_tokens_seen": 87927690,
      "step": 4083,
      "time_per_iteration": 2.6546945571899414
    },
    {
      "auxiliary_loss_clip": 0.01182691,
      "auxiliary_loss_mlp": 0.01021972,
      "balance_loss_clip": 1.05440807,
      "balance_loss_mlp": 1.01402688,
      "epoch": 0.4910719653700475,
      "flos": 23148377694720.0,
      "grad_norm": 2.191119774827125,
      "language_loss": 0.84567332,
      "learning_rate": 2.154860196843134e-06,
      "loss": 0.86772001,
      "num_input_tokens_seen": 87946885,
      "step": 4084,
      "time_per_iteration": 2.619154453277588
    },
    {
      "auxiliary_loss_clip": 0.01185383,
      "auxiliary_loss_mlp": 0.01033174,
      "balance_loss_clip": 1.05400372,
      "balance_loss_mlp": 1.02465034,
      "epoch": 0.4911922082606866,
      "flos": 23331845387520.0,
      "grad_norm": 2.105844446775019,
      "language_loss": 0.76851356,
      "learning_rate": 2.154083552111085e-06,
      "loss": 0.79069912,
      "num_input_tokens_seen": 87966055,
      "step": 4085,
      "time_per_iteration": 2.6981279850006104
    },
    {
      "auxiliary_loss_clip": 0.01183706,
      "auxiliary_loss_mlp": 0.01026621,
      "balance_loss_clip": 1.05159497,
      "balance_loss_mlp": 1.01736474,
      "epoch": 0.49131245115132566,
      "flos": 29203239542400.0,
      "grad_norm": 3.302096939571365,
      "language_loss": 0.81880867,
      "learning_rate": 2.1533068840046834e-06,
      "loss": 0.84091198,
      "num_input_tokens_seen": 87986320,
      "step": 4086,
      "time_per_iteration": 2.6434237957000732
    },
    {
      "auxiliary_loss_clip": 0.01185738,
      "auxiliary_loss_mlp": 0.01026218,
      "balance_loss_clip": 1.05561721,
      "balance_loss_mlp": 1.01745033,
      "epoch": 0.49143269404196477,
      "flos": 20147465986560.0,
      "grad_norm": 2.6512836937616058,
      "language_loss": 0.61623305,
      "learning_rate": 2.152530192641749e-06,
      "loss": 0.63835257,
      "num_input_tokens_seen": 88001230,
      "step": 4087,
      "time_per_iteration": 2.6707284450531006
    },
    {
      "auxiliary_loss_clip": 0.01182184,
      "auxiliary_loss_mlp": 0.01029623,
      "balance_loss_clip": 1.05260837,
      "balance_loss_mlp": 1.02042031,
      "epoch": 0.4915529369326039,
      "flos": 24389809597440.0,
      "grad_norm": 2.267375031588598,
      "language_loss": 0.72226381,
      "learning_rate": 2.1517534781401068e-06,
      "loss": 0.7443819,
      "num_input_tokens_seen": 88019110,
      "step": 4088,
      "time_per_iteration": 2.630150556564331
    },
    {
      "auxiliary_loss_clip": 0.01185344,
      "auxiliary_loss_mlp": 0.01031499,
      "balance_loss_clip": 1.05481243,
      "balance_loss_mlp": 1.02279079,
      "epoch": 0.49167317982324293,
      "flos": 10524305197440.0,
      "grad_norm": 2.285534330296271,
      "language_loss": 0.69344497,
      "learning_rate": 2.150976740617581e-06,
      "loss": 0.71561342,
      "num_input_tokens_seen": 88035670,
      "step": 4089,
      "time_per_iteration": 2.6041195392608643
    },
    {
      "auxiliary_loss_clip": 0.01184954,
      "auxiliary_loss_mlp": 0.01025121,
      "balance_loss_clip": 1.05480981,
      "balance_loss_mlp": 1.01652646,
      "epoch": 0.49179342271388204,
      "flos": 25593427457280.0,
      "grad_norm": 2.405631794545671,
      "language_loss": 0.71404022,
      "learning_rate": 2.150199980192006e-06,
      "loss": 0.73614097,
      "num_input_tokens_seen": 88054790,
      "step": 4090,
      "time_per_iteration": 2.6629762649536133
    },
    {
      "auxiliary_loss_clip": 0.01183435,
      "auxiliary_loss_mlp": 0.01024508,
      "balance_loss_clip": 1.05410111,
      "balance_loss_mlp": 1.01638436,
      "epoch": 0.49191366560452116,
      "flos": 21102043875840.0,
      "grad_norm": 1.7298091075625275,
      "language_loss": 0.80698174,
      "learning_rate": 2.1494231969812114e-06,
      "loss": 0.82906115,
      "num_input_tokens_seen": 88073780,
      "step": 4091,
      "time_per_iteration": 2.63332462310791
    },
    {
      "auxiliary_loss_clip": 0.01186458,
      "auxiliary_loss_mlp": 0.01032631,
      "balance_loss_clip": 1.05690277,
      "balance_loss_mlp": 1.02324343,
      "epoch": 0.4920339084951602,
      "flos": 26067520091520.0,
      "grad_norm": 2.413748190721538,
      "language_loss": 0.81036961,
      "learning_rate": 2.1486463911030372e-06,
      "loss": 0.83256054,
      "num_input_tokens_seen": 88094430,
      "step": 4092,
      "time_per_iteration": 2.6588757038116455
    },
    {
      "auxiliary_loss_clip": 0.01183749,
      "auxiliary_loss_mlp": 0.01028881,
      "balance_loss_clip": 1.05353403,
      "balance_loss_mlp": 1.02072716,
      "epoch": 0.4921541513857993,
      "flos": 25081269384960.0,
      "grad_norm": 2.201267560628401,
      "language_loss": 0.7453351,
      "learning_rate": 2.147869562675324e-06,
      "loss": 0.76746136,
      "num_input_tokens_seen": 88113400,
      "step": 4093,
      "time_per_iteration": 2.6940228939056396
    },
    {
      "auxiliary_loss_clip": 0.01185565,
      "auxiliary_loss_mlp": 0.01033271,
      "balance_loss_clip": 1.05248475,
      "balance_loss_mlp": 1.02307248,
      "epoch": 0.49227439427643843,
      "flos": 24389809597440.0,
      "grad_norm": 1.7847778832574932,
      "language_loss": 0.72445649,
      "learning_rate": 2.147092711815915e-06,
      "loss": 0.74664485,
      "num_input_tokens_seen": 88132750,
      "step": 4094,
      "time_per_iteration": 2.7263436317443848
    },
    {
      "auxiliary_loss_clip": 0.01186312,
      "auxiliary_loss_mlp": 0.01027796,
      "balance_loss_clip": 1.05509341,
      "balance_loss_mlp": 1.01878977,
      "epoch": 0.4923946371670775,
      "flos": 11363753018880.0,
      "grad_norm": 2.6013310754391195,
      "language_loss": 0.86011267,
      "learning_rate": 2.1463158386426593e-06,
      "loss": 0.88225377,
      "num_input_tokens_seen": 88150560,
      "step": 4095,
      "time_per_iteration": 2.607175350189209
    },
    {
      "auxiliary_loss_clip": 0.01188133,
      "auxiliary_loss_mlp": 0.01032968,
      "balance_loss_clip": 1.05556178,
      "balance_loss_mlp": 1.02348495,
      "epoch": 0.4925148800577166,
      "flos": 30445964334720.0,
      "grad_norm": 1.949751131900783,
      "language_loss": 0.77437729,
      "learning_rate": 2.145538943273407e-06,
      "loss": 0.7965883,
      "num_input_tokens_seen": 88170835,
      "step": 4096,
      "time_per_iteration": 3.6471095085144043
    },
    {
      "auxiliary_loss_clip": 0.011887,
      "auxiliary_loss_mlp": 0.0102958,
      "balance_loss_clip": 1.05771255,
      "balance_loss_mlp": 1.02013314,
      "epoch": 0.49263512294835565,
      "flos": 20850454039680.0,
      "grad_norm": 2.3873743968333976,
      "language_loss": 0.71877033,
      "learning_rate": 2.144762025826013e-06,
      "loss": 0.74095309,
      "num_input_tokens_seen": 88189925,
      "step": 4097,
      "time_per_iteration": 2.628052234649658
    },
    {
      "auxiliary_loss_clip": 0.01182995,
      "auxiliary_loss_mlp": 0.01032291,
      "balance_loss_clip": 1.0528357,
      "balance_loss_mlp": 1.02268875,
      "epoch": 0.49275536583899476,
      "flos": 23767477534080.0,
      "grad_norm": 2.2185277060472073,
      "language_loss": 0.86782348,
      "learning_rate": 2.143985086418334e-06,
      "loss": 0.88997626,
      "num_input_tokens_seen": 88205105,
      "step": 4098,
      "time_per_iteration": 2.6295621395111084
    },
    {
      "auxiliary_loss_clip": 0.01181719,
      "auxiliary_loss_mlp": 0.01027627,
      "balance_loss_clip": 1.05408597,
      "balance_loss_mlp": 1.01962829,
      "epoch": 0.4928756087296339,
      "flos": 22273522041600.0,
      "grad_norm": 1.4900440362422995,
      "language_loss": 0.76617539,
      "learning_rate": 2.1432081251682324e-06,
      "loss": 0.7882688,
      "num_input_tokens_seen": 88225475,
      "step": 4099,
      "time_per_iteration": 3.549997568130493
    },
    {
      "auxiliary_loss_clip": 0.0118592,
      "auxiliary_loss_mlp": 0.01034078,
      "balance_loss_clip": 1.05623388,
      "balance_loss_mlp": 1.02576947,
      "epoch": 0.49299585162027293,
      "flos": 19645471463040.0,
      "grad_norm": 1.6540284886098684,
      "language_loss": 0.8722111,
      "learning_rate": 2.142431142193572e-06,
      "loss": 0.89441103,
      "num_input_tokens_seen": 88243255,
      "step": 4100,
      "time_per_iteration": 2.594496965408325
    },
    {
      "auxiliary_loss_clip": 0.01183792,
      "auxiliary_loss_mlp": 0.01023221,
      "balance_loss_clip": 1.05542064,
      "balance_loss_mlp": 1.0150702,
      "epoch": 0.49311609451091204,
      "flos": 38837138497920.0,
      "grad_norm": 3.1360893922271624,
      "language_loss": 0.71887827,
      "learning_rate": 2.1416541376122207e-06,
      "loss": 0.74094844,
      "num_input_tokens_seen": 88263435,
      "step": 4101,
      "time_per_iteration": 3.631092071533203
    },
    {
      "auxiliary_loss_clip": 0.01182079,
      "auxiliary_loss_mlp": 0.01031724,
      "balance_loss_clip": 1.05118418,
      "balance_loss_mlp": 1.02275395,
      "epoch": 0.49323633740155115,
      "flos": 28329102161280.0,
      "grad_norm": 2.276690848976027,
      "language_loss": 0.73061132,
      "learning_rate": 2.1408771115420496e-06,
      "loss": 0.75274932,
      "num_input_tokens_seen": 88283295,
      "step": 4102,
      "time_per_iteration": 2.646578550338745
    },
    {
      "auxiliary_loss_clip": 0.01186791,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.05607796,
      "balance_loss_mlp": 1.01826584,
      "epoch": 0.4933565802921902,
      "flos": 21135584200320.0,
      "grad_norm": 145.92702448997434,
      "language_loss": 0.64764661,
      "learning_rate": 2.140100064100932e-06,
      "loss": 0.66978276,
      "num_input_tokens_seen": 88299270,
      "step": 4103,
      "time_per_iteration": 3.448239326477051
    },
    {
      "auxiliary_loss_clip": 0.01184011,
      "auxiliary_loss_mlp": 0.01029762,
      "balance_loss_clip": 1.05385208,
      "balance_loss_mlp": 1.02120256,
      "epoch": 0.4934768231828293,
      "flos": 18039007595520.0,
      "grad_norm": 2.074258770834796,
      "language_loss": 0.75867337,
      "learning_rate": 2.139322995406746e-06,
      "loss": 0.78081113,
      "num_input_tokens_seen": 88316905,
      "step": 4104,
      "time_per_iteration": 2.621829032897949
    },
    {
      "auxiliary_loss_clip": 0.01185932,
      "auxiliary_loss_mlp": 0.01027678,
      "balance_loss_clip": 1.05667365,
      "balance_loss_mlp": 1.01872575,
      "epoch": 0.4935970660734684,
      "flos": 23469957181440.0,
      "grad_norm": 2.1363306343265793,
      "language_loss": 0.80020696,
      "learning_rate": 2.1385459055773727e-06,
      "loss": 0.82234299,
      "num_input_tokens_seen": 88335095,
      "step": 4105,
      "time_per_iteration": 2.5697152614593506
    },
    {
      "auxiliary_loss_clip": 0.0118447,
      "auxiliary_loss_mlp": 0.01025196,
      "balance_loss_clip": 1.05741191,
      "balance_loss_mlp": 1.01747704,
      "epoch": 0.4937173089641075,
      "flos": 64479258840960.0,
      "grad_norm": 2.165512225297504,
      "language_loss": 0.73737085,
      "learning_rate": 2.137768794730696e-06,
      "loss": 0.75946748,
      "num_input_tokens_seen": 88358545,
      "step": 4106,
      "time_per_iteration": 3.046872854232788
    },
    {
      "auxiliary_loss_clip": 0.01190037,
      "auxiliary_loss_mlp": 0.01027702,
      "balance_loss_clip": 1.05779648,
      "balance_loss_mlp": 1.01812971,
      "epoch": 0.4938375518547466,
      "flos": 22346025644160.0,
      "grad_norm": 1.8770906250963786,
      "language_loss": 0.80397636,
      "learning_rate": 2.1369916629846026e-06,
      "loss": 0.82615376,
      "num_input_tokens_seen": 88378295,
      "step": 4107,
      "time_per_iteration": 2.6439578533172607
    },
    {
      "auxiliary_loss_clip": 0.01182004,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.05332565,
      "balance_loss_mlp": 1.01623487,
      "epoch": 0.4939577947453857,
      "flos": 17858700299520.0,
      "grad_norm": 2.9690788378403634,
      "language_loss": 0.75254732,
      "learning_rate": 2.136214510456983e-06,
      "loss": 0.77461946,
      "num_input_tokens_seen": 88396750,
      "step": 4108,
      "time_per_iteration": 2.594273090362549
    },
    {
      "auxiliary_loss_clip": 0.01083456,
      "auxiliary_loss_mlp": 0.01002214,
      "balance_loss_clip": 1.0192827,
      "balance_loss_mlp": 1.00092649,
      "epoch": 0.49407803763602476,
      "flos": 70066746875520.0,
      "grad_norm": 0.9073293495400417,
      "language_loss": 0.63150412,
      "learning_rate": 2.1354373372657296e-06,
      "loss": 0.65236086,
      "num_input_tokens_seen": 88455190,
      "step": 4109,
      "time_per_iteration": 3.1662545204162598
    },
    {
      "auxiliary_loss_clip": 0.01184005,
      "auxiliary_loss_mlp": 0.01029919,
      "balance_loss_clip": 1.05549514,
      "balance_loss_mlp": 1.02065599,
      "epoch": 0.49419828052666387,
      "flos": 24317485562880.0,
      "grad_norm": 1.527961952977814,
      "language_loss": 0.70999825,
      "learning_rate": 2.1346601435287404e-06,
      "loss": 0.73213744,
      "num_input_tokens_seen": 88477460,
      "step": 4110,
      "time_per_iteration": 2.6475656032562256
    },
    {
      "auxiliary_loss_clip": 0.01184163,
      "auxiliary_loss_mlp": 0.01027819,
      "balance_loss_clip": 1.05481148,
      "balance_loss_mlp": 1.01919436,
      "epoch": 0.494318523417303,
      "flos": 29386060790400.0,
      "grad_norm": 5.465118408704743,
      "language_loss": 0.8026098,
      "learning_rate": 2.1338829293639144e-06,
      "loss": 0.82472962,
      "num_input_tokens_seen": 88497820,
      "step": 4111,
      "time_per_iteration": 2.691612958908081
    },
    {
      "auxiliary_loss_clip": 0.01184135,
      "auxiliary_loss_mlp": 0.0102526,
      "balance_loss_clip": 1.05331194,
      "balance_loss_mlp": 1.01645684,
      "epoch": 0.49443876630794203,
      "flos": 15268284195840.0,
      "grad_norm": 2.4982232377062363,
      "language_loss": 0.83011937,
      "learning_rate": 2.1331056948891547e-06,
      "loss": 0.85221332,
      "num_input_tokens_seen": 88514920,
      "step": 4112,
      "time_per_iteration": 2.627516746520996
    },
    {
      "auxiliary_loss_clip": 0.01183542,
      "auxiliary_loss_mlp": 0.01025709,
      "balance_loss_clip": 1.05269647,
      "balance_loss_mlp": 1.0167799,
      "epoch": 0.49455900919858115,
      "flos": 12347453859840.0,
      "grad_norm": 2.9792971355902713,
      "language_loss": 0.76501536,
      "learning_rate": 2.1323284402223666e-06,
      "loss": 0.78710783,
      "num_input_tokens_seen": 88530910,
      "step": 4113,
      "time_per_iteration": 2.6805591583251953
    },
    {
      "auxiliary_loss_clip": 0.01184313,
      "auxiliary_loss_mlp": 0.01029767,
      "balance_loss_clip": 1.05906796,
      "balance_loss_mlp": 1.02174377,
      "epoch": 0.4946792520892202,
      "flos": 22779610715520.0,
      "grad_norm": 1.953025188981165,
      "language_loss": 0.87860936,
      "learning_rate": 2.1315511654814597e-06,
      "loss": 0.90075016,
      "num_input_tokens_seen": 88549320,
      "step": 4114,
      "time_per_iteration": 2.6213817596435547
    },
    {
      "auxiliary_loss_clip": 0.01182099,
      "auxiliary_loss_mlp": 0.01028015,
      "balance_loss_clip": 1.0541048,
      "balance_loss_mlp": 1.01992655,
      "epoch": 0.4947994949798593,
      "flos": 23148126299520.0,
      "grad_norm": 2.515762229338354,
      "language_loss": 0.78246659,
      "learning_rate": 2.1307738707843456e-06,
      "loss": 0.80456769,
      "num_input_tokens_seen": 88568985,
      "step": 4115,
      "time_per_iteration": 2.6811766624450684
    },
    {
      "auxiliary_loss_clip": 0.01190601,
      "auxiliary_loss_mlp": 0.01033202,
      "balance_loss_clip": 1.05923736,
      "balance_loss_mlp": 1.02385557,
      "epoch": 0.4949197378704984,
      "flos": 23659997063040.0,
      "grad_norm": 4.919894660649589,
      "language_loss": 0.68724215,
      "learning_rate": 2.1299965562489385e-06,
      "loss": 0.70948017,
      "num_input_tokens_seen": 88588790,
      "step": 4116,
      "time_per_iteration": 2.6206274032592773
    },
    {
      "auxiliary_loss_clip": 0.01187589,
      "auxiliary_loss_mlp": 0.01027711,
      "balance_loss_clip": 1.05711985,
      "balance_loss_mlp": 1.01847816,
      "epoch": 0.4950399807611375,
      "flos": 26911493026560.0,
      "grad_norm": 1.6155470951418769,
      "language_loss": 0.78960776,
      "learning_rate": 2.129219221993158e-06,
      "loss": 0.81176078,
      "num_input_tokens_seen": 88613575,
      "step": 4117,
      "time_per_iteration": 2.729682683944702
    },
    {
      "auxiliary_loss_clip": 0.01084133,
      "auxiliary_loss_mlp": 0.01002736,
      "balance_loss_clip": 1.01953316,
      "balance_loss_mlp": 1.00148416,
      "epoch": 0.4951602236517766,
      "flos": 67315270187520.0,
      "grad_norm": 0.793237249859253,
      "language_loss": 0.59986305,
      "learning_rate": 2.128441868134924e-06,
      "loss": 0.62073171,
      "num_input_tokens_seen": 88675510,
      "step": 4118,
      "time_per_iteration": 3.214071273803711
    },
    {
      "auxiliary_loss_clip": 0.01181639,
      "auxiliary_loss_mlp": 0.01029335,
      "balance_loss_clip": 1.0538218,
      "balance_loss_mlp": 1.02069855,
      "epoch": 0.4952804665424157,
      "flos": 19901442758400.0,
      "grad_norm": 2.325195316673003,
      "language_loss": 0.82401454,
      "learning_rate": 2.1276644947921606e-06,
      "loss": 0.84612423,
      "num_input_tokens_seen": 88694425,
      "step": 4119,
      "time_per_iteration": 2.6439414024353027
    },
    {
      "auxiliary_loss_clip": 0.01185698,
      "auxiliary_loss_mlp": 0.01030437,
      "balance_loss_clip": 1.05454659,
      "balance_loss_mlp": 1.0215441,
      "epoch": 0.49540070943305475,
      "flos": 18806813740800.0,
      "grad_norm": 2.5509462716122733,
      "language_loss": 0.82447773,
      "learning_rate": 2.126887102082795e-06,
      "loss": 0.84663904,
      "num_input_tokens_seen": 88714450,
      "step": 4120,
      "time_per_iteration": 2.609943151473999
    },
    {
      "auxiliary_loss_clip": 0.01183209,
      "auxiliary_loss_mlp": 0.01030567,
      "balance_loss_clip": 1.05452967,
      "balance_loss_mlp": 1.02209735,
      "epoch": 0.49552095232369386,
      "flos": 24934179191040.0,
      "grad_norm": 1.907563240822655,
      "language_loss": 0.70379865,
      "learning_rate": 2.126109690124757e-06,
      "loss": 0.72593641,
      "num_input_tokens_seen": 88735265,
      "step": 4121,
      "time_per_iteration": 2.7045037746429443
    },
    {
      "auxiliary_loss_clip": 0.0118464,
      "auxiliary_loss_mlp": 0.01026926,
      "balance_loss_clip": 1.05399835,
      "balance_loss_mlp": 1.01826584,
      "epoch": 0.495641195214333,
      "flos": 22857249962880.0,
      "grad_norm": 1.9371903221142945,
      "language_loss": 0.71233547,
      "learning_rate": 2.1253322590359786e-06,
      "loss": 0.73445106,
      "num_input_tokens_seen": 88754600,
      "step": 4122,
      "time_per_iteration": 3.631962299346924
    },
    {
      "auxiliary_loss_clip": 0.01185388,
      "auxiliary_loss_mlp": 0.01032257,
      "balance_loss_clip": 1.05536509,
      "balance_loss_mlp": 1.02387071,
      "epoch": 0.49576143810497203,
      "flos": 25769748343680.0,
      "grad_norm": 3.0837606209197865,
      "language_loss": 0.74250686,
      "learning_rate": 2.124554808934397e-06,
      "loss": 0.76468337,
      "num_input_tokens_seen": 88775180,
      "step": 4123,
      "time_per_iteration": 2.7060649394989014
    },
    {
      "auxiliary_loss_clip": 0.01187543,
      "auxiliary_loss_mlp": 0.01025069,
      "balance_loss_clip": 1.0578413,
      "balance_loss_mlp": 1.01651597,
      "epoch": 0.49588168099561114,
      "flos": 22128838058880.0,
      "grad_norm": 2.2806921074156428,
      "language_loss": 0.73219121,
      "learning_rate": 2.1237773399379496e-06,
      "loss": 0.75431728,
      "num_input_tokens_seen": 88796145,
      "step": 4124,
      "time_per_iteration": 2.7097718715667725
    },
    {
      "auxiliary_loss_clip": 0.01184422,
      "auxiliary_loss_mlp": 0.01024611,
      "balance_loss_clip": 1.05192554,
      "balance_loss_mlp": 1.01533091,
      "epoch": 0.49600192388625025,
      "flos": 24387331559040.0,
      "grad_norm": 2.0567084834709544,
      "language_loss": 0.86959857,
      "learning_rate": 2.122999852164578e-06,
      "loss": 0.89168894,
      "num_input_tokens_seen": 88816765,
      "step": 4125,
      "time_per_iteration": 2.666076421737671
    },
    {
      "auxiliary_loss_clip": 0.01189436,
      "auxiliary_loss_mlp": 0.01031166,
      "balance_loss_clip": 1.05786049,
      "balance_loss_mlp": 1.02181363,
      "epoch": 0.4961221667768893,
      "flos": 22857429530880.0,
      "grad_norm": 2.736933554304298,
      "language_loss": 0.58039093,
      "learning_rate": 2.122222345732227e-06,
      "loss": 0.60259694,
      "num_input_tokens_seen": 88836680,
      "step": 4126,
      "time_per_iteration": 2.6619365215301514
    },
    {
      "auxiliary_loss_clip": 0.01185887,
      "auxiliary_loss_mlp": 0.0102927,
      "balance_loss_clip": 1.05480325,
      "balance_loss_mlp": 1.0201329,
      "epoch": 0.4962424096675284,
      "flos": 17858089768320.0,
      "grad_norm": 1.8386219672696447,
      "language_loss": 0.83060282,
      "learning_rate": 2.121444820758843e-06,
      "loss": 0.85275441,
      "num_input_tokens_seen": 88855320,
      "step": 4127,
      "time_per_iteration": 4.447176456451416
    },
    {
      "auxiliary_loss_clip": 0.01184583,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.05360162,
      "balance_loss_mlp": 1.02018738,
      "epoch": 0.49636265255816747,
      "flos": 21793611404160.0,
      "grad_norm": 2.4265399113790527,
      "language_loss": 0.78833884,
      "learning_rate": 2.120667277362376e-06,
      "loss": 0.81048065,
      "num_input_tokens_seen": 88874035,
      "step": 4128,
      "time_per_iteration": 2.6058075428009033
    },
    {
      "auxiliary_loss_clip": 0.01187212,
      "auxiliary_loss_mlp": 0.01031555,
      "balance_loss_clip": 1.05709815,
      "balance_loss_mlp": 1.02270401,
      "epoch": 0.4964828954488066,
      "flos": 16358603581440.0,
      "grad_norm": 2.5260085865555704,
      "language_loss": 0.84941995,
      "learning_rate": 2.1198897156607796e-06,
      "loss": 0.87160766,
      "num_input_tokens_seen": 88891390,
      "step": 4129,
      "time_per_iteration": 2.6020877361297607
    },
    {
      "auxiliary_loss_clip": 0.01185551,
      "auxiliary_loss_mlp": 0.01033192,
      "balance_loss_clip": 1.05482221,
      "balance_loss_mlp": 1.02382815,
      "epoch": 0.4966031383394457,
      "flos": 24711101775360.0,
      "grad_norm": 2.0116927352928258,
      "language_loss": 0.73630059,
      "learning_rate": 2.1191121357720085e-06,
      "loss": 0.758488,
      "num_input_tokens_seen": 88909450,
      "step": 4130,
      "time_per_iteration": 3.5939724445343018
    },
    {
      "auxiliary_loss_clip": 0.01184384,
      "auxiliary_loss_mlp": 0.01026088,
      "balance_loss_clip": 1.05411863,
      "balance_loss_mlp": 1.01617551,
      "epoch": 0.49672338123008475,
      "flos": 22930615491840.0,
      "grad_norm": 2.527179458412612,
      "language_loss": 0.74802011,
      "learning_rate": 2.1183345378140206e-06,
      "loss": 0.77012485,
      "num_input_tokens_seen": 88929195,
      "step": 4131,
      "time_per_iteration": 2.6605076789855957
    },
    {
      "auxiliary_loss_clip": 0.01085904,
      "auxiliary_loss_mlp": 0.01003801,
      "balance_loss_clip": 1.0198257,
      "balance_loss_mlp": 1.00250173,
      "epoch": 0.49684362412072386,
      "flos": 65976736844160.0,
      "grad_norm": 0.8566817517960529,
      "language_loss": 0.62002075,
      "learning_rate": 2.1175569219047783e-06,
      "loss": 0.64091778,
      "num_input_tokens_seen": 88990635,
      "step": 4132,
      "time_per_iteration": 3.2981526851654053
    },
    {
      "auxiliary_loss_clip": 0.01184462,
      "auxiliary_loss_mlp": 0.01027108,
      "balance_loss_clip": 1.05444276,
      "balance_loss_mlp": 1.01809549,
      "epoch": 0.49696386701136297,
      "flos": 19971288754560.0,
      "grad_norm": 3.1298252216485567,
      "language_loss": 0.73452723,
      "learning_rate": 2.1167792881622437e-06,
      "loss": 0.75664294,
      "num_input_tokens_seen": 89009655,
      "step": 4133,
      "time_per_iteration": 2.590299129486084
    },
    {
      "auxiliary_loss_clip": 0.01184201,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.05398893,
      "balance_loss_mlp": 1.01756382,
      "epoch": 0.497084109902002,
      "flos": 24750819239040.0,
      "grad_norm": 1.7032378159705683,
      "language_loss": 0.8101415,
      "learning_rate": 2.116001636704384e-06,
      "loss": 0.83224511,
      "num_input_tokens_seen": 89030040,
      "step": 4134,
      "time_per_iteration": 2.7474944591522217
    },
    {
      "auxiliary_loss_clip": 0.01186606,
      "auxiliary_loss_mlp": 0.01032366,
      "balance_loss_clip": 1.05715084,
      "balance_loss_mlp": 1.02381849,
      "epoch": 0.49720435279264114,
      "flos": 21871825269120.0,
      "grad_norm": 3.3187336120583426,
      "language_loss": 0.80024034,
      "learning_rate": 2.1152239676491685e-06,
      "loss": 0.82243007,
      "num_input_tokens_seen": 89048145,
      "step": 4135,
      "time_per_iteration": 2.6116278171539307
    },
    {
      "auxiliary_loss_clip": 0.01183106,
      "auxiliary_loss_mlp": 0.01025306,
      "balance_loss_clip": 1.05330133,
      "balance_loss_mlp": 1.01643109,
      "epoch": 0.49732459568328025,
      "flos": 23805794367360.0,
      "grad_norm": 1.9246778657051238,
      "language_loss": 0.73495084,
      "learning_rate": 2.114446281114569e-06,
      "loss": 0.75703496,
      "num_input_tokens_seen": 89067165,
      "step": 4136,
      "time_per_iteration": 2.732290029525757
    },
    {
      "auxiliary_loss_clip": 0.01184767,
      "auxiliary_loss_mlp": 0.0102651,
      "balance_loss_clip": 1.05676484,
      "balance_loss_mlp": 1.01764727,
      "epoch": 0.4974448385739193,
      "flos": 20047742853120.0,
      "grad_norm": 1.847943431935048,
      "language_loss": 0.76346755,
      "learning_rate": 2.1136685772185587e-06,
      "loss": 0.78558034,
      "num_input_tokens_seen": 89086190,
      "step": 4137,
      "time_per_iteration": 2.631620168685913
    },
    {
      "auxiliary_loss_clip": 0.01185744,
      "auxiliary_loss_mlp": 0.01027782,
      "balance_loss_clip": 1.05345845,
      "balance_loss_mlp": 1.01865029,
      "epoch": 0.4975650814645584,
      "flos": 24821347593600.0,
      "grad_norm": 2.2448082530522377,
      "language_loss": 0.77998763,
      "learning_rate": 2.1128908560791163e-06,
      "loss": 0.80212289,
      "num_input_tokens_seen": 89106020,
      "step": 4138,
      "time_per_iteration": 2.6649913787841797
    },
    {
      "auxiliary_loss_clip": 0.01184188,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.05521488,
      "balance_loss_mlp": 1.01675534,
      "epoch": 0.4976853243551975,
      "flos": 19829477859840.0,
      "grad_norm": 2.2339594666579288,
      "language_loss": 0.78073204,
      "learning_rate": 2.1121131178142203e-06,
      "loss": 0.80282736,
      "num_input_tokens_seen": 89125385,
      "step": 4139,
      "time_per_iteration": 2.5822842121124268
    },
    {
      "auxiliary_loss_clip": 0.01181066,
      "auxiliary_loss_mlp": 0.01027455,
      "balance_loss_clip": 1.05200911,
      "balance_loss_mlp": 1.01899087,
      "epoch": 0.4978055672458366,
      "flos": 23142990654720.0,
      "grad_norm": 3.863941238480939,
      "language_loss": 0.82510519,
      "learning_rate": 2.1113353625418544e-06,
      "loss": 0.84719038,
      "num_input_tokens_seen": 89143935,
      "step": 4140,
      "time_per_iteration": 2.6458401679992676
    },
    {
      "auxiliary_loss_clip": 0.01180735,
      "auxiliary_loss_mlp": 0.0103294,
      "balance_loss_clip": 1.05565727,
      "balance_loss_mlp": 1.02475095,
      "epoch": 0.4979258101364757,
      "flos": 15559914718080.0,
      "grad_norm": 1.7038771562767887,
      "language_loss": 0.79022884,
      "learning_rate": 2.1105575903800017e-06,
      "loss": 0.81236559,
      "num_input_tokens_seen": 89162655,
      "step": 4141,
      "time_per_iteration": 2.574063777923584
    },
    {
      "auxiliary_loss_clip": 0.01185206,
      "auxiliary_loss_mlp": 0.01027485,
      "balance_loss_clip": 1.05382597,
      "balance_loss_mlp": 1.01830626,
      "epoch": 0.4980460530271148,
      "flos": 26356169784960.0,
      "grad_norm": 2.2530164639909547,
      "language_loss": 0.85011756,
      "learning_rate": 2.1097798014466502e-06,
      "loss": 0.87224442,
      "num_input_tokens_seen": 89182255,
      "step": 4142,
      "time_per_iteration": 2.7008965015411377
    },
    {
      "auxiliary_loss_clip": 0.01185858,
      "auxiliary_loss_mlp": 0.01026578,
      "balance_loss_clip": 1.05396986,
      "balance_loss_mlp": 1.01762533,
      "epoch": 0.49816629591775385,
      "flos": 17274541415040.0,
      "grad_norm": 2.405463688851801,
      "language_loss": 0.59037399,
      "learning_rate": 2.109001995859791e-06,
      "loss": 0.61249828,
      "num_input_tokens_seen": 89201155,
      "step": 4143,
      "time_per_iteration": 2.5761005878448486
    },
    {
      "auxiliary_loss_clip": 0.01092025,
      "auxiliary_loss_mlp": 0.01005368,
      "balance_loss_clip": 1.02214193,
      "balance_loss_mlp": 1.00391996,
      "epoch": 0.49828653880839296,
      "flos": 64930947344640.0,
      "grad_norm": 0.8038991035236797,
      "language_loss": 0.60090578,
      "learning_rate": 2.108224173737415e-06,
      "loss": 0.6218797,
      "num_input_tokens_seen": 89264455,
      "step": 4144,
      "time_per_iteration": 3.201909065246582
    },
    {
      "auxiliary_loss_clip": 0.01184824,
      "auxiliary_loss_mlp": 0.010323,
      "balance_loss_clip": 1.05333161,
      "balance_loss_mlp": 1.02300143,
      "epoch": 0.498406781699032,
      "flos": 27484806003840.0,
      "grad_norm": 3.042622044352236,
      "language_loss": 0.75918436,
      "learning_rate": 2.1074463351975183e-06,
      "loss": 0.78135556,
      "num_input_tokens_seen": 89283340,
      "step": 4145,
      "time_per_iteration": 2.6779580116271973
    },
    {
      "auxiliary_loss_clip": 0.01182298,
      "auxiliary_loss_mlp": 0.01025469,
      "balance_loss_clip": 1.05319333,
      "balance_loss_mlp": 1.01686811,
      "epoch": 0.49852702458967113,
      "flos": 31499870307840.0,
      "grad_norm": 2.8785224776417158,
      "language_loss": 0.71466404,
      "learning_rate": 2.106668480358098e-06,
      "loss": 0.73674166,
      "num_input_tokens_seen": 89303565,
      "step": 4146,
      "time_per_iteration": 2.7706611156463623
    },
    {
      "auxiliary_loss_clip": 0.011864,
      "auxiliary_loss_mlp": 0.01026527,
      "balance_loss_clip": 1.05410337,
      "balance_loss_mlp": 1.01641226,
      "epoch": 0.49864726748031024,
      "flos": 22852868503680.0,
      "grad_norm": 2.271280853187627,
      "language_loss": 0.70604503,
      "learning_rate": 2.105890609337154e-06,
      "loss": 0.72817433,
      "num_input_tokens_seen": 89322080,
      "step": 4147,
      "time_per_iteration": 2.682781219482422
    },
    {
      "auxiliary_loss_clip": 0.01095725,
      "auxiliary_loss_mlp": 0.01003555,
      "balance_loss_clip": 1.02412271,
      "balance_loss_mlp": 1.0020833,
      "epoch": 0.4987675103709493,
      "flos": 70405708544640.0,
      "grad_norm": 0.7029401478259328,
      "language_loss": 0.63821077,
      "learning_rate": 2.1051127222526883e-06,
      "loss": 0.65920353,
      "num_input_tokens_seen": 89394195,
      "step": 4148,
      "time_per_iteration": 3.3079771995544434
    },
    {
      "auxiliary_loss_clip": 0.011866,
      "auxiliary_loss_mlp": 0.01025049,
      "balance_loss_clip": 1.0570308,
      "balance_loss_mlp": 1.01685905,
      "epoch": 0.4988877532615884,
      "flos": 28767571482240.0,
      "grad_norm": 1.6046942736203162,
      "language_loss": 0.80887264,
      "learning_rate": 2.1043348192227067e-06,
      "loss": 0.83098912,
      "num_input_tokens_seen": 89414565,
      "step": 4149,
      "time_per_iteration": 3.61830735206604
    },
    {
      "auxiliary_loss_clip": 0.01185453,
      "auxiliary_loss_mlp": 0.01027628,
      "balance_loss_clip": 1.05649257,
      "balance_loss_mlp": 1.01892531,
      "epoch": 0.4990079961522275,
      "flos": 16872700988160.0,
      "grad_norm": 1.907179784914694,
      "language_loss": 0.62100983,
      "learning_rate": 2.1035569003652156e-06,
      "loss": 0.64314061,
      "num_input_tokens_seen": 89433195,
      "step": 4150,
      "time_per_iteration": 2.6431491374969482
    },
    {
      "auxiliary_loss_clip": 0.01189479,
      "auxiliary_loss_mlp": 0.01033208,
      "balance_loss_clip": 1.05490232,
      "balance_loss_mlp": 1.02335525,
      "epoch": 0.4991282390428666,
      "flos": 13291042187520.0,
      "grad_norm": 2.7274877503386614,
      "language_loss": 0.81219733,
      "learning_rate": 2.1027789657982255e-06,
      "loss": 0.83442426,
      "num_input_tokens_seen": 89447410,
      "step": 4151,
      "time_per_iteration": 2.583574056625366
    },
    {
      "auxiliary_loss_clip": 0.01186704,
      "auxiliary_loss_mlp": 0.01033382,
      "balance_loss_clip": 1.05524158,
      "balance_loss_mlp": 1.02456021,
      "epoch": 0.4992484819335057,
      "flos": 21537496454400.0,
      "grad_norm": 1.9742069872746555,
      "language_loss": 0.77066398,
      "learning_rate": 2.1020010156397482e-06,
      "loss": 0.79286486,
      "num_input_tokens_seen": 89464630,
      "step": 4152,
      "time_per_iteration": 2.6672935485839844
    },
    {
      "auxiliary_loss_clip": 0.01185765,
      "auxiliary_loss_mlp": 0.01028907,
      "balance_loss_clip": 1.05501318,
      "balance_loss_mlp": 1.01959121,
      "epoch": 0.4993687248241448,
      "flos": 24860095390080.0,
      "grad_norm": 2.1317950846937426,
      "language_loss": 0.77328598,
      "learning_rate": 2.101223050007797e-06,
      "loss": 0.79543275,
      "num_input_tokens_seen": 89483180,
      "step": 4153,
      "time_per_iteration": 3.492973804473877
    },
    {
      "auxiliary_loss_clip": 0.01102942,
      "auxiliary_loss_mlp": 0.01006469,
      "balance_loss_clip": 1.02687502,
      "balance_loss_mlp": 1.00458574,
      "epoch": 0.49948896771478385,
      "flos": 62941602453120.0,
      "grad_norm": 0.8149864302708608,
      "language_loss": 0.53767538,
      "learning_rate": 2.1004450690203904e-06,
      "loss": 0.55876952,
      "num_input_tokens_seen": 89539260,
      "step": 4154,
      "time_per_iteration": 4.146761655807495
    },
    {
      "auxiliary_loss_clip": 0.01102563,
      "auxiliary_loss_mlp": 0.01006221,
      "balance_loss_clip": 1.02590632,
      "balance_loss_mlp": 1.00425363,
      "epoch": 0.49960921060542296,
      "flos": 68284213516800.0,
      "grad_norm": 0.8544101340586043,
      "language_loss": 0.63274729,
      "learning_rate": 2.099667072795546e-06,
      "loss": 0.65383518,
      "num_input_tokens_seen": 89601380,
      "step": 4155,
      "time_per_iteration": 3.2404987812042236
    },
    {
      "auxiliary_loss_clip": 0.01186837,
      "auxiliary_loss_mlp": 0.01029823,
      "balance_loss_clip": 1.05454183,
      "balance_loss_mlp": 1.02033353,
      "epoch": 0.49972945349606207,
      "flos": 23659350618240.0,
      "grad_norm": 4.695895960165325,
      "language_loss": 0.79674315,
      "learning_rate": 2.0988890614512864e-06,
      "loss": 0.81890976,
      "num_input_tokens_seen": 89621270,
      "step": 4156,
      "time_per_iteration": 3.639258861541748
    },
    {
      "auxiliary_loss_clip": 0.01189789,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.05791962,
      "balance_loss_mlp": 1.01552892,
      "epoch": 0.4998496963867011,
      "flos": 19755825022080.0,
      "grad_norm": 3.0585485954754894,
      "language_loss": 0.84251475,
      "learning_rate": 2.098111035105635e-06,
      "loss": 0.86466056,
      "num_input_tokens_seen": 89639695,
      "step": 4157,
      "time_per_iteration": 2.584723472595215
    },
    {
      "auxiliary_loss_clip": 0.01188886,
      "auxiliary_loss_mlp": 0.01032994,
      "balance_loss_clip": 1.05667257,
      "balance_loss_mlp": 1.02310514,
      "epoch": 0.49996993927734024,
      "flos": 22265728790400.0,
      "grad_norm": 1.9248747568937992,
      "language_loss": 0.73241955,
      "learning_rate": 2.0973329938766176e-06,
      "loss": 0.75463831,
      "num_input_tokens_seen": 89657125,
      "step": 4158,
      "time_per_iteration": 2.644535779953003
    },
    {
      "auxiliary_loss_clip": 0.01192944,
      "auxiliary_loss_mlp": 0.01029082,
      "balance_loss_clip": 1.05950594,
      "balance_loss_mlp": 1.0188657,
      "epoch": 0.5000901821679793,
      "flos": 23327212533120.0,
      "grad_norm": 2.213023639578084,
      "language_loss": 0.78987706,
      "learning_rate": 2.0965549378822618e-06,
      "loss": 0.81209731,
      "num_input_tokens_seen": 89678415,
      "step": 4159,
      "time_per_iteration": 2.6353256702423096
    },
    {
      "auxiliary_loss_clip": 0.01189041,
      "auxiliary_loss_mlp": 0.0103242,
      "balance_loss_clip": 1.05761707,
      "balance_loss_mlp": 1.02310967,
      "epoch": 0.5002104250586185,
      "flos": 20339014239360.0,
      "grad_norm": 3.3883533850252716,
      "language_loss": 0.83795547,
      "learning_rate": 2.095776867240599e-06,
      "loss": 0.86017007,
      "num_input_tokens_seen": 89695405,
      "step": 4160,
      "time_per_iteration": 2.6555094718933105
    },
    {
      "auxiliary_loss_clip": 0.01189646,
      "auxiliary_loss_mlp": 0.01024913,
      "balance_loss_clip": 1.05860043,
      "balance_loss_mlp": 1.01592469,
      "epoch": 0.5003306679492575,
      "flos": 13991372634240.0,
      "grad_norm": 7.043145012813491,
      "language_loss": 0.82302952,
      "learning_rate": 2.094998782069661e-06,
      "loss": 0.84517509,
      "num_input_tokens_seen": 89713110,
      "step": 4161,
      "time_per_iteration": 2.601011037826538
    },
    {
      "auxiliary_loss_clip": 0.01188161,
      "auxiliary_loss_mlp": 0.01032298,
      "balance_loss_clip": 1.0572238,
      "balance_loss_mlp": 1.02249944,
      "epoch": 0.5004509108398966,
      "flos": 27672762896640.0,
      "grad_norm": 1.6726528993670804,
      "language_loss": 0.75759608,
      "learning_rate": 2.0942206824874845e-06,
      "loss": 0.77980065,
      "num_input_tokens_seen": 89735885,
      "step": 4162,
      "time_per_iteration": 2.718170404434204
    },
    {
      "auxiliary_loss_clip": 0.01188492,
      "auxiliary_loss_mlp": 0.01030568,
      "balance_loss_clip": 1.05504465,
      "balance_loss_mlp": 1.02117455,
      "epoch": 0.5005711537305357,
      "flos": 14976186796800.0,
      "grad_norm": 7.066804906414022,
      "language_loss": 0.79312479,
      "learning_rate": 2.093442568612105e-06,
      "loss": 0.81531537,
      "num_input_tokens_seen": 89753690,
      "step": 4163,
      "time_per_iteration": 2.6964094638824463
    },
    {
      "auxiliary_loss_clip": 0.01189481,
      "auxiliary_loss_mlp": 0.01027807,
      "balance_loss_clip": 1.05532169,
      "balance_loss_mlp": 1.01745963,
      "epoch": 0.5006913966211748,
      "flos": 26503259978880.0,
      "grad_norm": 2.5211363290242588,
      "language_loss": 0.85060406,
      "learning_rate": 2.0926644405615613e-06,
      "loss": 0.87277699,
      "num_input_tokens_seen": 89774590,
      "step": 4164,
      "time_per_iteration": 2.7091829776763916
    },
    {
      "auxiliary_loss_clip": 0.0118664,
      "auxiliary_loss_mlp": 0.01023748,
      "balance_loss_clip": 1.05484462,
      "balance_loss_mlp": 1.01458073,
      "epoch": 0.5008116395118138,
      "flos": 20449295971200.0,
      "grad_norm": 2.152587746061348,
      "language_loss": 0.81567872,
      "learning_rate": 2.091886298453897e-06,
      "loss": 0.83778262,
      "num_input_tokens_seen": 89792775,
      "step": 4165,
      "time_per_iteration": 2.574129581451416
    },
    {
      "auxiliary_loss_clip": 0.0118946,
      "auxiliary_loss_mlp": 0.01030143,
      "balance_loss_clip": 1.05883956,
      "balance_loss_mlp": 1.02079654,
      "epoch": 0.500931882402453,
      "flos": 21579871524480.0,
      "grad_norm": 2.622732878816384,
      "language_loss": 0.73265511,
      "learning_rate": 2.091108142407153e-06,
      "loss": 0.75485116,
      "num_input_tokens_seen": 89811515,
      "step": 4166,
      "time_per_iteration": 2.66825532913208
    },
    {
      "auxiliary_loss_clip": 0.01110901,
      "auxiliary_loss_mlp": 0.01002264,
      "balance_loss_clip": 1.03125632,
      "balance_loss_mlp": 0.99992734,
      "epoch": 0.5010521252930921,
      "flos": 57785011925760.0,
      "grad_norm": 0.885342407529933,
      "language_loss": 0.62380183,
      "learning_rate": 2.090329972539377e-06,
      "loss": 0.64493346,
      "num_input_tokens_seen": 89870080,
      "step": 4167,
      "time_per_iteration": 3.222283363342285
    },
    {
      "auxiliary_loss_clip": 0.01188801,
      "auxiliary_loss_mlp": 0.01035955,
      "balance_loss_clip": 1.05617833,
      "balance_loss_mlp": 1.02672255,
      "epoch": 0.5011723681837311,
      "flos": 18625500864000.0,
      "grad_norm": 2.455580637724034,
      "language_loss": 0.68578362,
      "learning_rate": 2.089551788968616e-06,
      "loss": 0.70803124,
      "num_input_tokens_seen": 89888045,
      "step": 4168,
      "time_per_iteration": 2.6856489181518555
    },
    {
      "auxiliary_loss_clip": 0.01106329,
      "auxiliary_loss_mlp": 0.01002195,
      "balance_loss_clip": 1.02810454,
      "balance_loss_mlp": 0.99982244,
      "epoch": 0.5012926110743702,
      "flos": 55883146608000.0,
      "grad_norm": 0.8444809034169007,
      "language_loss": 0.60729766,
      "learning_rate": 2.08877359181292e-06,
      "loss": 0.62838292,
      "num_input_tokens_seen": 89944610,
      "step": 4169,
      "time_per_iteration": 3.1491854190826416
    },
    {
      "auxiliary_loss_clip": 0.01188516,
      "auxiliary_loss_mlp": 0.010237,
      "balance_loss_clip": 1.05647695,
      "balance_loss_mlp": 1.01478958,
      "epoch": 0.5014128539650093,
      "flos": 24238266117120.0,
      "grad_norm": 2.2163348502127884,
      "language_loss": 0.85075116,
      "learning_rate": 2.0879953811903396e-06,
      "loss": 0.87287331,
      "num_input_tokens_seen": 89959495,
      "step": 4170,
      "time_per_iteration": 2.644787311553955
    },
    {
      "auxiliary_loss_clip": 0.011882,
      "auxiliary_loss_mlp": 0.01027048,
      "balance_loss_clip": 1.05506551,
      "balance_loss_mlp": 1.01781511,
      "epoch": 0.5015330968556484,
      "flos": 27527468382720.0,
      "grad_norm": 1.8895476035509937,
      "language_loss": 0.78245455,
      "learning_rate": 2.08721715721893e-06,
      "loss": 0.80460703,
      "num_input_tokens_seen": 89978820,
      "step": 4171,
      "time_per_iteration": 2.70023775100708
    },
    {
      "auxiliary_loss_clip": 0.01188066,
      "auxiliary_loss_mlp": 0.01026268,
      "balance_loss_clip": 1.05580497,
      "balance_loss_mlp": 1.01695752,
      "epoch": 0.5016533397462875,
      "flos": 23800802376960.0,
      "grad_norm": 2.46085580295577,
      "language_loss": 0.76683921,
      "learning_rate": 2.0864389200167477e-06,
      "loss": 0.78898251,
      "num_input_tokens_seen": 89997075,
      "step": 4172,
      "time_per_iteration": 2.5983335971832275
    },
    {
      "auxiliary_loss_clip": 0.01187091,
      "auxiliary_loss_mlp": 0.01030644,
      "balance_loss_clip": 1.05739093,
      "balance_loss_mlp": 1.02097631,
      "epoch": 0.5017735826369266,
      "flos": 25295009264640.0,
      "grad_norm": 4.0970122177753465,
      "language_loss": 0.78883034,
      "learning_rate": 2.0856606697018504e-06,
      "loss": 0.81100774,
      "num_input_tokens_seen": 90015085,
      "step": 4173,
      "time_per_iteration": 2.684432029724121
    },
    {
      "auxiliary_loss_clip": 0.01187784,
      "auxiliary_loss_mlp": 0.01025997,
      "balance_loss_clip": 1.05471015,
      "balance_loss_mlp": 1.01669264,
      "epoch": 0.5018938255275657,
      "flos": 16873203778560.0,
      "grad_norm": 2.21233495747977,
      "language_loss": 0.73459041,
      "learning_rate": 2.084882406392297e-06,
      "loss": 0.75672823,
      "num_input_tokens_seen": 90033045,
      "step": 4174,
      "time_per_iteration": 2.6052281856536865
    },
    {
      "auxiliary_loss_clip": 0.01186715,
      "auxiliary_loss_mlp": 0.01027801,
      "balance_loss_clip": 1.05517626,
      "balance_loss_mlp": 1.01869988,
      "epoch": 0.5020140684182047,
      "flos": 25515429073920.0,
      "grad_norm": 3.4215389248724546,
      "language_loss": 0.71207649,
      "learning_rate": 2.0841041302061496e-06,
      "loss": 0.7342217,
      "num_input_tokens_seen": 90052505,
      "step": 4175,
      "time_per_iteration": 2.642899513244629
    },
    {
      "auxiliary_loss_clip": 0.01188206,
      "auxiliary_loss_mlp": 0.01031993,
      "balance_loss_clip": 1.05660355,
      "balance_loss_mlp": 1.02234316,
      "epoch": 0.5021343113088439,
      "flos": 23659278791040.0,
      "grad_norm": 3.7632727909165213,
      "language_loss": 0.75221938,
      "learning_rate": 2.083325841261473e-06,
      "loss": 0.77442133,
      "num_input_tokens_seen": 90071565,
      "step": 4176,
      "time_per_iteration": 3.5872209072113037
    },
    {
      "auxiliary_loss_clip": 0.01189314,
      "auxiliary_loss_mlp": 0.01028303,
      "balance_loss_clip": 1.0564971,
      "balance_loss_mlp": 1.01845634,
      "epoch": 0.502254554199483,
      "flos": 24534673148160.0,
      "grad_norm": 1.8783587516152525,
      "language_loss": 0.66172928,
      "learning_rate": 2.0825475396763322e-06,
      "loss": 0.68390548,
      "num_input_tokens_seen": 90092215,
      "step": 4177,
      "time_per_iteration": 2.6245102882385254
    },
    {
      "auxiliary_loss_clip": 0.01189993,
      "auxiliary_loss_mlp": 0.01027056,
      "balance_loss_clip": 1.05802441,
      "balance_loss_mlp": 1.01822841,
      "epoch": 0.502374797090122,
      "flos": 34240285607040.0,
      "grad_norm": 1.710901753325786,
      "language_loss": 0.65593868,
      "learning_rate": 2.081769225568796e-06,
      "loss": 0.67810917,
      "num_input_tokens_seen": 90114665,
      "step": 4178,
      "time_per_iteration": 2.745014190673828
    },
    {
      "auxiliary_loss_clip": 0.01185444,
      "auxiliary_loss_mlp": 0.01029311,
      "balance_loss_clip": 1.05425167,
      "balance_loss_mlp": 1.01988196,
      "epoch": 0.5024950399807612,
      "flos": 26031106679040.0,
      "grad_norm": 2.635025957370969,
      "language_loss": 0.76298261,
      "learning_rate": 2.0809908990569327e-06,
      "loss": 0.78513014,
      "num_input_tokens_seen": 90136445,
      "step": 4179,
      "time_per_iteration": 2.7325329780578613
    },
    {
      "auxiliary_loss_clip": 0.01189425,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.05732489,
      "balance_loss_mlp": 1.02066898,
      "epoch": 0.5026152828714002,
      "flos": 21252438120960.0,
      "grad_norm": 2.3528441063876833,
      "language_loss": 0.7937066,
      "learning_rate": 2.0802125602588146e-06,
      "loss": 0.81590056,
      "num_input_tokens_seen": 90155710,
      "step": 4180,
      "time_per_iteration": 3.61641263961792
    },
    {
      "auxiliary_loss_clip": 0.01186092,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.0543468,
      "balance_loss_mlp": 1.01978421,
      "epoch": 0.5027355257620393,
      "flos": 30956111245440.0,
      "grad_norm": 1.8587513964645985,
      "language_loss": 0.6621272,
      "learning_rate": 2.0794342092925146e-06,
      "loss": 0.68427271,
      "num_input_tokens_seen": 90176845,
      "step": 4181,
      "time_per_iteration": 3.5809831619262695
    },
    {
      "auxiliary_loss_clip": 0.01186357,
      "auxiliary_loss_mlp": 0.01028846,
      "balance_loss_clip": 1.05715036,
      "balance_loss_mlp": 1.02013135,
      "epoch": 0.5028557686526784,
      "flos": 24791147233920.0,
      "grad_norm": 2.0930843431815624,
      "language_loss": 0.67786539,
      "learning_rate": 2.078655846276108e-06,
      "loss": 0.70001745,
      "num_input_tokens_seen": 90197175,
      "step": 4182,
      "time_per_iteration": 3.6149051189422607
    },
    {
      "auxiliary_loss_clip": 0.01184589,
      "auxiliary_loss_mlp": 0.01024651,
      "balance_loss_clip": 1.05376768,
      "balance_loss_mlp": 1.01629448,
      "epoch": 0.5029760115433175,
      "flos": 22966992990720.0,
      "grad_norm": 2.0276565845421537,
      "language_loss": 0.68621153,
      "learning_rate": 2.0778774713276727e-06,
      "loss": 0.70830393,
      "num_input_tokens_seen": 90216650,
      "step": 4183,
      "time_per_iteration": 2.6476755142211914
    },
    {
      "auxiliary_loss_clip": 0.01185714,
      "auxiliary_loss_mlp": 0.01025162,
      "balance_loss_clip": 1.05305243,
      "balance_loss_mlp": 1.01604223,
      "epoch": 0.5030962544339566,
      "flos": 15305164485120.0,
      "grad_norm": 4.236661412887899,
      "language_loss": 0.6783185,
      "learning_rate": 2.077099084565287e-06,
      "loss": 0.70042729,
      "num_input_tokens_seen": 90234055,
      "step": 4184,
      "time_per_iteration": 2.6250433921813965
    },
    {
      "auxiliary_loss_clip": 0.01185886,
      "auxiliary_loss_mlp": 0.01026602,
      "balance_loss_clip": 1.05332208,
      "balance_loss_mlp": 1.01731586,
      "epoch": 0.5032164973245957,
      "flos": 24494847943680.0,
      "grad_norm": 2.3923521493574036,
      "language_loss": 0.64922738,
      "learning_rate": 2.0763206861070313e-06,
      "loss": 0.67135227,
      "num_input_tokens_seen": 90253115,
      "step": 4185,
      "time_per_iteration": 2.643373966217041
    },
    {
      "auxiliary_loss_clip": 0.01189139,
      "auxiliary_loss_mlp": 0.0102747,
      "balance_loss_clip": 1.05684292,
      "balance_loss_mlp": 1.01821375,
      "epoch": 0.5033367402152348,
      "flos": 16213452721920.0,
      "grad_norm": 2.0754756320862078,
      "language_loss": 0.75423616,
      "learning_rate": 2.0755422760709876e-06,
      "loss": 0.77640224,
      "num_input_tokens_seen": 90270515,
      "step": 4186,
      "time_per_iteration": 2.637749671936035
    },
    {
      "auxiliary_loss_clip": 0.01188099,
      "auxiliary_loss_mlp": 0.01031439,
      "balance_loss_clip": 1.05626357,
      "balance_loss_mlp": 1.02227759,
      "epoch": 0.5034569831058738,
      "flos": 21391375927680.0,
      "grad_norm": 3.149329370072541,
      "language_loss": 0.76646984,
      "learning_rate": 2.0747638545752417e-06,
      "loss": 0.78866518,
      "num_input_tokens_seen": 90289075,
      "step": 4187,
      "time_per_iteration": 2.633737802505493
    },
    {
      "auxiliary_loss_clip": 0.01183647,
      "auxiliary_loss_mlp": 0.01029751,
      "balance_loss_clip": 1.05410171,
      "balance_loss_mlp": 1.02114999,
      "epoch": 0.503577225996513,
      "flos": 20558751690240.0,
      "grad_norm": 2.5233262056444725,
      "language_loss": 0.8338089,
      "learning_rate": 2.073985421737878e-06,
      "loss": 0.85594296,
      "num_input_tokens_seen": 90306385,
      "step": 4188,
      "time_per_iteration": 2.626577377319336
    },
    {
      "auxiliary_loss_clip": 0.01186774,
      "auxiliary_loss_mlp": 0.01024747,
      "balance_loss_clip": 1.05491042,
      "balance_loss_mlp": 1.014853,
      "epoch": 0.5036974688871521,
      "flos": 27229157930880.0,
      "grad_norm": 6.1632700585569005,
      "language_loss": 0.73948312,
      "learning_rate": 2.0732069776769844e-06,
      "loss": 0.76159829,
      "num_input_tokens_seen": 90323795,
      "step": 4189,
      "time_per_iteration": 2.642078161239624
    },
    {
      "auxiliary_loss_clip": 0.0118824,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.05715632,
      "balance_loss_mlp": 1.01800466,
      "epoch": 0.5038177117777911,
      "flos": 20412164286720.0,
      "grad_norm": 5.4133390324790565,
      "language_loss": 0.73203266,
      "learning_rate": 2.072428522510651e-06,
      "loss": 0.75419146,
      "num_input_tokens_seen": 90340360,
      "step": 4190,
      "time_per_iteration": 2.707937717437744
    },
    {
      "auxiliary_loss_clip": 0.01186679,
      "auxiliary_loss_mlp": 0.01029533,
      "balance_loss_clip": 1.05542612,
      "balance_loss_mlp": 1.02080095,
      "epoch": 0.5039379546684303,
      "flos": 21907987286400.0,
      "grad_norm": 2.171262794763968,
      "language_loss": 0.76526475,
      "learning_rate": 2.071650056356968e-06,
      "loss": 0.78742689,
      "num_input_tokens_seen": 90357900,
      "step": 4191,
      "time_per_iteration": 2.6180481910705566
    },
    {
      "auxiliary_loss_clip": 0.01189214,
      "auxiliary_loss_mlp": 0.01034102,
      "balance_loss_clip": 1.05736172,
      "balance_loss_mlp": 1.02528071,
      "epoch": 0.5040581975590693,
      "flos": 20010718909440.0,
      "grad_norm": 2.394867337090465,
      "language_loss": 0.79807091,
      "learning_rate": 2.070871579334028e-06,
      "loss": 0.82030404,
      "num_input_tokens_seen": 90377010,
      "step": 4192,
      "time_per_iteration": 2.618746042251587
    },
    {
      "auxiliary_loss_clip": 0.01184202,
      "auxiliary_loss_mlp": 0.01028595,
      "balance_loss_clip": 1.05263078,
      "balance_loss_mlp": 1.01934433,
      "epoch": 0.5041784404497084,
      "flos": 20959837931520.0,
      "grad_norm": 4.188528009281382,
      "language_loss": 0.71590179,
      "learning_rate": 2.0700930915599264e-06,
      "loss": 0.73802984,
      "num_input_tokens_seen": 90396740,
      "step": 4193,
      "time_per_iteration": 2.5949883460998535
    },
    {
      "auxiliary_loss_clip": 0.01184137,
      "auxiliary_loss_mlp": 0.01027481,
      "balance_loss_clip": 1.05324483,
      "balance_loss_mlp": 1.0184691,
      "epoch": 0.5042986833403476,
      "flos": 12495082757760.0,
      "grad_norm": 2.758960565252597,
      "language_loss": 0.78885531,
      "learning_rate": 2.0693145931527583e-06,
      "loss": 0.8109715,
      "num_input_tokens_seen": 90413220,
      "step": 4194,
      "time_per_iteration": 2.6463723182678223
    },
    {
      "auxiliary_loss_clip": 0.01186617,
      "auxiliary_loss_mlp": 0.01031287,
      "balance_loss_clip": 1.05404234,
      "balance_loss_mlp": 1.02251339,
      "epoch": 0.5044189262309866,
      "flos": 29202305788800.0,
      "grad_norm": 1.7309306464887602,
      "language_loss": 0.77962416,
      "learning_rate": 2.068536084230622e-06,
      "loss": 0.80180323,
      "num_input_tokens_seen": 90435085,
      "step": 4195,
      "time_per_iteration": 2.644894599914551
    },
    {
      "auxiliary_loss_clip": 0.01187927,
      "auxiliary_loss_mlp": 0.01036157,
      "balance_loss_clip": 1.05489206,
      "balance_loss_mlp": 1.0254519,
      "epoch": 0.5045391691216257,
      "flos": 23873198238720.0,
      "grad_norm": 17.36725164966176,
      "language_loss": 0.88273299,
      "learning_rate": 2.067757564911616e-06,
      "loss": 0.90497386,
      "num_input_tokens_seen": 90453660,
      "step": 4196,
      "time_per_iteration": 2.655620813369751
    },
    {
      "auxiliary_loss_clip": 0.01184331,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.05350578,
      "balance_loss_mlp": 1.01608396,
      "epoch": 0.5046594120122648,
      "flos": 24644990793600.0,
      "grad_norm": 2.6769694298583677,
      "language_loss": 0.92370665,
      "learning_rate": 2.0669790353138407e-06,
      "loss": 0.94580233,
      "num_input_tokens_seen": 90472625,
      "step": 4197,
      "time_per_iteration": 2.6097612380981445
    },
    {
      "auxiliary_loss_clip": 0.01188067,
      "auxiliary_loss_mlp": 0.01032062,
      "balance_loss_clip": 1.05631924,
      "balance_loss_mlp": 1.02239394,
      "epoch": 0.5047796549029039,
      "flos": 23362835846400.0,
      "grad_norm": 3.4847410367568044,
      "language_loss": 0.7306416,
      "learning_rate": 2.0662004955553995e-06,
      "loss": 0.7528429,
      "num_input_tokens_seen": 90492325,
      "step": 4198,
      "time_per_iteration": 2.659881830215454
    },
    {
      "auxiliary_loss_clip": 0.01184508,
      "auxiliary_loss_mlp": 0.01025385,
      "balance_loss_clip": 1.05534482,
      "balance_loss_mlp": 1.01715922,
      "epoch": 0.5048998977935429,
      "flos": 17304095329920.0,
      "grad_norm": 1.9577236718029472,
      "language_loss": 0.76840508,
      "learning_rate": 2.065421945754395e-06,
      "loss": 0.79050398,
      "num_input_tokens_seen": 90510055,
      "step": 4199,
      "time_per_iteration": 2.575349807739258
    },
    {
      "auxiliary_loss_clip": 0.01185678,
      "auxiliary_loss_mlp": 0.01035977,
      "balance_loss_clip": 1.05656362,
      "balance_loss_mlp": 1.02627337,
      "epoch": 0.505020140684182,
      "flos": 34856979235200.0,
      "grad_norm": 6.834612455672516,
      "language_loss": 0.77910757,
      "learning_rate": 2.0646433860289344e-06,
      "loss": 0.80132413,
      "num_input_tokens_seen": 90528980,
      "step": 4200,
      "time_per_iteration": 2.7051587104797363
    },
    {
      "auxiliary_loss_clip": 0.01188282,
      "auxiliary_loss_mlp": 0.01025398,
      "balance_loss_clip": 1.05372596,
      "balance_loss_mlp": 1.01591432,
      "epoch": 0.5051403835748212,
      "flos": 24863974058880.0,
      "grad_norm": 2.553470857596031,
      "language_loss": 0.82464206,
      "learning_rate": 2.0638648164971233e-06,
      "loss": 0.84677893,
      "num_input_tokens_seen": 90547445,
      "step": 4201,
      "time_per_iteration": 2.6231088638305664
    },
    {
      "auxiliary_loss_clip": 0.01183738,
      "auxiliary_loss_mlp": 0.01027924,
      "balance_loss_clip": 1.05451047,
      "balance_loss_mlp": 1.01899517,
      "epoch": 0.5052606264654602,
      "flos": 20959694277120.0,
      "grad_norm": 2.373540675406167,
      "language_loss": 0.88706195,
      "learning_rate": 2.06308623727707e-06,
      "loss": 0.90917861,
      "num_input_tokens_seen": 90567545,
      "step": 4202,
      "time_per_iteration": 3.6049530506134033
    },
    {
      "auxiliary_loss_clip": 0.01186613,
      "auxiliary_loss_mlp": 0.01025815,
      "balance_loss_clip": 1.05434418,
      "balance_loss_mlp": 1.01674342,
      "epoch": 0.5053808693560993,
      "flos": 19642382893440.0,
      "grad_norm": 2.9169348082869995,
      "language_loss": 0.7665745,
      "learning_rate": 2.0623076484868846e-06,
      "loss": 0.78869879,
      "num_input_tokens_seen": 90585000,
      "step": 4203,
      "time_per_iteration": 2.651094913482666
    },
    {
      "auxiliary_loss_clip": 0.01092661,
      "auxiliary_loss_mlp": 0.01002503,
      "balance_loss_clip": 1.02364993,
      "balance_loss_mlp": 1.00098884,
      "epoch": 0.5055011122467384,
      "flos": 67504915019520.0,
      "grad_norm": 0.8326746673375135,
      "language_loss": 0.6064893,
      "learning_rate": 2.061529050244679e-06,
      "loss": 0.62744093,
      "num_input_tokens_seen": 90644745,
      "step": 4204,
      "time_per_iteration": 3.1465139389038086
    },
    {
      "auxiliary_loss_clip": 0.01185197,
      "auxiliary_loss_mlp": 0.0103124,
      "balance_loss_clip": 1.05462301,
      "balance_loss_mlp": 1.02223957,
      "epoch": 0.5056213551373775,
      "flos": 16872952383360.0,
      "grad_norm": 13.557591985983361,
      "language_loss": 0.73834544,
      "learning_rate": 2.060750442668565e-06,
      "loss": 0.76050985,
      "num_input_tokens_seen": 90662500,
      "step": 4205,
      "time_per_iteration": 2.6242222785949707
    },
    {
      "auxiliary_loss_clip": 0.01186839,
      "auxiliary_loss_mlp": 0.01032661,
      "balance_loss_clip": 1.05712175,
      "balance_loss_mlp": 1.02335668,
      "epoch": 0.5057415980280165,
      "flos": 15334179696000.0,
      "grad_norm": 3.1892176235988168,
      "language_loss": 0.63813341,
      "learning_rate": 2.059971825876657e-06,
      "loss": 0.66032839,
      "num_input_tokens_seen": 90677010,
      "step": 4206,
      "time_per_iteration": 3.5669116973876953
    },
    {
      "auxiliary_loss_clip": 0.01184454,
      "auxiliary_loss_mlp": 0.01025443,
      "balance_loss_clip": 1.05448604,
      "balance_loss_mlp": 1.01743197,
      "epoch": 0.5058618409186557,
      "flos": 19025976574080.0,
      "grad_norm": 1.823055746991044,
      "language_loss": 0.76338696,
      "learning_rate": 2.0591931999870713e-06,
      "loss": 0.78548598,
      "num_input_tokens_seen": 90695935,
      "step": 4207,
      "time_per_iteration": 3.476412534713745
    },
    {
      "auxiliary_loss_clip": 0.01090458,
      "auxiliary_loss_mlp": 0.01001564,
      "balance_loss_clip": 1.02271557,
      "balance_loss_mlp": 1.00024068,
      "epoch": 0.5059820838092948,
      "flos": 63453114080640.0,
      "grad_norm": 0.8299836936852405,
      "language_loss": 0.5753805,
      "learning_rate": 2.0584145651179234e-06,
      "loss": 0.59630078,
      "num_input_tokens_seen": 90751645,
      "step": 4208,
      "time_per_iteration": 4.0338475704193115
    },
    {
      "auxiliary_loss_clip": 0.01185321,
      "auxiliary_loss_mlp": 0.01038957,
      "balance_loss_clip": 1.05570161,
      "balance_loss_mlp": 1.02989757,
      "epoch": 0.5061023266999338,
      "flos": 15441803821440.0,
      "grad_norm": 3.319867265176537,
      "language_loss": 0.7992447,
      "learning_rate": 2.0576359213873327e-06,
      "loss": 0.82148749,
      "num_input_tokens_seen": 90766795,
      "step": 4209,
      "time_per_iteration": 2.6302595138549805
    },
    {
      "auxiliary_loss_clip": 0.01188102,
      "auxiliary_loss_mlp": 0.01027613,
      "balance_loss_clip": 1.05270672,
      "balance_loss_mlp": 1.01787972,
      "epoch": 0.506222569590573,
      "flos": 22451063990400.0,
      "grad_norm": 2.572968694193856,
      "language_loss": 0.70600283,
      "learning_rate": 2.056857268913419e-06,
      "loss": 0.72815996,
      "num_input_tokens_seen": 90786845,
      "step": 4210,
      "time_per_iteration": 2.6279540061950684
    },
    {
      "auxiliary_loss_clip": 0.01188207,
      "auxiliary_loss_mlp": 0.01027129,
      "balance_loss_clip": 1.05737829,
      "balance_loss_mlp": 1.01833177,
      "epoch": 0.506342812481212,
      "flos": 17558665994880.0,
      "grad_norm": 2.404995858474789,
      "language_loss": 0.84800327,
      "learning_rate": 2.056078607814303e-06,
      "loss": 0.87015665,
      "num_input_tokens_seen": 90802630,
      "step": 4211,
      "time_per_iteration": 2.612330198287964
    },
    {
      "auxiliary_loss_clip": 0.01184689,
      "auxiliary_loss_mlp": 0.01027094,
      "balance_loss_clip": 1.05316293,
      "balance_loss_mlp": 1.01702666,
      "epoch": 0.5064630553718511,
      "flos": 23402050519680.0,
      "grad_norm": 2.057559887311814,
      "language_loss": 0.78032184,
      "learning_rate": 2.055299938208106e-06,
      "loss": 0.80243969,
      "num_input_tokens_seen": 90823620,
      "step": 4212,
      "time_per_iteration": 2.6772449016571045
    },
    {
      "auxiliary_loss_clip": 0.01189213,
      "auxiliary_loss_mlp": 0.01035274,
      "balance_loss_clip": 1.05522978,
      "balance_loss_mlp": 1.02621436,
      "epoch": 0.5065832982624903,
      "flos": 23987035416960.0,
      "grad_norm": 2.441147980477992,
      "language_loss": 0.86011928,
      "learning_rate": 2.0545212602129526e-06,
      "loss": 0.88236415,
      "num_input_tokens_seen": 90843475,
      "step": 4213,
      "time_per_iteration": 2.698103189468384
    },
    {
      "auxiliary_loss_clip": 0.01186361,
      "auxiliary_loss_mlp": 0.01029872,
      "balance_loss_clip": 1.05460298,
      "balance_loss_mlp": 1.02022195,
      "epoch": 0.5067035411531293,
      "flos": 21503058289920.0,
      "grad_norm": 2.9291151211183384,
      "language_loss": 0.66433942,
      "learning_rate": 2.0537425739469673e-06,
      "loss": 0.68650174,
      "num_input_tokens_seen": 90862410,
      "step": 4214,
      "time_per_iteration": 2.592956066131592
    },
    {
      "auxiliary_loss_clip": 0.0108765,
      "auxiliary_loss_mlp": 0.010018,
      "balance_loss_clip": 1.01956224,
      "balance_loss_mlp": 1.00044656,
      "epoch": 0.5068237840437684,
      "flos": 65934397687680.0,
      "grad_norm": 0.8388783265953,
      "language_loss": 0.59423339,
      "learning_rate": 2.052963879528276e-06,
      "loss": 0.61512786,
      "num_input_tokens_seen": 90922280,
      "step": 4215,
      "time_per_iteration": 3.1637215614318848
    },
    {
      "auxiliary_loss_clip": 0.01189605,
      "auxiliary_loss_mlp": 0.01025373,
      "balance_loss_clip": 1.0566783,
      "balance_loss_mlp": 1.01644421,
      "epoch": 0.5069440269344075,
      "flos": 27264206626560.0,
      "grad_norm": 2.389169970309749,
      "language_loss": 0.7663461,
      "learning_rate": 2.052185177075007e-06,
      "loss": 0.7884959,
      "num_input_tokens_seen": 90941850,
      "step": 4216,
      "time_per_iteration": 2.652885913848877
    },
    {
      "auxiliary_loss_clip": 0.01188718,
      "auxiliary_loss_mlp": 0.01032684,
      "balance_loss_clip": 1.05542374,
      "balance_loss_mlp": 1.02385044,
      "epoch": 0.5070642698250466,
      "flos": 23366319465600.0,
      "grad_norm": 2.06278477232312,
      "language_loss": 0.83116674,
      "learning_rate": 2.051406466705288e-06,
      "loss": 0.85338074,
      "num_input_tokens_seen": 90961390,
      "step": 4217,
      "time_per_iteration": 2.6571993827819824
    },
    {
      "auxiliary_loss_clip": 0.01184393,
      "auxiliary_loss_mlp": 0.0102758,
      "balance_loss_clip": 1.05274463,
      "balance_loss_mlp": 1.01931882,
      "epoch": 0.5071845127156857,
      "flos": 20340127560960.0,
      "grad_norm": 2.2281288993038637,
      "language_loss": 0.80890286,
      "learning_rate": 2.0506277485372486e-06,
      "loss": 0.83102262,
      "num_input_tokens_seen": 90980215,
      "step": 4218,
      "time_per_iteration": 2.6085622310638428
    },
    {
      "auxiliary_loss_clip": 0.01188452,
      "auxiliary_loss_mlp": 0.01027269,
      "balance_loss_clip": 1.05480826,
      "balance_loss_mlp": 1.01803005,
      "epoch": 0.5073047556063248,
      "flos": 12092955022080.0,
      "grad_norm": 2.4583121914279165,
      "language_loss": 0.67200959,
      "learning_rate": 2.04984902268902e-06,
      "loss": 0.69416684,
      "num_input_tokens_seen": 90997415,
      "step": 4219,
      "time_per_iteration": 2.587432861328125
    },
    {
      "auxiliary_loss_clip": 0.0119262,
      "auxiliary_loss_mlp": 0.01030359,
      "balance_loss_clip": 1.05518341,
      "balance_loss_mlp": 1.02041674,
      "epoch": 0.5074249984969639,
      "flos": 19682854542720.0,
      "grad_norm": 2.5010002356760728,
      "language_loss": 0.75505984,
      "learning_rate": 2.0490702892787345e-06,
      "loss": 0.77728963,
      "num_input_tokens_seen": 91016475,
      "step": 4220,
      "time_per_iteration": 2.637355327606201
    },
    {
      "auxiliary_loss_clip": 0.01182091,
      "auxiliary_loss_mlp": 0.0102868,
      "balance_loss_clip": 1.05208313,
      "balance_loss_mlp": 1.01951909,
      "epoch": 0.5075452413876029,
      "flos": 28765703975040.0,
      "grad_norm": 1.8485838247731199,
      "language_loss": 0.62306571,
      "learning_rate": 2.0482915484245246e-06,
      "loss": 0.64517343,
      "num_input_tokens_seen": 91038095,
      "step": 4221,
      "time_per_iteration": 2.688246488571167
    },
    {
      "auxiliary_loss_clip": 0.01190157,
      "auxiliary_loss_mlp": 0.01034396,
      "balance_loss_clip": 1.05569124,
      "balance_loss_mlp": 1.0250082,
      "epoch": 0.5076654842782421,
      "flos": 20339445202560.0,
      "grad_norm": 9.600841953983215,
      "language_loss": 0.84053749,
      "learning_rate": 2.047512800244526e-06,
      "loss": 0.86278296,
      "num_input_tokens_seen": 91053360,
      "step": 4222,
      "time_per_iteration": 2.645620346069336
    },
    {
      "auxiliary_loss_clip": 0.01185159,
      "auxiliary_loss_mlp": 0.01030407,
      "balance_loss_clip": 1.05361772,
      "balance_loss_mlp": 1.0211271,
      "epoch": 0.5077857271688812,
      "flos": 26359653404160.0,
      "grad_norm": 2.2351310988691915,
      "language_loss": 0.79415369,
      "learning_rate": 2.046734044856873e-06,
      "loss": 0.81630939,
      "num_input_tokens_seen": 91072770,
      "step": 4223,
      "time_per_iteration": 2.656696081161499
    },
    {
      "auxiliary_loss_clip": 0.01185727,
      "auxiliary_loss_mlp": 0.010278,
      "balance_loss_clip": 1.0540396,
      "balance_loss_mlp": 1.01863265,
      "epoch": 0.5079059700595202,
      "flos": 21798962530560.0,
      "grad_norm": 2.3057101563574713,
      "language_loss": 0.81276232,
      "learning_rate": 2.045955282379702e-06,
      "loss": 0.83489758,
      "num_input_tokens_seen": 91091430,
      "step": 4224,
      "time_per_iteration": 2.726820945739746
    },
    {
      "auxiliary_loss_clip": 0.01184617,
      "auxiliary_loss_mlp": 0.01031884,
      "balance_loss_clip": 1.05231738,
      "balance_loss_mlp": 1.02207899,
      "epoch": 0.5080262129501594,
      "flos": 13187943175680.0,
      "grad_norm": 3.1938387306137317,
      "language_loss": 0.76452589,
      "learning_rate": 2.045176512931152e-06,
      "loss": 0.78669083,
      "num_input_tokens_seen": 91106060,
      "step": 4225,
      "time_per_iteration": 2.596259832382202
    },
    {
      "auxiliary_loss_clip": 0.01184878,
      "auxiliary_loss_mlp": 0.01024167,
      "balance_loss_clip": 1.05487132,
      "balance_loss_mlp": 1.0158577,
      "epoch": 0.5081464558407984,
      "flos": 25301473712640.0,
      "grad_norm": 2.3194564850593182,
      "language_loss": 0.7600702,
      "learning_rate": 2.0443977366293604e-06,
      "loss": 0.78216064,
      "num_input_tokens_seen": 91124100,
      "step": 4226,
      "time_per_iteration": 2.64382266998291
    },
    {
      "auxiliary_loss_clip": 0.0119017,
      "auxiliary_loss_mlp": 0.01029599,
      "balance_loss_clip": 1.05504155,
      "balance_loss_mlp": 1.01934671,
      "epoch": 0.5082666987314375,
      "flos": 30951226995840.0,
      "grad_norm": 2.2646885946402193,
      "language_loss": 0.7696799,
      "learning_rate": 2.043618953592468e-06,
      "loss": 0.79187751,
      "num_input_tokens_seen": 91146555,
      "step": 4227,
      "time_per_iteration": 2.6764049530029297
    },
    {
      "auxiliary_loss_clip": 0.01187865,
      "auxiliary_loss_mlp": 0.01030455,
      "balance_loss_clip": 1.05578148,
      "balance_loss_mlp": 1.01969063,
      "epoch": 0.5083869416220766,
      "flos": 19682495406720.0,
      "grad_norm": 10.328976493291874,
      "language_loss": 0.81433797,
      "learning_rate": 2.0428401639386144e-06,
      "loss": 0.83652121,
      "num_input_tokens_seen": 91167120,
      "step": 4228,
      "time_per_iteration": 2.656233787536621
    },
    {
      "auxiliary_loss_clip": 0.01088784,
      "auxiliary_loss_mlp": 0.0100731,
      "balance_loss_clip": 1.01836681,
      "balance_loss_mlp": 1.00571215,
      "epoch": 0.5085071845127157,
      "flos": 71817535589760.0,
      "grad_norm": 0.8187226004343748,
      "language_loss": 0.58078974,
      "learning_rate": 2.042061367785943e-06,
      "loss": 0.60175073,
      "num_input_tokens_seen": 91220260,
      "step": 4229,
      "time_per_iteration": 4.131144046783447
    },
    {
      "auxiliary_loss_clip": 0.01185486,
      "auxiliary_loss_mlp": 0.01028742,
      "balance_loss_clip": 1.05210841,
      "balance_loss_mlp": 1.01928234,
      "epoch": 0.5086274274033548,
      "flos": 35951608252800.0,
      "grad_norm": 2.633658980131054,
      "language_loss": 0.75411308,
      "learning_rate": 2.041282565252594e-06,
      "loss": 0.77625537,
      "num_input_tokens_seen": 91240425,
      "step": 4230,
      "time_per_iteration": 2.704841375350952
    },
    {
      "auxiliary_loss_clip": 0.01182953,
      "auxiliary_loss_mlp": 0.01028269,
      "balance_loss_clip": 1.05334854,
      "balance_loss_mlp": 1.01890492,
      "epoch": 0.5087476702939938,
      "flos": 23513732881920.0,
      "grad_norm": 1.876294666954577,
      "language_loss": 0.7735374,
      "learning_rate": 2.040503756456714e-06,
      "loss": 0.79564965,
      "num_input_tokens_seen": 91259635,
      "step": 4231,
      "time_per_iteration": 2.618431329727173
    },
    {
      "auxiliary_loss_clip": 0.01184691,
      "auxiliary_loss_mlp": 0.01027271,
      "balance_loss_clip": 1.05387032,
      "balance_loss_mlp": 1.01784778,
      "epoch": 0.508867913184633,
      "flos": 15122091841920.0,
      "grad_norm": 2.372933108150753,
      "language_loss": 0.78925651,
      "learning_rate": 2.0397249415164456e-06,
      "loss": 0.81137615,
      "num_input_tokens_seen": 91276990,
      "step": 4232,
      "time_per_iteration": 3.506561517715454
    },
    {
      "auxiliary_loss_clip": 0.01185395,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 1.05377185,
      "balance_loss_mlp": 1.02091956,
      "epoch": 0.508988156075272,
      "flos": 25885309374720.0,
      "grad_norm": 2.0038859543707943,
      "language_loss": 0.80392396,
      "learning_rate": 2.0389461205499354e-06,
      "loss": 0.8260783,
      "num_input_tokens_seen": 91296125,
      "step": 4233,
      "time_per_iteration": 3.5489253997802734
    },
    {
      "auxiliary_loss_clip": 0.01186597,
      "auxiliary_loss_mlp": 0.01025873,
      "balance_loss_clip": 1.0555774,
      "balance_loss_mlp": 1.01723039,
      "epoch": 0.5091083989659111,
      "flos": 13844857057920.0,
      "grad_norm": 2.3682538258529573,
      "language_loss": 0.73398983,
      "learning_rate": 2.03816729367533e-06,
      "loss": 0.75611454,
      "num_input_tokens_seen": 91314280,
      "step": 4234,
      "time_per_iteration": 2.7089991569519043
    },
    {
      "auxiliary_loss_clip": 0.01190388,
      "auxiliary_loss_mlp": 0.01027959,
      "balance_loss_clip": 1.05652547,
      "balance_loss_mlp": 1.01863718,
      "epoch": 0.5092286418565503,
      "flos": 21104881050240.0,
      "grad_norm": 2.2175766016506646,
      "language_loss": 0.71522093,
      "learning_rate": 2.0373884610107765e-06,
      "loss": 0.73740435,
      "num_input_tokens_seen": 91334595,
      "step": 4235,
      "time_per_iteration": 3.545377492904663
    },
    {
      "auxiliary_loss_clip": 0.01182561,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.05039072,
      "balance_loss_mlp": 1.01926661,
      "epoch": 0.5093488847471893,
      "flos": 18621298972800.0,
      "grad_norm": 3.635398698485761,
      "language_loss": 0.69482827,
      "learning_rate": 2.0366096226744225e-06,
      "loss": 0.71693867,
      "num_input_tokens_seen": 91349790,
      "step": 4236,
      "time_per_iteration": 2.6361310482025146
    },
    {
      "auxiliary_loss_clip": 0.01185545,
      "auxiliary_loss_mlp": 0.01037028,
      "balance_loss_clip": 1.05459976,
      "balance_loss_mlp": 1.02762794,
      "epoch": 0.5094691276378284,
      "flos": 23803783205760.0,
      "grad_norm": 1.8721161888134374,
      "language_loss": 0.76926291,
      "learning_rate": 2.035830778784418e-06,
      "loss": 0.79148853,
      "num_input_tokens_seen": 91370465,
      "step": 4237,
      "time_per_iteration": 2.640392541885376
    },
    {
      "auxiliary_loss_clip": 0.01189715,
      "auxiliary_loss_mlp": 0.01035814,
      "balance_loss_clip": 1.05852139,
      "balance_loss_mlp": 1.02622938,
      "epoch": 0.5095893705284675,
      "flos": 17420410546560.0,
      "grad_norm": 3.9605057932036414,
      "language_loss": 0.80118024,
      "learning_rate": 2.0350519294589134e-06,
      "loss": 0.82343554,
      "num_input_tokens_seen": 91388505,
      "step": 4238,
      "time_per_iteration": 2.625807046890259
    },
    {
      "auxiliary_loss_clip": 0.0118685,
      "auxiliary_loss_mlp": 0.01033309,
      "balance_loss_clip": 1.05231643,
      "balance_loss_mlp": 1.02402258,
      "epoch": 0.5097096134191066,
      "flos": 25849362839040.0,
      "grad_norm": 2.3849439517602877,
      "language_loss": 0.8247208,
      "learning_rate": 2.0342730748160588e-06,
      "loss": 0.8469224,
      "num_input_tokens_seen": 91408970,
      "step": 4239,
      "time_per_iteration": 2.6513354778289795
    },
    {
      "auxiliary_loss_clip": 0.01185971,
      "auxiliary_loss_mlp": 0.01029053,
      "balance_loss_clip": 1.05340242,
      "balance_loss_mlp": 1.01921272,
      "epoch": 0.5098298563097456,
      "flos": 27745122844800.0,
      "grad_norm": 2.320239475975508,
      "language_loss": 0.70711601,
      "learning_rate": 2.033494214974006e-06,
      "loss": 0.72926617,
      "num_input_tokens_seen": 91430115,
      "step": 4240,
      "time_per_iteration": 2.6559817790985107
    },
    {
      "auxiliary_loss_clip": 0.0118504,
      "auxiliary_loss_mlp": 0.01030399,
      "balance_loss_clip": 1.05531514,
      "balance_loss_mlp": 1.02218556,
      "epoch": 0.5099500992003848,
      "flos": 21358913011200.0,
      "grad_norm": 2.119352640197743,
      "language_loss": 0.83660877,
      "learning_rate": 2.0327153500509067e-06,
      "loss": 0.85876316,
      "num_input_tokens_seen": 91449140,
      "step": 4241,
      "time_per_iteration": 2.5567572116851807
    },
    {
      "auxiliary_loss_clip": 0.01187434,
      "auxiliary_loss_mlp": 0.01027394,
      "balance_loss_clip": 1.05575347,
      "balance_loss_mlp": 1.01835203,
      "epoch": 0.5100703420910239,
      "flos": 19865999013120.0,
      "grad_norm": 2.0918384689825738,
      "language_loss": 0.8465122,
      "learning_rate": 2.031936480164916e-06,
      "loss": 0.86866051,
      "num_input_tokens_seen": 91466880,
      "step": 4242,
      "time_per_iteration": 2.589578628540039
    },
    {
      "auxiliary_loss_clip": 0.01189255,
      "auxiliary_loss_mlp": 0.01033256,
      "balance_loss_clip": 1.05495024,
      "balance_loss_mlp": 1.02433324,
      "epoch": 0.5101905849816629,
      "flos": 24648797635200.0,
      "grad_norm": 2.817027814942451,
      "language_loss": 0.80359513,
      "learning_rate": 2.0311576054341857e-06,
      "loss": 0.82582021,
      "num_input_tokens_seen": 91487495,
      "step": 4243,
      "time_per_iteration": 2.6594667434692383
    },
    {
      "auxiliary_loss_clip": 0.01187642,
      "auxiliary_loss_mlp": 0.01028959,
      "balance_loss_clip": 1.05633366,
      "balance_loss_mlp": 1.01933265,
      "epoch": 0.5103108278723021,
      "flos": 22930076787840.0,
      "grad_norm": 1.7265789402685239,
      "language_loss": 0.62730134,
      "learning_rate": 2.0303787259768715e-06,
      "loss": 0.64946729,
      "num_input_tokens_seen": 91508395,
      "step": 4244,
      "time_per_iteration": 2.5905747413635254
    },
    {
      "auxiliary_loss_clip": 0.01190369,
      "auxiliary_loss_mlp": 0.01025356,
      "balance_loss_clip": 1.05823207,
      "balance_loss_mlp": 1.01607525,
      "epoch": 0.5104310707629411,
      "flos": 21506613736320.0,
      "grad_norm": 3.406695953587381,
      "language_loss": 0.69531167,
      "learning_rate": 2.0295998419111294e-06,
      "loss": 0.71746898,
      "num_input_tokens_seen": 91525685,
      "step": 4245,
      "time_per_iteration": 2.642869472503662
    },
    {
      "auxiliary_loss_clip": 0.011857,
      "auxiliary_loss_mlp": 0.0102922,
      "balance_loss_clip": 1.05555284,
      "balance_loss_mlp": 1.0199337,
      "epoch": 0.5105513136535802,
      "flos": 14903180403840.0,
      "grad_norm": 5.1717452925447,
      "language_loss": 0.73284072,
      "learning_rate": 2.028820953355115e-06,
      "loss": 0.75498986,
      "num_input_tokens_seen": 91543785,
      "step": 4246,
      "time_per_iteration": 2.686002016067505
    },
    {
      "auxiliary_loss_clip": 0.0119005,
      "auxiliary_loss_mlp": 0.01028296,
      "balance_loss_clip": 1.05508435,
      "balance_loss_mlp": 1.01807988,
      "epoch": 0.5106715565442194,
      "flos": 22602212421120.0,
      "grad_norm": 3.6870487392728544,
      "language_loss": 0.7851004,
      "learning_rate": 2.0280420604269834e-06,
      "loss": 0.80728382,
      "num_input_tokens_seen": 91563325,
      "step": 4247,
      "time_per_iteration": 2.6368606090545654
    },
    {
      "auxiliary_loss_clip": 0.01098105,
      "auxiliary_loss_mlp": 0.01003941,
      "balance_loss_clip": 1.02478361,
      "balance_loss_mlp": 1.00232017,
      "epoch": 0.5107917994348584,
      "flos": 71027645558400.0,
      "grad_norm": 0.7144270608218471,
      "language_loss": 0.58942223,
      "learning_rate": 2.027263163244895e-06,
      "loss": 0.6104427,
      "num_input_tokens_seen": 91632450,
      "step": 4248,
      "time_per_iteration": 3.3348374366760254
    },
    {
      "auxiliary_loss_clip": 0.01186722,
      "auxiliary_loss_mlp": 0.01030138,
      "balance_loss_clip": 1.05581594,
      "balance_loss_mlp": 1.02079153,
      "epoch": 0.5109120423254975,
      "flos": 24827416992000.0,
      "grad_norm": 1.6490518139084116,
      "language_loss": 0.74595118,
      "learning_rate": 2.026484261927005e-06,
      "loss": 0.76811981,
      "num_input_tokens_seen": 91651945,
      "step": 4249,
      "time_per_iteration": 2.6787381172180176
    },
    {
      "auxiliary_loss_clip": 0.01193616,
      "auxiliary_loss_mlp": 0.01036772,
      "balance_loss_clip": 1.05917716,
      "balance_loss_mlp": 1.02718163,
      "epoch": 0.5110322852161366,
      "flos": 21247661612160.0,
      "grad_norm": 2.2526346596618105,
      "language_loss": 0.74217975,
      "learning_rate": 2.025705356591475e-06,
      "loss": 0.76448357,
      "num_input_tokens_seen": 91669635,
      "step": 4250,
      "time_per_iteration": 2.6223819255828857
    },
    {
      "auxiliary_loss_clip": 0.01092176,
      "auxiliary_loss_mlp": 0.01001747,
      "balance_loss_clip": 1.02253294,
      "balance_loss_mlp": 1.00016165,
      "epoch": 0.5111525281067757,
      "flos": 66457114358400.0,
      "grad_norm": 0.7590335293506631,
      "language_loss": 0.57964039,
      "learning_rate": 2.024926447356462e-06,
      "loss": 0.60057962,
      "num_input_tokens_seen": 91731920,
      "step": 4251,
      "time_per_iteration": 3.230369806289673
    },
    {
      "auxiliary_loss_clip": 0.01190389,
      "auxiliary_loss_mlp": 0.01026249,
      "balance_loss_clip": 1.05452681,
      "balance_loss_mlp": 1.01679564,
      "epoch": 0.5112727709974147,
      "flos": 14866731077760.0,
      "grad_norm": 3.20968602263192,
      "language_loss": 0.78747106,
      "learning_rate": 2.024147534340127e-06,
      "loss": 0.80963737,
      "num_input_tokens_seen": 91749780,
      "step": 4252,
      "time_per_iteration": 2.67020320892334
    },
    {
      "auxiliary_loss_clip": 0.01185279,
      "auxiliary_loss_mlp": 0.01026297,
      "balance_loss_clip": 1.05242324,
      "balance_loss_mlp": 1.01704597,
      "epoch": 0.5113930138880539,
      "flos": 21177600134400.0,
      "grad_norm": 1.750923061231129,
      "language_loss": 0.80129802,
      "learning_rate": 2.02336861766063e-06,
      "loss": 0.82341385,
      "num_input_tokens_seen": 91768840,
      "step": 4253,
      "time_per_iteration": 2.5928127765655518
    },
    {
      "auxiliary_loss_clip": 0.01191827,
      "auxiliary_loss_mlp": 0.0102672,
      "balance_loss_clip": 1.05799699,
      "balance_loss_mlp": 1.01686096,
      "epoch": 0.511513256778693,
      "flos": 20409111630720.0,
      "grad_norm": 1.8923831276569765,
      "language_loss": 0.78696978,
      "learning_rate": 2.0225896974361327e-06,
      "loss": 0.80915529,
      "num_input_tokens_seen": 91788945,
      "step": 4254,
      "time_per_iteration": 2.6696441173553467
    },
    {
      "auxiliary_loss_clip": 0.0108796,
      "auxiliary_loss_mlp": 0.0100195,
      "balance_loss_clip": 1.02107263,
      "balance_loss_mlp": 1.00069261,
      "epoch": 0.511633499669332,
      "flos": 69879975131520.0,
      "grad_norm": 0.8571931972695002,
      "language_loss": 0.59942997,
      "learning_rate": 2.0218107737847962e-06,
      "loss": 0.62032902,
      "num_input_tokens_seen": 91850990,
      "step": 4255,
      "time_per_iteration": 4.151374101638794
    },
    {
      "auxiliary_loss_clip": 0.01186393,
      "auxiliary_loss_mlp": 0.01028492,
      "balance_loss_clip": 1.05532992,
      "balance_loss_mlp": 1.01955736,
      "epoch": 0.5117537425599712,
      "flos": 24097855852800.0,
      "grad_norm": 2.1681395269587456,
      "language_loss": 0.74700356,
      "learning_rate": 2.0210318468247826e-06,
      "loss": 0.7691524,
      "num_input_tokens_seen": 91869960,
      "step": 4256,
      "time_per_iteration": 2.679680824279785
    },
    {
      "auxiliary_loss_clip": 0.01183463,
      "auxiliary_loss_mlp": 0.01029115,
      "balance_loss_clip": 1.05282736,
      "balance_loss_mlp": 1.0201087,
      "epoch": 0.5118739854506102,
      "flos": 20959550622720.0,
      "grad_norm": 2.605938881580977,
      "language_loss": 0.82137907,
      "learning_rate": 2.020252916674255e-06,
      "loss": 0.84350485,
      "num_input_tokens_seen": 91889075,
      "step": 4257,
      "time_per_iteration": 2.6903038024902344
    },
    {
      "auxiliary_loss_clip": 0.01188951,
      "auxiliary_loss_mlp": 0.01032942,
      "balance_loss_clip": 1.05718338,
      "balance_loss_mlp": 1.023054,
      "epoch": 0.5119942283412493,
      "flos": 17457326749440.0,
      "grad_norm": 2.137038628319388,
      "language_loss": 0.81042731,
      "learning_rate": 2.019473983451375e-06,
      "loss": 0.83264625,
      "num_input_tokens_seen": 91907495,
      "step": 4258,
      "time_per_iteration": 2.586223602294922
    },
    {
      "auxiliary_loss_clip": 0.01190075,
      "auxiliary_loss_mlp": 0.01033059,
      "balance_loss_clip": 1.05651057,
      "balance_loss_mlp": 1.02321839,
      "epoch": 0.5121144712318885,
      "flos": 21066743784960.0,
      "grad_norm": 40.943721123177376,
      "language_loss": 0.71518648,
      "learning_rate": 2.0186950472743076e-06,
      "loss": 0.73741782,
      "num_input_tokens_seen": 91927400,
      "step": 4259,
      "time_per_iteration": 3.525528907775879
    },
    {
      "auxiliary_loss_clip": 0.01187928,
      "auxiliary_loss_mlp": 0.0102732,
      "balance_loss_clip": 1.05545127,
      "balance_loss_mlp": 1.0187192,
      "epoch": 0.5122347141225275,
      "flos": 19860791541120.0,
      "grad_norm": 1.932267589353743,
      "language_loss": 0.7417661,
      "learning_rate": 2.0179161082612162e-06,
      "loss": 0.76391864,
      "num_input_tokens_seen": 91946790,
      "step": 4260,
      "time_per_iteration": 3.5691375732421875
    },
    {
      "auxiliary_loss_clip": 0.01185716,
      "auxiliary_loss_mlp": 0.01025716,
      "balance_loss_clip": 1.05228746,
      "balance_loss_mlp": 1.01572001,
      "epoch": 0.5123549570131666,
      "flos": 22528487756160.0,
      "grad_norm": 5.752960974016962,
      "language_loss": 0.72836983,
      "learning_rate": 2.017137166530266e-06,
      "loss": 0.75048417,
      "num_input_tokens_seen": 91966325,
      "step": 4261,
      "time_per_iteration": 2.6572744846343994
    },
    {
      "auxiliary_loss_clip": 0.01186822,
      "auxiliary_loss_mlp": 0.01026667,
      "balance_loss_clip": 1.05536485,
      "balance_loss_mlp": 1.01756573,
      "epoch": 0.5124751999038056,
      "flos": 20333375804160.0,
      "grad_norm": 2.252799967753973,
      "language_loss": 0.79607821,
      "learning_rate": 2.0163582221996213e-06,
      "loss": 0.81821311,
      "num_input_tokens_seen": 91984700,
      "step": 4262,
      "time_per_iteration": 3.6406383514404297
    },
    {
      "auxiliary_loss_clip": 0.0118906,
      "auxiliary_loss_mlp": 0.01029377,
      "balance_loss_clip": 1.05504549,
      "balance_loss_mlp": 1.01901174,
      "epoch": 0.5125954427944448,
      "flos": 39785970211200.0,
      "grad_norm": 3.3985579588068386,
      "language_loss": 0.67942095,
      "learning_rate": 2.015579275387446e-06,
      "loss": 0.70160532,
      "num_input_tokens_seen": 92010020,
      "step": 4263,
      "time_per_iteration": 2.7624733448028564
    },
    {
      "auxiliary_loss_clip": 0.01187197,
      "auxiliary_loss_mlp": 0.01033361,
      "balance_loss_clip": 1.05694866,
      "balance_loss_mlp": 1.02415252,
      "epoch": 0.5127156856850839,
      "flos": 29205394358400.0,
      "grad_norm": 3.7641171914329683,
      "language_loss": 0.68643343,
      "learning_rate": 2.0148003262119085e-06,
      "loss": 0.70863903,
      "num_input_tokens_seen": 92030990,
      "step": 4264,
      "time_per_iteration": 2.7481040954589844
    },
    {
      "auxiliary_loss_clip": 0.01190072,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.05599451,
      "balance_loss_mlp": 1.01818252,
      "epoch": 0.5128359285757229,
      "flos": 13553693412480.0,
      "grad_norm": 2.8336232853986036,
      "language_loss": 0.76658612,
      "learning_rate": 2.0140213747911728e-06,
      "loss": 0.78876442,
      "num_input_tokens_seen": 92049525,
      "step": 4265,
      "time_per_iteration": 2.5855441093444824
    },
    {
      "auxiliary_loss_clip": 0.01186869,
      "auxiliary_loss_mlp": 0.01027827,
      "balance_loss_clip": 1.05429363,
      "balance_loss_mlp": 1.01845157,
      "epoch": 0.5129561714663621,
      "flos": 25192089820800.0,
      "grad_norm": 2.417597844397982,
      "language_loss": 0.80514586,
      "learning_rate": 2.013242421243406e-06,
      "loss": 0.82729292,
      "num_input_tokens_seen": 92068430,
      "step": 4266,
      "time_per_iteration": 2.6815195083618164
    },
    {
      "auxiliary_loss_clip": 0.01185285,
      "auxiliary_loss_mlp": 0.01026818,
      "balance_loss_clip": 1.05517054,
      "balance_loss_mlp": 1.01745439,
      "epoch": 0.5130764143570011,
      "flos": 18150223080960.0,
      "grad_norm": 2.1876739749748184,
      "language_loss": 0.79080844,
      "learning_rate": 2.012463465686774e-06,
      "loss": 0.81292951,
      "num_input_tokens_seen": 92088180,
      "step": 4267,
      "time_per_iteration": 2.6473746299743652
    },
    {
      "auxiliary_loss_clip": 0.01084733,
      "auxiliary_loss_mlp": 0.01004141,
      "balance_loss_clip": 1.02034712,
      "balance_loss_mlp": 1.00297284,
      "epoch": 0.5131966572476402,
      "flos": 59794896418560.0,
      "grad_norm": 0.7769226436369217,
      "language_loss": 0.54696,
      "learning_rate": 2.0116845082394446e-06,
      "loss": 0.56784874,
      "num_input_tokens_seen": 92153015,
      "step": 4268,
      "time_per_iteration": 3.2421391010284424
    },
    {
      "auxiliary_loss_clip": 0.01188423,
      "auxiliary_loss_mlp": 0.01029704,
      "balance_loss_clip": 1.05639136,
      "balance_loss_mlp": 1.02058411,
      "epoch": 0.5133169001382794,
      "flos": 18515219132160.0,
      "grad_norm": 2.4125439378217317,
      "language_loss": 0.78732491,
      "learning_rate": 2.0109055490195836e-06,
      "loss": 0.80950618,
      "num_input_tokens_seen": 92171470,
      "step": 4269,
      "time_per_iteration": 2.624600887298584
    },
    {
      "auxiliary_loss_clip": 0.01186022,
      "auxiliary_loss_mlp": 0.01023034,
      "balance_loss_clip": 1.05303478,
      "balance_loss_mlp": 1.01431394,
      "epoch": 0.5134371430289184,
      "flos": 15523537219200.0,
      "grad_norm": 4.037350469939436,
      "language_loss": 0.64496845,
      "learning_rate": 2.0101265881453605e-06,
      "loss": 0.66705906,
      "num_input_tokens_seen": 92189945,
      "step": 4270,
      "time_per_iteration": 2.575106143951416
    },
    {
      "auxiliary_loss_clip": 0.01188262,
      "auxiliary_loss_mlp": 0.01025651,
      "balance_loss_clip": 1.05875611,
      "balance_loss_mlp": 1.01663256,
      "epoch": 0.5135573859195575,
      "flos": 21433786911360.0,
      "grad_norm": 2.9826543801827605,
      "language_loss": 0.77986085,
      "learning_rate": 2.009347625734941e-06,
      "loss": 0.80199993,
      "num_input_tokens_seen": 92209855,
      "step": 4271,
      "time_per_iteration": 2.620882749557495
    },
    {
      "auxiliary_loss_clip": 0.01193049,
      "auxiliary_loss_mlp": 0.0102593,
      "balance_loss_clip": 1.05859792,
      "balance_loss_mlp": 1.01684046,
      "epoch": 0.5136776288101966,
      "flos": 17712651600000.0,
      "grad_norm": 3.668742881124715,
      "language_loss": 0.75119352,
      "learning_rate": 2.0085686619064954e-06,
      "loss": 0.77338326,
      "num_input_tokens_seen": 92226295,
      "step": 4272,
      "time_per_iteration": 2.582627058029175
    },
    {
      "auxiliary_loss_clip": 0.01189798,
      "auxiliary_loss_mlp": 0.01029361,
      "balance_loss_clip": 1.05841339,
      "balance_loss_mlp": 1.02002072,
      "epoch": 0.5137978717008357,
      "flos": 16581680997120.0,
      "grad_norm": 2.3217421639924285,
      "language_loss": 0.82591814,
      "learning_rate": 2.00778969677819e-06,
      "loss": 0.84810972,
      "num_input_tokens_seen": 92243330,
      "step": 4273,
      "time_per_iteration": 2.634982109069824
    },
    {
      "auxiliary_loss_clip": 0.01184005,
      "auxiliary_loss_mlp": 0.01030148,
      "balance_loss_clip": 1.05336702,
      "balance_loss_mlp": 1.02114761,
      "epoch": 0.5139181145914747,
      "flos": 20668243322880.0,
      "grad_norm": 2.0825084242670777,
      "language_loss": 0.64049423,
      "learning_rate": 2.0070107304681934e-06,
      "loss": 0.6626358,
      "num_input_tokens_seen": 92262285,
      "step": 4274,
      "time_per_iteration": 2.602396249771118
    },
    {
      "auxiliary_loss_clip": 0.01185491,
      "auxiliary_loss_mlp": 0.0102785,
      "balance_loss_clip": 1.05420077,
      "balance_loss_mlp": 1.01936841,
      "epoch": 0.5140383574821139,
      "flos": 32926996546560.0,
      "grad_norm": 2.3587849864371724,
      "language_loss": 0.78353751,
      "learning_rate": 2.006231763094675e-06,
      "loss": 0.80567092,
      "num_input_tokens_seen": 92283305,
      "step": 4275,
      "time_per_iteration": 2.7194409370422363
    },
    {
      "auxiliary_loss_clip": 0.01185612,
      "auxiliary_loss_mlp": 0.01028672,
      "balance_loss_clip": 1.05773008,
      "balance_loss_mlp": 1.01942766,
      "epoch": 0.514158600372753,
      "flos": 19537093152000.0,
      "grad_norm": 2.9309238473957846,
      "language_loss": 0.87513965,
      "learning_rate": 2.0054527947758027e-06,
      "loss": 0.89728248,
      "num_input_tokens_seen": 92302105,
      "step": 4276,
      "time_per_iteration": 2.59159255027771
    },
    {
      "auxiliary_loss_clip": 0.01085357,
      "auxiliary_loss_mlp": 0.0100128,
      "balance_loss_clip": 1.0209353,
      "balance_loss_mlp": 1.00023675,
      "epoch": 0.514278843263392,
      "flos": 62523855279360.0,
      "grad_norm": 0.7279918095637039,
      "language_loss": 0.55837536,
      "learning_rate": 2.004673825629746e-06,
      "loss": 0.57924175,
      "num_input_tokens_seen": 92362885,
      "step": 4277,
      "time_per_iteration": 3.211117744445801
    },
    {
      "auxiliary_loss_clip": 0.0118496,
      "auxiliary_loss_mlp": 0.01023673,
      "balance_loss_clip": 1.05393493,
      "balance_loss_mlp": 1.01489878,
      "epoch": 0.5143990861540312,
      "flos": 25882328545920.0,
      "grad_norm": 1.7447758737620918,
      "language_loss": 0.72542477,
      "learning_rate": 2.0038948557746744e-06,
      "loss": 0.74751109,
      "num_input_tokens_seen": 92384740,
      "step": 4278,
      "time_per_iteration": 2.6411802768707275
    },
    {
      "auxiliary_loss_clip": 0.0118397,
      "auxiliary_loss_mlp": 0.01024393,
      "balance_loss_clip": 1.05369449,
      "balance_loss_mlp": 1.01517785,
      "epoch": 0.5145193290446702,
      "flos": 23330660238720.0,
      "grad_norm": 4.201214062769527,
      "language_loss": 0.74977785,
      "learning_rate": 2.0031158853287558e-06,
      "loss": 0.77186143,
      "num_input_tokens_seen": 92405175,
      "step": 4279,
      "time_per_iteration": 2.7069644927978516
    },
    {
      "auxiliary_loss_clip": 0.01187578,
      "auxiliary_loss_mlp": 0.01031434,
      "balance_loss_clip": 1.05609381,
      "balance_loss_mlp": 1.02228487,
      "epoch": 0.5146395719353093,
      "flos": 22856603518080.0,
      "grad_norm": 3.502734098118463,
      "language_loss": 0.70623696,
      "learning_rate": 2.0023369144101593e-06,
      "loss": 0.72842705,
      "num_input_tokens_seen": 92423345,
      "step": 4280,
      "time_per_iteration": 2.6770081520080566
    },
    {
      "auxiliary_loss_clip": 0.01186321,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.05328345,
      "balance_loss_mlp": 1.01955903,
      "epoch": 0.5147598148259485,
      "flos": 26391577616640.0,
      "grad_norm": 3.1866334616392513,
      "language_loss": 0.77049923,
      "learning_rate": 2.0015579431370555e-06,
      "loss": 0.79265249,
      "num_input_tokens_seen": 92445025,
      "step": 4281,
      "time_per_iteration": 2.667390823364258
    },
    {
      "auxiliary_loss_clip": 0.01186704,
      "auxiliary_loss_mlp": 0.01033787,
      "balance_loss_clip": 1.05698848,
      "balance_loss_mlp": 1.02532315,
      "epoch": 0.5148800577165875,
      "flos": 29965694561280.0,
      "grad_norm": 2.698588473794946,
      "language_loss": 0.70124626,
      "learning_rate": 2.000778971627612e-06,
      "loss": 0.72345114,
      "num_input_tokens_seen": 92464490,
      "step": 4282,
      "time_per_iteration": 3.7092883586883545
    },
    {
      "auxiliary_loss_clip": 0.01184041,
      "auxiliary_loss_mlp": 0.01033851,
      "balance_loss_clip": 1.05430782,
      "balance_loss_mlp": 1.02420676,
      "epoch": 0.5150003006072266,
      "flos": 17931383470080.0,
      "grad_norm": 2.948995223665226,
      "language_loss": 0.90231657,
      "learning_rate": 2e-06,
      "loss": 0.92449552,
      "num_input_tokens_seen": 92482085,
      "step": 4283,
      "time_per_iteration": 2.574784755706787
    },
    {
      "auxiliary_loss_clip": 0.01183707,
      "auxiliary_loss_mlp": 0.01023864,
      "balance_loss_clip": 1.05364585,
      "balance_loss_mlp": 1.01517928,
      "epoch": 0.5151205434978657,
      "flos": 18478733892480.0,
      "grad_norm": 4.683981775855628,
      "language_loss": 0.85785699,
      "learning_rate": 1.9992210283723878e-06,
      "loss": 0.8799327,
      "num_input_tokens_seen": 92499325,
      "step": 4284,
      "time_per_iteration": 2.6886792182922363
    },
    {
      "auxiliary_loss_clip": 0.01184369,
      "auxiliary_loss_mlp": 0.0102911,
      "balance_loss_clip": 1.05503464,
      "balance_loss_mlp": 1.0207355,
      "epoch": 0.5152407863885048,
      "flos": 25341263003520.0,
      "grad_norm": 2.036734558680635,
      "language_loss": 0.7967236,
      "learning_rate": 1.9984420568629448e-06,
      "loss": 0.81885839,
      "num_input_tokens_seen": 92522090,
      "step": 4285,
      "time_per_iteration": 2.622946262359619
    },
    {
      "auxiliary_loss_clip": 0.01187348,
      "auxiliary_loss_mlp": 0.01030059,
      "balance_loss_clip": 1.05535698,
      "balance_loss_mlp": 1.02099895,
      "epoch": 0.5153610292791438,
      "flos": 18329740277760.0,
      "grad_norm": 7.316492402760561,
      "language_loss": 0.7891627,
      "learning_rate": 1.9976630855898405e-06,
      "loss": 0.81133676,
      "num_input_tokens_seen": 92539845,
      "step": 4286,
      "time_per_iteration": 3.546104669570923
    },
    {
      "auxiliary_loss_clip": 0.01179381,
      "auxiliary_loss_mlp": 0.0102482,
      "balance_loss_clip": 1.05213451,
      "balance_loss_mlp": 1.01560473,
      "epoch": 0.515481272169783,
      "flos": 30409945971840.0,
      "grad_norm": 2.385550674058364,
      "language_loss": 0.74591064,
      "learning_rate": 1.9968841146712445e-06,
      "loss": 0.76795262,
      "num_input_tokens_seen": 92559460,
      "step": 4287,
      "time_per_iteration": 2.699084758758545
    },
    {
      "auxiliary_loss_clip": 0.01187736,
      "auxiliary_loss_mlp": 0.01028393,
      "balance_loss_clip": 1.05502474,
      "balance_loss_mlp": 1.01913071,
      "epoch": 0.5156015150604221,
      "flos": 23037305863680.0,
      "grad_norm": 1.789935002360248,
      "language_loss": 0.71421945,
      "learning_rate": 1.996105144225326e-06,
      "loss": 0.7363807,
      "num_input_tokens_seen": 92579695,
      "step": 4288,
      "time_per_iteration": 3.5274977684020996
    },
    {
      "auxiliary_loss_clip": 0.0118614,
      "auxiliary_loss_mlp": 0.01027736,
      "balance_loss_clip": 1.05611026,
      "balance_loss_mlp": 1.0191468,
      "epoch": 0.5157217579510611,
      "flos": 17858556645120.0,
      "grad_norm": 3.260710892768341,
      "language_loss": 0.78681809,
      "learning_rate": 1.995326174370254e-06,
      "loss": 0.80895686,
      "num_input_tokens_seen": 92598795,
      "step": 4289,
      "time_per_iteration": 3.516402244567871
    },
    {
      "auxiliary_loss_clip": 0.01178921,
      "auxiliary_loss_mlp": 0.01024542,
      "balance_loss_clip": 1.05046976,
      "balance_loss_mlp": 1.01569128,
      "epoch": 0.5158420008417003,
      "flos": 19171486569600.0,
      "grad_norm": 2.384955060659728,
      "language_loss": 0.73013169,
      "learning_rate": 1.994547205224197e-06,
      "loss": 0.75216633,
      "num_input_tokens_seen": 92617700,
      "step": 4290,
      "time_per_iteration": 2.6787188053131104
    },
    {
      "auxiliary_loss_clip": 0.01184102,
      "auxiliary_loss_mlp": 0.01026693,
      "balance_loss_clip": 1.05373108,
      "balance_loss_mlp": 1.01816952,
      "epoch": 0.5159622437323393,
      "flos": 22419534827520.0,
      "grad_norm": 5.3746503805220796,
      "language_loss": 0.67607796,
      "learning_rate": 1.993768236905325e-06,
      "loss": 0.69818592,
      "num_input_tokens_seen": 92638370,
      "step": 4291,
      "time_per_iteration": 2.6361496448516846
    },
    {
      "auxiliary_loss_clip": 0.01181156,
      "auxiliary_loss_mlp": 0.01024385,
      "balance_loss_clip": 1.05160308,
      "balance_loss_mlp": 1.01574278,
      "epoch": 0.5160824866229784,
      "flos": 24603010773120.0,
      "grad_norm": 4.300606307773687,
      "language_loss": 0.66016895,
      "learning_rate": 1.992989269531807e-06,
      "loss": 0.68222433,
      "num_input_tokens_seen": 92657180,
      "step": 4292,
      "time_per_iteration": 2.6717841625213623
    },
    {
      "auxiliary_loss_clip": 0.01183419,
      "auxiliary_loss_mlp": 0.01034163,
      "balance_loss_clip": 1.05284905,
      "balance_loss_mlp": 1.02507877,
      "epoch": 0.5162027295136175,
      "flos": 18002737837440.0,
      "grad_norm": 3.1415996421223182,
      "language_loss": 0.68036556,
      "learning_rate": 1.99221030322181e-06,
      "loss": 0.70254135,
      "num_input_tokens_seen": 92673985,
      "step": 4293,
      "time_per_iteration": 2.6003661155700684
    },
    {
      "auxiliary_loss_clip": 0.01181219,
      "auxiliary_loss_mlp": 0.01027452,
      "balance_loss_clip": 1.05241501,
      "balance_loss_mlp": 1.01865458,
      "epoch": 0.5163229724042566,
      "flos": 27344611221120.0,
      "grad_norm": 2.8831199478917564,
      "language_loss": 0.81114852,
      "learning_rate": 1.991431338093505e-06,
      "loss": 0.83323526,
      "num_input_tokens_seen": 92696340,
      "step": 4294,
      "time_per_iteration": 2.666588068008423
    },
    {
      "auxiliary_loss_clip": 0.01184478,
      "auxiliary_loss_mlp": 0.01024633,
      "balance_loss_clip": 1.05593204,
      "balance_loss_mlp": 1.01606202,
      "epoch": 0.5164432152948957,
      "flos": 21762764599680.0,
      "grad_norm": 2.3537352057203083,
      "language_loss": 0.78983903,
      "learning_rate": 1.9906523742650587e-06,
      "loss": 0.81193006,
      "num_input_tokens_seen": 92715200,
      "step": 4295,
      "time_per_iteration": 2.6215450763702393
    },
    {
      "auxiliary_loss_clip": 0.01183031,
      "auxiliary_loss_mlp": 0.01028516,
      "balance_loss_clip": 1.05060887,
      "balance_loss_mlp": 1.01960492,
      "epoch": 0.5165634581855347,
      "flos": 25550334115200.0,
      "grad_norm": 2.431955537461845,
      "language_loss": 0.77437663,
      "learning_rate": 1.9898734118546397e-06,
      "loss": 0.7964921,
      "num_input_tokens_seen": 92735150,
      "step": 4296,
      "time_per_iteration": 2.6364378929138184
    },
    {
      "auxiliary_loss_clip": 0.01185054,
      "auxiliary_loss_mlp": 0.01029653,
      "balance_loss_clip": 1.05473006,
      "balance_loss_mlp": 1.02107,
      "epoch": 0.5166837010761739,
      "flos": 19901191363200.0,
      "grad_norm": 2.701850086990341,
      "language_loss": 0.80360854,
      "learning_rate": 1.989094450980416e-06,
      "loss": 0.8257556,
      "num_input_tokens_seen": 92755250,
      "step": 4297,
      "time_per_iteration": 2.5938525199890137
    },
    {
      "auxiliary_loss_clip": 0.01181617,
      "auxiliary_loss_mlp": 0.01025371,
      "balance_loss_clip": 1.05169702,
      "balance_loss_mlp": 1.01657319,
      "epoch": 0.516803943966813,
      "flos": 26646076454400.0,
      "grad_norm": 2.933212564529706,
      "language_loss": 0.76574886,
      "learning_rate": 1.9883154917605556e-06,
      "loss": 0.78781873,
      "num_input_tokens_seen": 92774460,
      "step": 4298,
      "time_per_iteration": 2.7241744995117188
    },
    {
      "auxiliary_loss_clip": 0.01180093,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 1.05134892,
      "balance_loss_mlp": 1.01721215,
      "epoch": 0.516924186857452,
      "flos": 19682854542720.0,
      "grad_norm": 1.9946170309433324,
      "language_loss": 0.8328737,
      "learning_rate": 1.9875365343132262e-06,
      "loss": 0.85493386,
      "num_input_tokens_seen": 92791580,
      "step": 4299,
      "time_per_iteration": 2.609396457672119
    },
    {
      "auxiliary_loss_clip": 0.01184628,
      "auxiliary_loss_mlp": 0.01027683,
      "balance_loss_clip": 1.0542376,
      "balance_loss_mlp": 1.01892149,
      "epoch": 0.5170444297480912,
      "flos": 15956583586560.0,
      "grad_norm": 2.787709743522977,
      "language_loss": 0.84577155,
      "learning_rate": 1.9867575787565946e-06,
      "loss": 0.86789465,
      "num_input_tokens_seen": 92806240,
      "step": 4300,
      "time_per_iteration": 2.5846002101898193
    },
    {
      "auxiliary_loss_clip": 0.01184667,
      "auxiliary_loss_mlp": 0.01028887,
      "balance_loss_clip": 1.05396223,
      "balance_loss_mlp": 1.01920104,
      "epoch": 0.5171646726387302,
      "flos": 14174157968640.0,
      "grad_norm": 3.183373942427346,
      "language_loss": 0.85971117,
      "learning_rate": 1.9859786252088275e-06,
      "loss": 0.88184673,
      "num_input_tokens_seen": 92823420,
      "step": 4301,
      "time_per_iteration": 2.5643296241760254
    },
    {
      "auxiliary_loss_clip": 0.01187625,
      "auxiliary_loss_mlp": 0.01026746,
      "balance_loss_clip": 1.05423474,
      "balance_loss_mlp": 1.01846099,
      "epoch": 0.5172849155293693,
      "flos": 23578550974080.0,
      "grad_norm": 3.909723249687514,
      "language_loss": 0.6649543,
      "learning_rate": 1.9851996737880914e-06,
      "loss": 0.68709803,
      "num_input_tokens_seen": 92838605,
      "step": 4302,
      "time_per_iteration": 2.6579251289367676
    },
    {
      "auxiliary_loss_clip": 0.0118626,
      "auxiliary_loss_mlp": 0.01031852,
      "balance_loss_clip": 1.05356109,
      "balance_loss_mlp": 1.02263725,
      "epoch": 0.5174051584200084,
      "flos": 14283541860480.0,
      "grad_norm": 2.0741655980646607,
      "language_loss": 0.74308091,
      "learning_rate": 1.9844207246125537e-06,
      "loss": 0.76526201,
      "num_input_tokens_seen": 92855185,
      "step": 4303,
      "time_per_iteration": 2.5466928482055664
    },
    {
      "auxiliary_loss_clip": 0.01181607,
      "auxiliary_loss_mlp": 0.0102706,
      "balance_loss_clip": 1.05383897,
      "balance_loss_mlp": 1.01802945,
      "epoch": 0.5175254013106475,
      "flos": 37889384192640.0,
      "grad_norm": 2.8639880826315895,
      "language_loss": 0.68525982,
      "learning_rate": 1.983641777800379e-06,
      "loss": 0.70734644,
      "num_input_tokens_seen": 92877830,
      "step": 4304,
      "time_per_iteration": 2.743872880935669
    },
    {
      "auxiliary_loss_clip": 0.01082965,
      "auxiliary_loss_mlp": 0.01000841,
      "balance_loss_clip": 1.01818347,
      "balance_loss_mlp": 0.99984533,
      "epoch": 0.5176456442012866,
      "flos": 68549737829760.0,
      "grad_norm": 0.7437327211101652,
      "language_loss": 0.58759588,
      "learning_rate": 1.9828628334697343e-06,
      "loss": 0.60843396,
      "num_input_tokens_seen": 92945040,
      "step": 4305,
      "time_per_iteration": 3.3368725776672363
    },
    {
      "auxiliary_loss_clip": 0.01083343,
      "auxiliary_loss_mlp": 0.01001913,
      "balance_loss_clip": 1.01842201,
      "balance_loss_mlp": 1.00088739,
      "epoch": 0.5177658870919257,
      "flos": 64084137235200.0,
      "grad_norm": 0.7623447460613092,
      "language_loss": 0.5465858,
      "learning_rate": 1.982083891738784e-06,
      "loss": 0.56743836,
      "num_input_tokens_seen": 93005910,
      "step": 4306,
      "time_per_iteration": 3.239051580429077
    },
    {
      "auxiliary_loss_clip": 0.01178894,
      "auxiliary_loss_mlp": 0.01029022,
      "balance_loss_clip": 1.05248022,
      "balance_loss_mlp": 1.02076101,
      "epoch": 0.5178861299825648,
      "flos": 26651248012800.0,
      "grad_norm": 1.7684457153496211,
      "language_loss": 0.82775718,
      "learning_rate": 1.9813049527256923e-06,
      "loss": 0.84983635,
      "num_input_tokens_seen": 93026305,
      "step": 4307,
      "time_per_iteration": 2.6449313163757324
    },
    {
      "auxiliary_loss_clip": 0.01184515,
      "auxiliary_loss_mlp": 0.01027659,
      "balance_loss_clip": 1.05309534,
      "balance_loss_mlp": 1.01834881,
      "epoch": 0.5180063728732038,
      "flos": 17931886260480.0,
      "grad_norm": 4.023328429643238,
      "language_loss": 0.82374257,
      "learning_rate": 1.9805260165486252e-06,
      "loss": 0.8458643,
      "num_input_tokens_seen": 93045675,
      "step": 4308,
      "time_per_iteration": 3.537696599960327
    },
    {
      "auxiliary_loss_clip": 0.01182536,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.05285597,
      "balance_loss_mlp": 1.01645398,
      "epoch": 0.518126615763843,
      "flos": 19500895221120.0,
      "grad_norm": 2.4147034193773864,
      "language_loss": 0.86582875,
      "learning_rate": 1.9797470833257457e-06,
      "loss": 0.88790441,
      "num_input_tokens_seen": 93065375,
      "step": 4309,
      "time_per_iteration": 2.603299379348755
    },
    {
      "auxiliary_loss_clip": 0.01184698,
      "auxiliary_loss_mlp": 0.01028515,
      "balance_loss_clip": 1.05553794,
      "balance_loss_mlp": 1.01986647,
      "epoch": 0.5182468586544821,
      "flos": 20704082117760.0,
      "grad_norm": 2.5756852812979645,
      "language_loss": 0.76976013,
      "learning_rate": 1.9789681531752177e-06,
      "loss": 0.79189229,
      "num_input_tokens_seen": 93085595,
      "step": 4310,
      "time_per_iteration": 2.6553757190704346
    },
    {
      "auxiliary_loss_clip": 0.01179367,
      "auxiliary_loss_mlp": 0.01022912,
      "balance_loss_clip": 1.05169153,
      "balance_loss_mlp": 1.01453805,
      "epoch": 0.5183671015451211,
      "flos": 23112107936640.0,
      "grad_norm": 3.247892094288944,
      "language_loss": 0.72764504,
      "learning_rate": 1.978189226215204e-06,
      "loss": 0.74966776,
      "num_input_tokens_seen": 93106140,
      "step": 4311,
      "time_per_iteration": 3.5337743759155273
    },
    {
      "auxiliary_loss_clip": 0.01182325,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 1.05278718,
      "balance_loss_mlp": 1.01840878,
      "epoch": 0.5184873444357603,
      "flos": 17597090568960.0,
      "grad_norm": 2.1436129528910817,
      "language_loss": 0.77369905,
      "learning_rate": 1.9774103025638675e-06,
      "loss": 0.79579264,
      "num_input_tokens_seen": 93124265,
      "step": 4312,
      "time_per_iteration": 2.6017308235168457
    },
    {
      "auxiliary_loss_clip": 0.01185215,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 1.0554812,
      "balance_loss_mlp": 1.01965261,
      "epoch": 0.5186075873263993,
      "flos": 24936800883840.0,
      "grad_norm": 1.8123529402879204,
      "language_loss": 0.76666892,
      "learning_rate": 1.9766313823393696e-06,
      "loss": 0.78880507,
      "num_input_tokens_seen": 93145130,
      "step": 4313,
      "time_per_iteration": 2.7522530555725098
    },
    {
      "auxiliary_loss_clip": 0.01180368,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.05206454,
      "balance_loss_mlp": 1.01729918,
      "epoch": 0.5187278302170384,
      "flos": 15190106244480.0,
      "grad_norm": 2.2377930733396267,
      "language_loss": 0.69027478,
      "learning_rate": 1.975852465659873e-06,
      "loss": 0.71233964,
      "num_input_tokens_seen": 93161110,
      "step": 4314,
      "time_per_iteration": 3.503753423690796
    },
    {
      "auxiliary_loss_clip": 0.01186725,
      "auxiliary_loss_mlp": 0.01030871,
      "balance_loss_clip": 1.05432868,
      "balance_loss_mlp": 1.02210355,
      "epoch": 0.5188480731076776,
      "flos": 25009412227200.0,
      "grad_norm": 3.0860136519596506,
      "language_loss": 0.70143449,
      "learning_rate": 1.9750735526435377e-06,
      "loss": 0.72361046,
      "num_input_tokens_seen": 93178055,
      "step": 4315,
      "time_per_iteration": 3.5693392753601074
    },
    {
      "auxiliary_loss_clip": 0.01182555,
      "auxiliary_loss_mlp": 0.010255,
      "balance_loss_clip": 1.05235672,
      "balance_loss_mlp": 1.0172565,
      "epoch": 0.5189683159983166,
      "flos": 24790141653120.0,
      "grad_norm": 2.3801465154378647,
      "language_loss": 0.7966857,
      "learning_rate": 1.974294643408525e-06,
      "loss": 0.81876624,
      "num_input_tokens_seen": 93195850,
      "step": 4316,
      "time_per_iteration": 2.6300442218780518
    },
    {
      "auxiliary_loss_clip": 0.01183649,
      "auxiliary_loss_mlp": 0.01030423,
      "balance_loss_clip": 1.05345929,
      "balance_loss_mlp": 1.02127373,
      "epoch": 0.5190885588889557,
      "flos": 24754266944640.0,
      "grad_norm": 2.488973381492028,
      "language_loss": 0.67293799,
      "learning_rate": 1.9735157380729947e-06,
      "loss": 0.69507867,
      "num_input_tokens_seen": 93216260,
      "step": 4317,
      "time_per_iteration": 2.7176342010498047
    },
    {
      "auxiliary_loss_clip": 0.01181917,
      "auxiliary_loss_mlp": 0.01027023,
      "balance_loss_clip": 1.05372322,
      "balance_loss_mlp": 1.0184164,
      "epoch": 0.5192088017795948,
      "flos": 24712646060160.0,
      "grad_norm": 3.3125394027263946,
      "language_loss": 0.84140933,
      "learning_rate": 1.9727368367551053e-06,
      "loss": 0.86349869,
      "num_input_tokens_seen": 93234810,
      "step": 4318,
      "time_per_iteration": 2.6251797676086426
    },
    {
      "auxiliary_loss_clip": 0.01183058,
      "auxiliary_loss_mlp": 0.01024595,
      "balance_loss_clip": 1.05238473,
      "balance_loss_mlp": 1.01610684,
      "epoch": 0.5193290446702339,
      "flos": 27229588894080.0,
      "grad_norm": 2.272134398125069,
      "language_loss": 0.68822098,
      "learning_rate": 1.9719579395730164e-06,
      "loss": 0.71029747,
      "num_input_tokens_seen": 93254185,
      "step": 4319,
      "time_per_iteration": 2.665964365005493
    },
    {
      "auxiliary_loss_clip": 0.01180959,
      "auxiliary_loss_mlp": 0.01032993,
      "balance_loss_clip": 1.05252516,
      "balance_loss_mlp": 1.02415967,
      "epoch": 0.5194492875608729,
      "flos": 11473352392320.0,
      "grad_norm": 5.934877191737592,
      "language_loss": 0.93934834,
      "learning_rate": 1.9711790466448854e-06,
      "loss": 0.96148789,
      "num_input_tokens_seen": 93268205,
      "step": 4320,
      "time_per_iteration": 2.607588052749634
    },
    {
      "auxiliary_loss_clip": 0.01187368,
      "auxiliary_loss_mlp": 0.01031664,
      "balance_loss_clip": 1.05446398,
      "balance_loss_mlp": 1.02277136,
      "epoch": 0.5195695304515121,
      "flos": 20338906498560.0,
      "grad_norm": 6.492277754697397,
      "language_loss": 0.71385169,
      "learning_rate": 1.9704001580888704e-06,
      "loss": 0.73604202,
      "num_input_tokens_seen": 93286945,
      "step": 4321,
      "time_per_iteration": 2.67257022857666
    },
    {
      "auxiliary_loss_clip": 0.01178742,
      "auxiliary_loss_mlp": 0.01025968,
      "balance_loss_clip": 1.0507493,
      "balance_loss_mlp": 1.01695001,
      "epoch": 0.5196897733421512,
      "flos": 20048317470720.0,
      "grad_norm": 6.54328769510025,
      "language_loss": 0.86657798,
      "learning_rate": 1.9696212740231283e-06,
      "loss": 0.88862509,
      "num_input_tokens_seen": 93305595,
      "step": 4322,
      "time_per_iteration": 2.627859592437744
    },
    {
      "auxiliary_loss_clip": 0.01185533,
      "auxiliary_loss_mlp": 0.0102715,
      "balance_loss_clip": 1.05245185,
      "balance_loss_mlp": 1.01772618,
      "epoch": 0.5198100162327902,
      "flos": 23805507058560.0,
      "grad_norm": 2.2921916538995815,
      "language_loss": 0.82163322,
      "learning_rate": 1.9688423945658146e-06,
      "loss": 0.84376007,
      "num_input_tokens_seen": 93326460,
      "step": 4323,
      "time_per_iteration": 2.6669087409973145
    },
    {
      "auxiliary_loss_clip": 0.01181203,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.05125809,
      "balance_loss_mlp": 1.01948178,
      "epoch": 0.5199302591234293,
      "flos": 24023951619840.0,
      "grad_norm": 3.152749238703709,
      "language_loss": 0.71927798,
      "learning_rate": 1.9680635198350845e-06,
      "loss": 0.74137276,
      "num_input_tokens_seen": 93346170,
      "step": 4324,
      "time_per_iteration": 2.731935501098633
    },
    {
      "auxiliary_loss_clip": 0.0118341,
      "auxiliary_loss_mlp": 0.01029693,
      "balance_loss_clip": 1.05310404,
      "balance_loss_mlp": 1.02050769,
      "epoch": 0.5200505020140684,
      "flos": 26359366095360.0,
      "grad_norm": 2.494909072019263,
      "language_loss": 0.72646314,
      "learning_rate": 1.967284649949093e-06,
      "loss": 0.74859416,
      "num_input_tokens_seen": 93365380,
      "step": 4325,
      "time_per_iteration": 2.6454570293426514
    },
    {
      "auxiliary_loss_clip": 0.01180556,
      "auxiliary_loss_mlp": 0.01024189,
      "balance_loss_clip": 1.05081201,
      "balance_loss_mlp": 1.01593971,
      "epoch": 0.5201707449047075,
      "flos": 39604262284800.0,
      "grad_norm": 2.7085991055027874,
      "language_loss": 0.7270515,
      "learning_rate": 1.966505785025994e-06,
      "loss": 0.74909902,
      "num_input_tokens_seen": 93387285,
      "step": 4326,
      "time_per_iteration": 2.791578769683838
    },
    {
      "auxiliary_loss_clip": 0.01183703,
      "auxiliary_loss_mlp": 0.01027776,
      "balance_loss_clip": 1.05301547,
      "balance_loss_mlp": 1.01949072,
      "epoch": 0.5202909877953465,
      "flos": 53682788292480.0,
      "grad_norm": 1.8245958959089958,
      "language_loss": 0.75872111,
      "learning_rate": 1.965726925183941e-06,
      "loss": 0.78083593,
      "num_input_tokens_seen": 93410390,
      "step": 4327,
      "time_per_iteration": 2.8636891841888428
    },
    {
      "auxiliary_loss_clip": 0.01182622,
      "auxiliary_loss_mlp": 0.01033614,
      "balance_loss_clip": 1.053231,
      "balance_loss_mlp": 1.02439332,
      "epoch": 0.5204112306859857,
      "flos": 19537021324800.0,
      "grad_norm": 2.0146398771580616,
      "language_loss": 0.8512997,
      "learning_rate": 1.964948070541087e-06,
      "loss": 0.87346208,
      "num_input_tokens_seen": 93429050,
      "step": 4328,
      "time_per_iteration": 2.5868659019470215
    },
    {
      "auxiliary_loss_clip": 0.01179265,
      "auxiliary_loss_mlp": 0.01029468,
      "balance_loss_clip": 1.05038249,
      "balance_loss_mlp": 1.02134943,
      "epoch": 0.5205314735766248,
      "flos": 15304697608320.0,
      "grad_norm": 2.7002711018010324,
      "language_loss": 0.70060372,
      "learning_rate": 1.9641692212155816e-06,
      "loss": 0.72269106,
      "num_input_tokens_seen": 93446815,
      "step": 4329,
      "time_per_iteration": 2.6204171180725098
    },
    {
      "auxiliary_loss_clip": 0.01182579,
      "auxiliary_loss_mlp": 0.01026805,
      "balance_loss_clip": 1.05375004,
      "balance_loss_mlp": 1.01836538,
      "epoch": 0.5206517164672638,
      "flos": 59263701160320.0,
      "grad_norm": 2.341042089523334,
      "language_loss": 0.72521305,
      "learning_rate": 1.9633903773255777e-06,
      "loss": 0.74730682,
      "num_input_tokens_seen": 93469130,
      "step": 4330,
      "time_per_iteration": 2.9321889877319336
    },
    {
      "auxiliary_loss_clip": 0.0118041,
      "auxiliary_loss_mlp": 0.01028415,
      "balance_loss_clip": 1.052037,
      "balance_loss_mlp": 1.01959348,
      "epoch": 0.520771959357903,
      "flos": 26871129118080.0,
      "grad_norm": 1.6148209764356178,
      "language_loss": 0.74730289,
      "learning_rate": 1.9626115389892237e-06,
      "loss": 0.76939118,
      "num_input_tokens_seen": 93489920,
      "step": 4331,
      "time_per_iteration": 2.691171407699585
    },
    {
      "auxiliary_loss_clip": 0.01183373,
      "auxiliary_loss_mlp": 0.01026885,
      "balance_loss_clip": 1.05387425,
      "balance_loss_mlp": 1.01898158,
      "epoch": 0.520892202248542,
      "flos": 26907075653760.0,
      "grad_norm": 2.6035440727427153,
      "language_loss": 0.85533226,
      "learning_rate": 1.96183270632467e-06,
      "loss": 0.87743479,
      "num_input_tokens_seen": 93509770,
      "step": 4332,
      "time_per_iteration": 2.625643014907837
    },
    {
      "auxiliary_loss_clip": 0.01182474,
      "auxiliary_loss_mlp": 0.01027489,
      "balance_loss_clip": 1.052001,
      "balance_loss_mlp": 1.01836395,
      "epoch": 0.5210124451391811,
      "flos": 25849434666240.0,
      "grad_norm": 1.9076733873625584,
      "language_loss": 0.79127157,
      "learning_rate": 1.9610538794500644e-06,
      "loss": 0.81337124,
      "num_input_tokens_seen": 93529320,
      "step": 4333,
      "time_per_iteration": 2.620067596435547
    },
    {
      "auxiliary_loss_clip": 0.01083418,
      "auxiliary_loss_mlp": 0.01003375,
      "balance_loss_clip": 1.01692617,
      "balance_loss_mlp": 1.00215268,
      "epoch": 0.5211326880298203,
      "flos": 70553804319360.0,
      "grad_norm": 0.7979856942590564,
      "language_loss": 0.59427202,
      "learning_rate": 1.9602750584835542e-06,
      "loss": 0.61513996,
      "num_input_tokens_seen": 93595255,
      "step": 4334,
      "time_per_iteration": 4.232523441314697
    },
    {
      "auxiliary_loss_clip": 0.01181042,
      "auxiliary_loss_mlp": 0.01026522,
      "balance_loss_clip": 1.05285883,
      "balance_loss_mlp": 1.01879168,
      "epoch": 0.5212529309204593,
      "flos": 15628898787840.0,
      "grad_norm": 2.2223431422326647,
      "language_loss": 0.83000469,
      "learning_rate": 1.959496243543286e-06,
      "loss": 0.85208035,
      "num_input_tokens_seen": 93613135,
      "step": 4335,
      "time_per_iteration": 2.6622655391693115
    },
    {
      "auxiliary_loss_clip": 0.01189408,
      "auxiliary_loss_mlp": 0.01034351,
      "balance_loss_clip": 1.05833161,
      "balance_loss_mlp": 1.02529454,
      "epoch": 0.5213731738110984,
      "flos": 26242655829120.0,
      "grad_norm": 2.6396327971841513,
      "language_loss": 0.79154456,
      "learning_rate": 1.9587174347474057e-06,
      "loss": 0.81378222,
      "num_input_tokens_seen": 93629645,
      "step": 4336,
      "time_per_iteration": 2.7289295196533203
    },
    {
      "auxiliary_loss_clip": 0.01182935,
      "auxiliary_loss_mlp": 0.01032655,
      "balance_loss_clip": 1.05179882,
      "balance_loss_mlp": 1.02428079,
      "epoch": 0.5214934167017375,
      "flos": 19418407637760.0,
      "grad_norm": 2.899311731852909,
      "language_loss": 0.82432657,
      "learning_rate": 1.9579386322140574e-06,
      "loss": 0.84648246,
      "num_input_tokens_seen": 93645325,
      "step": 4337,
      "time_per_iteration": 3.5136725902557373
    },
    {
      "auxiliary_loss_clip": 0.01186096,
      "auxiliary_loss_mlp": 0.01027727,
      "balance_loss_clip": 1.05376792,
      "balance_loss_mlp": 1.01836348,
      "epoch": 0.5216136595923766,
      "flos": 30955788023040.0,
      "grad_norm": 2.500985808616258,
      "language_loss": 0.80640078,
      "learning_rate": 1.9571598360613854e-06,
      "loss": 0.82853901,
      "num_input_tokens_seen": 93668200,
      "step": 4338,
      "time_per_iteration": 2.7073864936828613
    },
    {
      "auxiliary_loss_clip": 0.0118347,
      "auxiliary_loss_mlp": 0.01033456,
      "balance_loss_clip": 1.05380392,
      "balance_loss_mlp": 1.02496219,
      "epoch": 0.5217339024830157,
      "flos": 21945047143680.0,
      "grad_norm": 2.400049507138396,
      "language_loss": 0.6974324,
      "learning_rate": 1.956381046407532e-06,
      "loss": 0.71960163,
      "num_input_tokens_seen": 93688495,
      "step": 4339,
      "time_per_iteration": 2.6117782592773438
    },
    {
      "auxiliary_loss_clip": 0.01179727,
      "auxiliary_loss_mlp": 0.01028667,
      "balance_loss_clip": 1.05201554,
      "balance_loss_mlp": 1.01994646,
      "epoch": 0.5218541453736548,
      "flos": 20923209037440.0,
      "grad_norm": 1.9362036320723182,
      "language_loss": 0.86140656,
      "learning_rate": 1.9556022633706394e-06,
      "loss": 0.88349044,
      "num_input_tokens_seen": 93707285,
      "step": 4340,
      "time_per_iteration": 2.62048077583313
    },
    {
      "auxiliary_loss_clip": 0.01183924,
      "auxiliary_loss_mlp": 0.01025145,
      "balance_loss_clip": 1.05367804,
      "balance_loss_mlp": 1.01591825,
      "epoch": 0.5219743882642939,
      "flos": 23951663498880.0,
      "grad_norm": 5.8243751674215245,
      "language_loss": 0.7985965,
      "learning_rate": 1.954823487068848e-06,
      "loss": 0.82068717,
      "num_input_tokens_seen": 93727495,
      "step": 4341,
      "time_per_iteration": 3.552344799041748
    },
    {
      "auxiliary_loss_clip": 0.01182582,
      "auxiliary_loss_mlp": 0.01031318,
      "balance_loss_clip": 1.05431807,
      "balance_loss_mlp": 1.02291429,
      "epoch": 0.5220946311549329,
      "flos": 28799280213120.0,
      "grad_norm": 2.459359106028317,
      "language_loss": 0.81287694,
      "learning_rate": 1.9540447176202976e-06,
      "loss": 0.83501595,
      "num_input_tokens_seen": 93748740,
      "step": 4342,
      "time_per_iteration": 3.5418953895568848
    },
    {
      "auxiliary_loss_clip": 0.01082276,
      "auxiliary_loss_mlp": 0.01001443,
      "balance_loss_clip": 1.01659596,
      "balance_loss_mlp": 1.00028062,
      "epoch": 0.5222148740455721,
      "flos": 67189369017600.0,
      "grad_norm": 0.8652588983991361,
      "language_loss": 0.60659593,
      "learning_rate": 1.9532659551431272e-06,
      "loss": 0.62743312,
      "num_input_tokens_seen": 93815770,
      "step": 4343,
      "time_per_iteration": 3.3572030067443848
    },
    {
      "auxiliary_loss_clip": 0.01182381,
      "auxiliary_loss_mlp": 0.01024067,
      "balance_loss_clip": 1.05322313,
      "balance_loss_mlp": 1.01616979,
      "epoch": 0.5223351169362112,
      "flos": 61856164339200.0,
      "grad_norm": 6.974280738094692,
      "language_loss": 0.67609215,
      "learning_rate": 1.9524871997554744e-06,
      "loss": 0.69815665,
      "num_input_tokens_seen": 93843530,
      "step": 4344,
      "time_per_iteration": 2.9296069145202637
    },
    {
      "auxiliary_loss_clip": 0.01183251,
      "auxiliary_loss_mlp": 0.01027519,
      "balance_loss_clip": 1.0531249,
      "balance_loss_mlp": 1.01913857,
      "epoch": 0.5224553598268502,
      "flos": 14647388676480.0,
      "grad_norm": 2.291890878006725,
      "language_loss": 0.81048357,
      "learning_rate": 1.951708451575475e-06,
      "loss": 0.8325913,
      "num_input_tokens_seen": 93860595,
      "step": 4345,
      "time_per_iteration": 2.6022632122039795
    },
    {
      "auxiliary_loss_clip": 0.01180092,
      "auxiliary_loss_mlp": 0.01023453,
      "balance_loss_clip": 1.05172706,
      "balance_loss_mlp": 1.01515007,
      "epoch": 0.5225756027174894,
      "flos": 14826043946880.0,
      "grad_norm": 2.36969016159589,
      "language_loss": 0.82333887,
      "learning_rate": 1.9509297107212657e-06,
      "loss": 0.84537429,
      "num_input_tokens_seen": 93877365,
      "step": 4346,
      "time_per_iteration": 2.6803510189056396
    },
    {
      "auxiliary_loss_clip": 0.0117996,
      "auxiliary_loss_mlp": 0.0102988,
      "balance_loss_clip": 1.05105948,
      "balance_loss_mlp": 1.02132118,
      "epoch": 0.5226958456081284,
      "flos": 23512009029120.0,
      "grad_norm": 1.7931822733843215,
      "language_loss": 0.79516578,
      "learning_rate": 1.95015097731098e-06,
      "loss": 0.8172642,
      "num_input_tokens_seen": 93896855,
      "step": 4347,
      "time_per_iteration": 2.5945310592651367
    },
    {
      "auxiliary_loss_clip": 0.01181934,
      "auxiliary_loss_mlp": 0.01028748,
      "balance_loss_clip": 1.05207968,
      "balance_loss_mlp": 1.01972389,
      "epoch": 0.5228160884987675,
      "flos": 19062928690560.0,
      "grad_norm": 3.6179826314894417,
      "language_loss": 0.8199451,
      "learning_rate": 1.949372251462751e-06,
      "loss": 0.84205192,
      "num_input_tokens_seen": 93914270,
      "step": 4348,
      "time_per_iteration": 2.591914415359497
    },
    {
      "auxiliary_loss_clip": 0.01184272,
      "auxiliary_loss_mlp": 0.0102297,
      "balance_loss_clip": 1.05532026,
      "balance_loss_mlp": 1.01498318,
      "epoch": 0.5229363313894067,
      "flos": 21063224252160.0,
      "grad_norm": 2.2296632598182393,
      "language_loss": 0.82407403,
      "learning_rate": 1.9485935332947124e-06,
      "loss": 0.84614646,
      "num_input_tokens_seen": 93932180,
      "step": 4349,
      "time_per_iteration": 2.649033784866333
    },
    {
      "auxiliary_loss_clip": 0.01177289,
      "auxiliary_loss_mlp": 0.01023754,
      "balance_loss_clip": 1.0500927,
      "balance_loss_mlp": 1.01523066,
      "epoch": 0.5230565742800457,
      "flos": 14830389492480.0,
      "grad_norm": 2.9751391701734002,
      "language_loss": 0.83296424,
      "learning_rate": 1.947814822924993e-06,
      "loss": 0.85497463,
      "num_input_tokens_seen": 93949690,
      "step": 4350,
      "time_per_iteration": 2.5655369758605957
    },
    {
      "auxiliary_loss_clip": 0.01181097,
      "auxiliary_loss_mlp": 0.01021536,
      "balance_loss_clip": 1.05315185,
      "balance_loss_mlp": 1.01336408,
      "epoch": 0.5231768171706848,
      "flos": 25813021253760.0,
      "grad_norm": 1.994258543329889,
      "language_loss": 0.82776648,
      "learning_rate": 1.9470361204717236e-06,
      "loss": 0.84979284,
      "num_input_tokens_seen": 93968830,
      "step": 4351,
      "time_per_iteration": 2.687917709350586
    },
    {
      "auxiliary_loss_clip": 0.01182717,
      "auxiliary_loss_mlp": 0.01027428,
      "balance_loss_clip": 1.05284464,
      "balance_loss_mlp": 1.01898241,
      "epoch": 0.5232970600613239,
      "flos": 22743807834240.0,
      "grad_norm": 2.3348978132876956,
      "language_loss": 0.80589557,
      "learning_rate": 1.9462574260530326e-06,
      "loss": 0.82799703,
      "num_input_tokens_seen": 93989110,
      "step": 4352,
      "time_per_iteration": 2.6537365913391113
    },
    {
      "auxiliary_loss_clip": 0.01176961,
      "auxiliary_loss_mlp": 0.01026134,
      "balance_loss_clip": 1.04879951,
      "balance_loss_mlp": 1.01716959,
      "epoch": 0.523417302951963,
      "flos": 17310703432320.0,
      "grad_norm": 1.9259828581034892,
      "language_loss": 0.80899739,
      "learning_rate": 1.9454787397870472e-06,
      "loss": 0.83102834,
      "num_input_tokens_seen": 94006430,
      "step": 4353,
      "time_per_iteration": 2.5694117546081543
    },
    {
      "auxiliary_loss_clip": 0.01183531,
      "auxiliary_loss_mlp": 0.01024791,
      "balance_loss_clip": 1.05416203,
      "balance_loss_mlp": 1.01662552,
      "epoch": 0.523537545842602,
      "flos": 18551740285440.0,
      "grad_norm": 1.9897801813273326,
      "language_loss": 0.72163403,
      "learning_rate": 1.944700061791894e-06,
      "loss": 0.74371725,
      "num_input_tokens_seen": 94024825,
      "step": 4354,
      "time_per_iteration": 2.647785186767578
    },
    {
      "auxiliary_loss_clip": 0.01181808,
      "auxiliary_loss_mlp": 0.01027607,
      "balance_loss_clip": 1.05170321,
      "balance_loss_mlp": 1.01892269,
      "epoch": 0.5236577887332411,
      "flos": 19719267955200.0,
      "grad_norm": 2.1387173321201876,
      "language_loss": 0.64776403,
      "learning_rate": 1.943921392185698e-06,
      "loss": 0.66985816,
      "num_input_tokens_seen": 94043450,
      "step": 4355,
      "time_per_iteration": 2.5806753635406494
    },
    {
      "auxiliary_loss_clip": 0.01181154,
      "auxiliary_loss_mlp": 0.01026922,
      "balance_loss_clip": 1.0523541,
      "balance_loss_mlp": 1.01820803,
      "epoch": 0.5237780316238803,
      "flos": 23550218121600.0,
      "grad_norm": 2.4051313629272517,
      "language_loss": 0.77155644,
      "learning_rate": 1.9431427310865814e-06,
      "loss": 0.79363722,
      "num_input_tokens_seen": 94063055,
      "step": 4356,
      "time_per_iteration": 2.6750307083129883
    },
    {
      "auxiliary_loss_clip": 0.01176161,
      "auxiliary_loss_mlp": 0.0102572,
      "balance_loss_clip": 1.05049491,
      "balance_loss_mlp": 1.01825774,
      "epoch": 0.5238982745145193,
      "flos": 22491894775680.0,
      "grad_norm": 2.1636619200994165,
      "language_loss": 0.78628314,
      "learning_rate": 1.942364078612667e-06,
      "loss": 0.80830193,
      "num_input_tokens_seen": 94081785,
      "step": 4357,
      "time_per_iteration": 2.6504242420196533
    },
    {
      "auxiliary_loss_clip": 0.01180329,
      "auxiliary_loss_mlp": 0.01029113,
      "balance_loss_clip": 1.05058002,
      "balance_loss_mlp": 1.02035129,
      "epoch": 0.5240185174051584,
      "flos": 27088927234560.0,
      "grad_norm": 1.7993398204027884,
      "language_loss": 0.75390428,
      "learning_rate": 1.9415854348820765e-06,
      "loss": 0.77599871,
      "num_input_tokens_seen": 94101635,
      "step": 4358,
      "time_per_iteration": 2.7454347610473633
    },
    {
      "auxiliary_loss_clip": 0.01181348,
      "auxiliary_loss_mlp": 0.010286,
      "balance_loss_clip": 1.05035806,
      "balance_loss_mlp": 1.01953447,
      "epoch": 0.5241387602957975,
      "flos": 22674680110080.0,
      "grad_norm": 2.8092814706572162,
      "language_loss": 0.6835351,
      "learning_rate": 1.940806800012929e-06,
      "loss": 0.70563459,
      "num_input_tokens_seen": 94121705,
      "step": 4359,
      "time_per_iteration": 2.5844504833221436
    },
    {
      "auxiliary_loss_clip": 0.01184394,
      "auxiliary_loss_mlp": 0.01027491,
      "balance_loss_clip": 1.05426729,
      "balance_loss_mlp": 1.01796663,
      "epoch": 0.5242590031864366,
      "flos": 40553453134080.0,
      "grad_norm": 1.671047450672029,
      "language_loss": 0.63587326,
      "learning_rate": 1.9400281741233432e-06,
      "loss": 0.65799212,
      "num_input_tokens_seen": 94146595,
      "step": 4360,
      "time_per_iteration": 2.819772720336914
    },
    {
      "auxiliary_loss_clip": 0.01082549,
      "auxiliary_loss_mlp": 0.01001492,
      "balance_loss_clip": 1.01630247,
      "balance_loss_mlp": 1.00033593,
      "epoch": 0.5243792460770756,
      "flos": 66676313105280.0,
      "grad_norm": 0.6570839503600685,
      "language_loss": 0.52511382,
      "learning_rate": 1.939249557331435e-06,
      "loss": 0.54595429,
      "num_input_tokens_seen": 94212410,
      "step": 4361,
      "time_per_iteration": 4.156051158905029
    },
    {
      "auxiliary_loss_clip": 0.01183928,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.05452895,
      "balance_loss_mlp": 1.01812673,
      "epoch": 0.5244994889677148,
      "flos": 28183663992960.0,
      "grad_norm": 3.0430175918814917,
      "language_loss": 0.72823036,
      "learning_rate": 1.938470949755321e-06,
      "loss": 0.75034022,
      "num_input_tokens_seen": 94232290,
      "step": 4362,
      "time_per_iteration": 2.6409618854522705
    },
    {
      "auxiliary_loss_clip": 0.01081907,
      "auxiliary_loss_mlp": 0.01001764,
      "balance_loss_clip": 1.01587772,
      "balance_loss_mlp": 1.00067353,
      "epoch": 0.5246197318583539,
      "flos": 65950379239680.0,
      "grad_norm": 0.9066092712704488,
      "language_loss": 0.55622423,
      "learning_rate": 1.937692351513115e-06,
      "loss": 0.57706094,
      "num_input_tokens_seen": 94291285,
      "step": 4363,
      "time_per_iteration": 3.9609711170196533
    },
    {
      "auxiliary_loss_clip": 0.01180817,
      "auxiliary_loss_mlp": 0.01024052,
      "balance_loss_clip": 1.05154347,
      "balance_loss_mlp": 1.0157553,
      "epoch": 0.5247399747489929,
      "flos": 21033490769280.0,
      "grad_norm": 2.014306359099883,
      "language_loss": 0.80652153,
      "learning_rate": 1.9369137627229297e-06,
      "loss": 0.82857025,
      "num_input_tokens_seen": 94309685,
      "step": 4364,
      "time_per_iteration": 2.5793919563293457
    },
    {
      "auxiliary_loss_clip": 0.01180524,
      "auxiliary_loss_mlp": 0.01027996,
      "balance_loss_clip": 1.05163288,
      "balance_loss_mlp": 1.01968098,
      "epoch": 0.5248602176396321,
      "flos": 19025940660480.0,
      "grad_norm": 3.0371920444752774,
      "language_loss": 0.88634694,
      "learning_rate": 1.936135183502877e-06,
      "loss": 0.90843213,
      "num_input_tokens_seen": 94326985,
      "step": 4365,
      "time_per_iteration": 2.6024880409240723
    },
    {
      "auxiliary_loss_clip": 0.01182944,
      "auxiliary_loss_mlp": 0.01030454,
      "balance_loss_clip": 1.05318832,
      "balance_loss_mlp": 1.0215373,
      "epoch": 0.5249804605302711,
      "flos": 22200084685440.0,
      "grad_norm": 2.2315138250076423,
      "language_loss": 0.80756348,
      "learning_rate": 1.935356613971066e-06,
      "loss": 0.82969749,
      "num_input_tokens_seen": 94347645,
      "step": 4366,
      "time_per_iteration": 2.5777523517608643
    },
    {
      "auxiliary_loss_clip": 0.0118007,
      "auxiliary_loss_mlp": 0.0102823,
      "balance_loss_clip": 1.05159414,
      "balance_loss_mlp": 1.01971841,
      "epoch": 0.5251007034209102,
      "flos": 23805686626560.0,
      "grad_norm": 1.8817263385922052,
      "language_loss": 0.76749134,
      "learning_rate": 1.9345780542456047e-06,
      "loss": 0.78957438,
      "num_input_tokens_seen": 94367020,
      "step": 4367,
      "time_per_iteration": 3.534548759460449
    },
    {
      "auxiliary_loss_clip": 0.01179616,
      "auxiliary_loss_mlp": 0.01032858,
      "balance_loss_clip": 1.05269217,
      "balance_loss_mlp": 1.02455544,
      "epoch": 0.5252209463115494,
      "flos": 23294605962240.0,
      "grad_norm": 2.848211892531178,
      "language_loss": 0.71378767,
      "learning_rate": 1.9337995044446007e-06,
      "loss": 0.73591244,
      "num_input_tokens_seen": 94385860,
      "step": 4368,
      "time_per_iteration": 3.566842555999756
    },
    {
      "auxiliary_loss_clip": 0.01181206,
      "auxiliary_loss_mlp": 0.0102848,
      "balance_loss_clip": 1.05231631,
      "balance_loss_mlp": 1.01940227,
      "epoch": 0.5253411892021884,
      "flos": 19828687760640.0,
      "grad_norm": 2.4685794297290222,
      "language_loss": 0.79747069,
      "learning_rate": 1.9330209646861596e-06,
      "loss": 0.81956756,
      "num_input_tokens_seen": 94405010,
      "step": 4369,
      "time_per_iteration": 2.6305224895477295
    },
    {
      "auxiliary_loss_clip": 0.01179094,
      "auxiliary_loss_mlp": 0.01026636,
      "balance_loss_clip": 1.04990852,
      "balance_loss_mlp": 1.01879179,
      "epoch": 0.5254614320928275,
      "flos": 24133730561280.0,
      "grad_norm": 1.7984504153003869,
      "language_loss": 0.77550375,
      "learning_rate": 1.9322424350883843e-06,
      "loss": 0.79756105,
      "num_input_tokens_seen": 94426845,
      "step": 4370,
      "time_per_iteration": 2.6395153999328613
    },
    {
      "auxiliary_loss_clip": 0.01180295,
      "auxiliary_loss_mlp": 0.01028825,
      "balance_loss_clip": 1.05199921,
      "balance_loss_mlp": 1.02068901,
      "epoch": 0.5255816749834666,
      "flos": 24644954880000.0,
      "grad_norm": 1.6682754802282829,
      "language_loss": 0.78836894,
      "learning_rate": 1.931463915769379e-06,
      "loss": 0.81046021,
      "num_input_tokens_seen": 94446960,
      "step": 4371,
      "time_per_iteration": 2.64393949508667
    },
    {
      "auxiliary_loss_clip": 0.01184329,
      "auxiliary_loss_mlp": 0.01030396,
      "balance_loss_clip": 1.05399454,
      "balance_loss_mlp": 1.02119911,
      "epoch": 0.5257019178741057,
      "flos": 14136595320960.0,
      "grad_norm": 3.636159980044421,
      "language_loss": 0.73656631,
      "learning_rate": 1.930685406847242e-06,
      "loss": 0.75871348,
      "num_input_tokens_seen": 94461535,
      "step": 4372,
      "time_per_iteration": 2.583716869354248
    },
    {
      "auxiliary_loss_clip": 0.0118171,
      "auxiliary_loss_mlp": 0.01023077,
      "balance_loss_clip": 1.05206132,
      "balance_loss_mlp": 1.01471162,
      "epoch": 0.5258221607647448,
      "flos": 23548961145600.0,
      "grad_norm": 4.655072569206606,
      "language_loss": 0.8134557,
      "learning_rate": 1.9299069084400734e-06,
      "loss": 0.83550358,
      "num_input_tokens_seen": 94482395,
      "step": 4373,
      "time_per_iteration": 2.6800906658172607
    },
    {
      "auxiliary_loss_clip": 0.01184096,
      "auxiliary_loss_mlp": 0.01024204,
      "balance_loss_clip": 1.05407834,
      "balance_loss_mlp": 1.01505446,
      "epoch": 0.5259424036553839,
      "flos": 24966103403520.0,
      "grad_norm": 1.9903432921694462,
      "language_loss": 0.69858617,
      "learning_rate": 1.9291284206659717e-06,
      "loss": 0.72066915,
      "num_input_tokens_seen": 94500580,
      "step": 4374,
      "time_per_iteration": 2.645270347595215
    },
    {
      "auxiliary_loss_clip": 0.01182937,
      "auxiliary_loss_mlp": 0.01025455,
      "balance_loss_clip": 1.05308354,
      "balance_loss_mlp": 1.01686049,
      "epoch": 0.526062646546023,
      "flos": 28763908295040.0,
      "grad_norm": 3.362317642907264,
      "language_loss": 0.71846926,
      "learning_rate": 1.928349943643032e-06,
      "loss": 0.74055314,
      "num_input_tokens_seen": 94519680,
      "step": 4375,
      "time_per_iteration": 2.663222551345825
    },
    {
      "auxiliary_loss_clip": 0.01180392,
      "auxiliary_loss_mlp": 0.01026488,
      "balance_loss_clip": 1.05319238,
      "balance_loss_mlp": 1.01845312,
      "epoch": 0.526182889436662,
      "flos": 22821375254400.0,
      "grad_norm": 1.7470242834622,
      "language_loss": 0.81598735,
      "learning_rate": 1.9275714774893493e-06,
      "loss": 0.83805621,
      "num_input_tokens_seen": 94539135,
      "step": 4376,
      "time_per_iteration": 2.650472640991211
    },
    {
      "auxiliary_loss_clip": 0.01181321,
      "auxiliary_loss_mlp": 0.01032738,
      "balance_loss_clip": 1.05282617,
      "balance_loss_mlp": 1.02369046,
      "epoch": 0.5263031323273012,
      "flos": 22929466256640.0,
      "grad_norm": 2.2853967098227286,
      "language_loss": 0.73153824,
      "learning_rate": 1.9267930223230154e-06,
      "loss": 0.7536788,
      "num_input_tokens_seen": 94557610,
      "step": 4377,
      "time_per_iteration": 2.6296939849853516
    },
    {
      "auxiliary_loss_clip": 0.01182193,
      "auxiliary_loss_mlp": 0.01026493,
      "balance_loss_clip": 1.05370355,
      "balance_loss_mlp": 1.01861358,
      "epoch": 0.5264233752179402,
      "flos": 17748634049280.0,
      "grad_norm": 2.8046930055764325,
      "language_loss": 0.78180754,
      "learning_rate": 1.9260145782621224e-06,
      "loss": 0.80389434,
      "num_input_tokens_seen": 94575390,
      "step": 4378,
      "time_per_iteration": 2.5954971313476562
    },
    {
      "auxiliary_loss_clip": 0.01182961,
      "auxiliary_loss_mlp": 0.01026081,
      "balance_loss_clip": 1.05375028,
      "balance_loss_mlp": 1.01807022,
      "epoch": 0.5265436181085793,
      "flos": 24421626069120.0,
      "grad_norm": 2.027312069024158,
      "language_loss": 0.8824628,
      "learning_rate": 1.925236145424758e-06,
      "loss": 0.90455317,
      "num_input_tokens_seen": 94594210,
      "step": 4379,
      "time_per_iteration": 2.6908352375030518
    },
    {
      "auxiliary_loss_clip": 0.01077469,
      "auxiliary_loss_mlp": 0.01000296,
      "balance_loss_clip": 1.01434922,
      "balance_loss_mlp": 0.99927634,
      "epoch": 0.5266638609992185,
      "flos": 69207298156800.0,
      "grad_norm": 0.6961325476096766,
      "language_loss": 0.57578969,
      "learning_rate": 1.924457723929012e-06,
      "loss": 0.59656727,
      "num_input_tokens_seen": 94665020,
      "step": 4380,
      "time_per_iteration": 3.3286068439483643
    },
    {
      "auxiliary_loss_clip": 0.01182174,
      "auxiliary_loss_mlp": 0.01024597,
      "balance_loss_clip": 1.05248106,
      "balance_loss_mlp": 1.01618052,
      "epoch": 0.5267841038898575,
      "flos": 20738699850240.0,
      "grad_norm": 1.7873479670464023,
      "language_loss": 0.82712078,
      "learning_rate": 1.9236793138929685e-06,
      "loss": 0.84918851,
      "num_input_tokens_seen": 94684290,
      "step": 4381,
      "time_per_iteration": 2.6495320796966553
    },
    {
      "auxiliary_loss_clip": 0.01184592,
      "auxiliary_loss_mlp": 0.01029919,
      "balance_loss_clip": 1.05418527,
      "balance_loss_mlp": 1.02072811,
      "epoch": 0.5269043467804966,
      "flos": 17234392988160.0,
      "grad_norm": 2.0522028527800975,
      "language_loss": 0.81372309,
      "learning_rate": 1.9229009154347133e-06,
      "loss": 0.83586824,
      "num_input_tokens_seen": 94701880,
      "step": 4382,
      "time_per_iteration": 2.5469186305999756
    },
    {
      "auxiliary_loss_clip": 0.0118104,
      "auxiliary_loss_mlp": 0.010254,
      "balance_loss_clip": 1.0548898,
      "balance_loss_mlp": 1.01737165,
      "epoch": 0.5270245896711357,
      "flos": 18223157646720.0,
      "grad_norm": 2.159189452353127,
      "language_loss": 0.80898774,
      "learning_rate": 1.922122528672327e-06,
      "loss": 0.83105218,
      "num_input_tokens_seen": 94720545,
      "step": 4383,
      "time_per_iteration": 2.637815475463867
    },
    {
      "auxiliary_loss_clip": 0.01179215,
      "auxiliary_loss_mlp": 0.0102602,
      "balance_loss_clip": 1.05153537,
      "balance_loss_mlp": 1.01814067,
      "epoch": 0.5271448325617748,
      "flos": 21287558643840.0,
      "grad_norm": 2.8246456177322816,
      "language_loss": 0.78366536,
      "learning_rate": 1.9213441537238914e-06,
      "loss": 0.80571777,
      "num_input_tokens_seen": 94737420,
      "step": 4384,
      "time_per_iteration": 2.6376001834869385
    },
    {
      "auxiliary_loss_clip": 0.01076891,
      "auxiliary_loss_mlp": 0.01003793,
      "balance_loss_clip": 1.01386654,
      "balance_loss_mlp": 1.00270772,
      "epoch": 0.5272650754524139,
      "flos": 65495497403520.0,
      "grad_norm": 0.8259177129136881,
      "language_loss": 0.57322407,
      "learning_rate": 1.920565790707485e-06,
      "loss": 0.59403086,
      "num_input_tokens_seen": 94802810,
      "step": 4385,
      "time_per_iteration": 3.3685643672943115
    },
    {
      "auxiliary_loss_clip": 0.01183692,
      "auxiliary_loss_mlp": 0.01026825,
      "balance_loss_clip": 1.05316484,
      "balance_loss_mlp": 1.01828408,
      "epoch": 0.527385318343053,
      "flos": 19676426008320.0,
      "grad_norm": 2.5127807748915303,
      "language_loss": 0.65881538,
      "learning_rate": 1.9197874397411853e-06,
      "loss": 0.68092054,
      "num_input_tokens_seen": 94819440,
      "step": 4386,
      "time_per_iteration": 2.642441511154175
    },
    {
      "auxiliary_loss_clip": 0.01180674,
      "auxiliary_loss_mlp": 0.01034963,
      "balance_loss_clip": 1.05164194,
      "balance_loss_mlp": 1.02614117,
      "epoch": 0.5275055612336921,
      "flos": 12712018947840.0,
      "grad_norm": 3.6840152423650503,
      "language_loss": 0.66865045,
      "learning_rate": 1.919009100943067e-06,
      "loss": 0.69080675,
      "num_input_tokens_seen": 94835130,
      "step": 4387,
      "time_per_iteration": 3.444821357727051
    },
    {
      "auxiliary_loss_clip": 0.01184993,
      "auxiliary_loss_mlp": 0.01025289,
      "balance_loss_clip": 1.05269217,
      "balance_loss_mlp": 1.01647925,
      "epoch": 0.5276258041243311,
      "flos": 17749029098880.0,
      "grad_norm": 2.417017351132885,
      "language_loss": 0.65628576,
      "learning_rate": 1.9182307744312043e-06,
      "loss": 0.6783886,
      "num_input_tokens_seen": 94852235,
      "step": 4388,
      "time_per_iteration": 2.62111496925354
    },
    {
      "auxiliary_loss_clip": 0.01180255,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.05105317,
      "balance_loss_mlp": 1.0174365,
      "epoch": 0.5277460470149702,
      "flos": 22710447077760.0,
      "grad_norm": 1.8785222168031415,
      "language_loss": 0.76382822,
      "learning_rate": 1.9174524603236676e-06,
      "loss": 0.78589451,
      "num_input_tokens_seen": 94871185,
      "step": 4389,
      "time_per_iteration": 3.499974250793457
    },
    {
      "auxiliary_loss_clip": 0.01183887,
      "auxiliary_loss_mlp": 0.01026271,
      "balance_loss_clip": 1.05245638,
      "balance_loss_mlp": 1.01692498,
      "epoch": 0.5278662899056094,
      "flos": 19902699734400.0,
      "grad_norm": 2.877039080344242,
      "language_loss": 0.76415867,
      "learning_rate": 1.916674158738527e-06,
      "loss": 0.78626025,
      "num_input_tokens_seen": 94890090,
      "step": 4390,
      "time_per_iteration": 2.5800302028656006
    },
    {
      "auxiliary_loss_clip": 0.01182779,
      "auxiliary_loss_mlp": 0.01024085,
      "balance_loss_clip": 1.05150819,
      "balance_loss_mlp": 1.01515031,
      "epoch": 0.5279865327962484,
      "flos": 18005215875840.0,
      "grad_norm": 2.517781415734341,
      "language_loss": 0.60088998,
      "learning_rate": 1.9158958697938506e-06,
      "loss": 0.62295866,
      "num_input_tokens_seen": 94908470,
      "step": 4391,
      "time_per_iteration": 2.6358251571655273
    },
    {
      "auxiliary_loss_clip": 0.01181511,
      "auxiliary_loss_mlp": 0.0102979,
      "balance_loss_clip": 1.05356669,
      "balance_loss_mlp": 1.02155256,
      "epoch": 0.5281067756868875,
      "flos": 15924443892480.0,
      "grad_norm": 3.329200506135843,
      "language_loss": 0.86087239,
      "learning_rate": 1.9151175936077032e-06,
      "loss": 0.88298541,
      "num_input_tokens_seen": 94923440,
      "step": 4392,
      "time_per_iteration": 2.6262307167053223
    },
    {
      "auxiliary_loss_clip": 0.01179727,
      "auxiliary_loss_mlp": 0.01028159,
      "balance_loss_clip": 1.05161381,
      "balance_loss_mlp": 1.01962316,
      "epoch": 0.5282270185775266,
      "flos": 19426488197760.0,
      "grad_norm": 2.0041986873072863,
      "language_loss": 0.79294837,
      "learning_rate": 1.9143393302981507e-06,
      "loss": 0.81502724,
      "num_input_tokens_seen": 94941125,
      "step": 4393,
      "time_per_iteration": 2.623157024383545
    },
    {
      "auxiliary_loss_clip": 0.01184346,
      "auxiliary_loss_mlp": 0.01024739,
      "balance_loss_clip": 1.05509138,
      "balance_loss_mlp": 1.01665032,
      "epoch": 0.5283472614681657,
      "flos": 16399613934720.0,
      "grad_norm": 2.0992648479434957,
      "language_loss": 0.83163822,
      "learning_rate": 1.913561079983252e-06,
      "loss": 0.85372907,
      "num_input_tokens_seen": 94959950,
      "step": 4394,
      "time_per_iteration": 4.504488468170166
    },
    {
      "auxiliary_loss_clip": 0.01185932,
      "auxiliary_loss_mlp": 0.01029981,
      "balance_loss_clip": 1.05482388,
      "balance_loss_mlp": 1.02087307,
      "epoch": 0.5284675043588047,
      "flos": 26760524163840.0,
      "grad_norm": 2.4046174423181124,
      "language_loss": 0.74940526,
      "learning_rate": 1.9127828427810693e-06,
      "loss": 0.77156436,
      "num_input_tokens_seen": 94980515,
      "step": 4395,
      "time_per_iteration": 2.650973081588745
    },
    {
      "auxiliary_loss_clip": 0.01183091,
      "auxiliary_loss_mlp": 0.01033033,
      "balance_loss_clip": 1.05353475,
      "balance_loss_mlp": 1.02427077,
      "epoch": 0.5285877472494439,
      "flos": 19899898473600.0,
      "grad_norm": 2.482514798345595,
      "language_loss": 0.81115782,
      "learning_rate": 1.9120046188096607e-06,
      "loss": 0.83331907,
      "num_input_tokens_seen": 94998560,
      "step": 4396,
      "time_per_iteration": 2.632270336151123
    },
    {
      "auxiliary_loss_clip": 0.01183777,
      "auxiliary_loss_mlp": 0.01031623,
      "balance_loss_clip": 1.05452991,
      "balance_loss_mlp": 1.02298045,
      "epoch": 0.528707990140083,
      "flos": 20011257613440.0,
      "grad_norm": 1.9881900645461437,
      "language_loss": 0.74567163,
      "learning_rate": 1.9112264081870804e-06,
      "loss": 0.7678256,
      "num_input_tokens_seen": 95016950,
      "step": 4397,
      "time_per_iteration": 2.626359462738037
    },
    {
      "auxiliary_loss_clip": 0.01184495,
      "auxiliary_loss_mlp": 0.01031051,
      "balance_loss_clip": 1.05280852,
      "balance_loss_mlp": 1.02149677,
      "epoch": 0.528828233030722,
      "flos": 20667956014080.0,
      "grad_norm": 8.262484006474985,
      "language_loss": 0.75837928,
      "learning_rate": 1.9104482110313843e-06,
      "loss": 0.78053474,
      "num_input_tokens_seen": 95036540,
      "step": 4398,
      "time_per_iteration": 2.6527798175811768
    },
    {
      "auxiliary_loss_clip": 0.01179888,
      "auxiliary_loss_mlp": 0.01025971,
      "balance_loss_clip": 1.05220556,
      "balance_loss_mlp": 1.01798987,
      "epoch": 0.5289484759213612,
      "flos": 25192448956800.0,
      "grad_norm": 1.8631288965973911,
      "language_loss": 0.74568307,
      "learning_rate": 1.909670027460623e-06,
      "loss": 0.76774168,
      "num_input_tokens_seen": 95053840,
      "step": 4399,
      "time_per_iteration": 2.6521828174591064
    },
    {
      "auxiliary_loss_clip": 0.01182568,
      "auxiliary_loss_mlp": 0.01024355,
      "balance_loss_clip": 1.05399632,
      "balance_loss_mlp": 1.01601624,
      "epoch": 0.5290687188120002,
      "flos": 31139255715840.0,
      "grad_norm": 2.6368977189108747,
      "language_loss": 0.71842331,
      "learning_rate": 1.908891857592847e-06,
      "loss": 0.74049252,
      "num_input_tokens_seen": 95074910,
      "step": 4400,
      "time_per_iteration": 2.7552740573883057
    },
    {
      "auxiliary_loss_clip": 0.0118213,
      "auxiliary_loss_mlp": 0.01026327,
      "balance_loss_clip": 1.05265319,
      "balance_loss_mlp": 1.01827443,
      "epoch": 0.5291889617026393,
      "flos": 20119851406080.0,
      "grad_norm": 2.362261886011266,
      "language_loss": 0.90044725,
      "learning_rate": 1.9081137015461034e-06,
      "loss": 0.92253178,
      "num_input_tokens_seen": 95090985,
      "step": 4401,
      "time_per_iteration": 2.576916456222534
    },
    {
      "auxiliary_loss_clip": 0.01183051,
      "auxiliary_loss_mlp": 0.01025816,
      "balance_loss_clip": 1.05411458,
      "balance_loss_mlp": 1.01771021,
      "epoch": 0.5293092045932785,
      "flos": 19643747610240.0,
      "grad_norm": 2.511036536914939,
      "language_loss": 0.90787518,
      "learning_rate": 1.9073355594384383e-06,
      "loss": 0.92996389,
      "num_input_tokens_seen": 95109225,
      "step": 4402,
      "time_per_iteration": 2.6456480026245117
    },
    {
      "auxiliary_loss_clip": 0.01181912,
      "auxiliary_loss_mlp": 0.01024525,
      "balance_loss_clip": 1.05361378,
      "balance_loss_mlp": 1.01607323,
      "epoch": 0.5294294474839175,
      "flos": 24317736958080.0,
      "grad_norm": 2.4592834616745223,
      "language_loss": 0.80312335,
      "learning_rate": 1.906557431387895e-06,
      "loss": 0.82518774,
      "num_input_tokens_seen": 95128215,
      "step": 4403,
      "time_per_iteration": 2.641906261444092
    },
    {
      "auxiliary_loss_clip": 0.01184108,
      "auxiliary_loss_mlp": 0.01027325,
      "balance_loss_clip": 1.05576169,
      "balance_loss_mlp": 1.0183723,
      "epoch": 0.5295496903745566,
      "flos": 18875941464960.0,
      "grad_norm": 4.958257578053625,
      "language_loss": 0.7913512,
      "learning_rate": 1.905779317512516e-06,
      "loss": 0.81346548,
      "num_input_tokens_seen": 95145760,
      "step": 4404,
      "time_per_iteration": 2.652799606323242
    },
    {
      "auxiliary_loss_clip": 0.01178569,
      "auxiliary_loss_mlp": 0.01022498,
      "balance_loss_clip": 1.05137038,
      "balance_loss_mlp": 1.01433265,
      "epoch": 0.5296699332651957,
      "flos": 20923101296640.0,
      "grad_norm": 5.627677974667963,
      "language_loss": 0.80702615,
      "learning_rate": 1.9050012179303385e-06,
      "loss": 0.82903683,
      "num_input_tokens_seen": 95164270,
      "step": 4405,
      "time_per_iteration": 2.644681930541992
    },
    {
      "auxiliary_loss_clip": 0.01184499,
      "auxiliary_loss_mlp": 0.01027145,
      "balance_loss_clip": 1.05396581,
      "balance_loss_mlp": 1.01835346,
      "epoch": 0.5297901761558348,
      "flos": 22046745525120.0,
      "grad_norm": 2.2195532872015713,
      "language_loss": 0.68805957,
      "learning_rate": 1.904223132759401e-06,
      "loss": 0.71017599,
      "num_input_tokens_seen": 95182870,
      "step": 4406,
      "time_per_iteration": 2.6400258541107178
    },
    {
      "auxiliary_loss_clip": 0.01183189,
      "auxiliary_loss_mlp": 0.01025389,
      "balance_loss_clip": 1.05249548,
      "balance_loss_mlp": 1.01686525,
      "epoch": 0.5299104190464738,
      "flos": 21798495653760.0,
      "grad_norm": 3.919384899389195,
      "language_loss": 0.69241744,
      "learning_rate": 1.9034450621177383e-06,
      "loss": 0.71450317,
      "num_input_tokens_seen": 95201190,
      "step": 4407,
      "time_per_iteration": 2.626492500305176
    },
    {
      "auxiliary_loss_clip": 0.01183188,
      "auxiliary_loss_mlp": 0.01033494,
      "balance_loss_clip": 1.05424237,
      "balance_loss_mlp": 1.0250839,
      "epoch": 0.530030661937113,
      "flos": 14720790119040.0,
      "grad_norm": 2.8113013354055374,
      "language_loss": 0.70505446,
      "learning_rate": 1.9026670061233824e-06,
      "loss": 0.72722131,
      "num_input_tokens_seen": 95218625,
      "step": 4408,
      "time_per_iteration": 2.6116771697998047
    },
    {
      "auxiliary_loss_clip": 0.01181869,
      "auxiliary_loss_mlp": 0.01026641,
      "balance_loss_clip": 1.05334473,
      "balance_loss_mlp": 1.01852858,
      "epoch": 0.5301509048277521,
      "flos": 21251504367360.0,
      "grad_norm": 1.6698548181746111,
      "language_loss": 0.8013041,
      "learning_rate": 1.901888964894365e-06,
      "loss": 0.82338917,
      "num_input_tokens_seen": 95237665,
      "step": 4409,
      "time_per_iteration": 2.6392083168029785
    },
    {
      "auxiliary_loss_clip": 0.01182516,
      "auxiliary_loss_mlp": 0.01026132,
      "balance_loss_clip": 1.05113292,
      "balance_loss_mlp": 1.0177691,
      "epoch": 0.5302711477183911,
      "flos": 25957058791680.0,
      "grad_norm": 2.288925712191963,
      "language_loss": 0.67965174,
      "learning_rate": 1.9011109385487134e-06,
      "loss": 0.70173818,
      "num_input_tokens_seen": 95258915,
      "step": 4410,
      "time_per_iteration": 2.677870988845825
    },
    {
      "auxiliary_loss_clip": 0.01183821,
      "auxiliary_loss_mlp": 0.01026361,
      "balance_loss_clip": 1.05399132,
      "balance_loss_mlp": 1.01775384,
      "epoch": 0.5303913906090303,
      "flos": 22273126992000.0,
      "grad_norm": 2.524564366107364,
      "language_loss": 0.6612097,
      "learning_rate": 1.900332927204454e-06,
      "loss": 0.68331146,
      "num_input_tokens_seen": 95277365,
      "step": 4411,
      "time_per_iteration": 2.6346914768218994
    },
    {
      "auxiliary_loss_clip": 0.01182486,
      "auxiliary_loss_mlp": 0.01029378,
      "balance_loss_clip": 1.05318904,
      "balance_loss_mlp": 1.02071142,
      "epoch": 0.5305116334996693,
      "flos": 24936010784640.0,
      "grad_norm": 5.874235089240123,
      "language_loss": 0.76743245,
      "learning_rate": 1.8995549309796097e-06,
      "loss": 0.78955114,
      "num_input_tokens_seen": 95296670,
      "step": 4412,
      "time_per_iteration": 2.6413726806640625
    },
    {
      "auxiliary_loss_clip": 0.01182129,
      "auxiliary_loss_mlp": 0.01024225,
      "balance_loss_clip": 1.05298913,
      "balance_loss_mlp": 1.01599371,
      "epoch": 0.5306318763903084,
      "flos": 20189338266240.0,
      "grad_norm": 1.9072725715209833,
      "language_loss": 0.76599491,
      "learning_rate": 1.8987769499922028e-06,
      "loss": 0.78805852,
      "num_input_tokens_seen": 95315640,
      "step": 4413,
      "time_per_iteration": 3.5287487506866455
    },
    {
      "auxiliary_loss_clip": 0.01181941,
      "auxiliary_loss_mlp": 0.01027561,
      "balance_loss_clip": 1.05409563,
      "balance_loss_mlp": 1.01929402,
      "epoch": 0.5307521192809476,
      "flos": 20266366982400.0,
      "grad_norm": 2.338415857009616,
      "language_loss": 0.70921719,
      "learning_rate": 1.897998984360252e-06,
      "loss": 0.73131227,
      "num_input_tokens_seen": 95334610,
      "step": 4414,
      "time_per_iteration": 2.616565227508545
    },
    {
      "auxiliary_loss_clip": 0.01180055,
      "auxiliary_loss_mlp": 0.0103029,
      "balance_loss_clip": 1.05428588,
      "balance_loss_mlp": 1.02247,
      "epoch": 0.5308723621715866,
      "flos": 28844276976000.0,
      "grad_norm": 1.5497910606031298,
      "language_loss": 0.78269148,
      "learning_rate": 1.897221034201775e-06,
      "loss": 0.80479491,
      "num_input_tokens_seen": 95358350,
      "step": 4415,
      "time_per_iteration": 2.7547459602355957
    },
    {
      "auxiliary_loss_clip": 0.01176461,
      "auxiliary_loss_mlp": 0.01027461,
      "balance_loss_clip": 1.05102491,
      "balance_loss_mlp": 1.01987922,
      "epoch": 0.5309926050622257,
      "flos": 27457766040960.0,
      "grad_norm": 1.9744279122767368,
      "language_loss": 0.66540504,
      "learning_rate": 1.8964430996347842e-06,
      "loss": 0.68744421,
      "num_input_tokens_seen": 95379900,
      "step": 4416,
      "time_per_iteration": 3.69169545173645
    },
    {
      "auxiliary_loss_clip": 0.01183391,
      "auxiliary_loss_mlp": 0.01028781,
      "balance_loss_clip": 1.05398834,
      "balance_loss_mlp": 1.01987028,
      "epoch": 0.5311128479528648,
      "flos": 20514545026560.0,
      "grad_norm": 1.9406057362714861,
      "language_loss": 0.8240096,
      "learning_rate": 1.8956651807772931e-06,
      "loss": 0.84613132,
      "num_input_tokens_seen": 95397935,
      "step": 4417,
      "time_per_iteration": 2.6268019676208496
    },
    {
      "auxiliary_loss_clip": 0.01178792,
      "auxiliary_loss_mlp": 0.01023568,
      "balance_loss_clip": 1.05326939,
      "balance_loss_mlp": 1.01543784,
      "epoch": 0.5312330908435039,
      "flos": 21397660807680.0,
      "grad_norm": 1.7842788962318896,
      "language_loss": 0.83526361,
      "learning_rate": 1.8948872777473115e-06,
      "loss": 0.85728717,
      "num_input_tokens_seen": 95415890,
      "step": 4418,
      "time_per_iteration": 2.6835107803344727
    },
    {
      "auxiliary_loss_clip": 0.01179884,
      "auxiliary_loss_mlp": 0.01029216,
      "balance_loss_clip": 1.05294466,
      "balance_loss_mlp": 1.02091289,
      "epoch": 0.531353333734143,
      "flos": 24717350741760.0,
      "grad_norm": 1.9813400215698695,
      "language_loss": 0.63457179,
      "learning_rate": 1.8941093906628458e-06,
      "loss": 0.65666282,
      "num_input_tokens_seen": 95433675,
      "step": 4419,
      "time_per_iteration": 2.6420083045959473
    },
    {
      "auxiliary_loss_clip": 0.01181736,
      "auxiliary_loss_mlp": 0.01026929,
      "balance_loss_clip": 1.0538528,
      "balance_loss_mlp": 1.0192343,
      "epoch": 0.531473576624782,
      "flos": 30480689808000.0,
      "grad_norm": 2.0240548546531727,
      "language_loss": 0.70907307,
      "learning_rate": 1.893331519641902e-06,
      "loss": 0.73115969,
      "num_input_tokens_seen": 95455820,
      "step": 4420,
      "time_per_iteration": 3.6284492015838623
    },
    {
      "auxiliary_loss_clip": 0.01180224,
      "auxiliary_loss_mlp": 0.01031624,
      "balance_loss_clip": 1.05125451,
      "balance_loss_mlp": 1.0234462,
      "epoch": 0.5315938195154212,
      "flos": 23002975440000.0,
      "grad_norm": 2.8102481197796956,
      "language_loss": 0.73864627,
      "learning_rate": 1.8925536648024815e-06,
      "loss": 0.76076484,
      "num_input_tokens_seen": 95473240,
      "step": 4421,
      "time_per_iteration": 3.572871208190918
    },
    {
      "auxiliary_loss_clip": 0.01180995,
      "auxiliary_loss_mlp": 0.01027743,
      "balance_loss_clip": 1.05306816,
      "balance_loss_mlp": 1.019786,
      "epoch": 0.5317140624060602,
      "flos": 22748584343040.0,
      "grad_norm": 2.5641453290892056,
      "language_loss": 0.75764233,
      "learning_rate": 1.8917758262625849e-06,
      "loss": 0.77972972,
      "num_input_tokens_seen": 95493480,
      "step": 4422,
      "time_per_iteration": 2.631061553955078
    },
    {
      "auxiliary_loss_clip": 0.01176867,
      "auxiliary_loss_mlp": 0.0103167,
      "balance_loss_clip": 1.05191731,
      "balance_loss_mlp": 1.02410579,
      "epoch": 0.5318343052966993,
      "flos": 22821087945600.0,
      "grad_norm": 1.8232334361781914,
      "language_loss": 0.80820203,
      "learning_rate": 1.8909980041402089e-06,
      "loss": 0.83028746,
      "num_input_tokens_seen": 95512075,
      "step": 4423,
      "time_per_iteration": 2.7020890712738037
    },
    {
      "auxiliary_loss_clip": 0.01180463,
      "auxiliary_loss_mlp": 0.01025983,
      "balance_loss_clip": 1.05205917,
      "balance_loss_mlp": 1.01727509,
      "epoch": 0.5319545481873384,
      "flos": 13626089274240.0,
      "grad_norm": 3.8679714490528383,
      "language_loss": 0.65893567,
      "learning_rate": 1.8902201985533494e-06,
      "loss": 0.68100017,
      "num_input_tokens_seen": 95529340,
      "step": 4424,
      "time_per_iteration": 2.585879325866699
    },
    {
      "auxiliary_loss_clip": 0.01181908,
      "auxiliary_loss_mlp": 0.0102708,
      "balance_loss_clip": 1.05382895,
      "balance_loss_mlp": 1.01893806,
      "epoch": 0.5320747910779775,
      "flos": 22162522037760.0,
      "grad_norm": 2.4870295104816336,
      "language_loss": 0.74864721,
      "learning_rate": 1.8894424096199983e-06,
      "loss": 0.77073705,
      "num_input_tokens_seen": 95548545,
      "step": 4425,
      "time_per_iteration": 2.7149431705474854
    },
    {
      "auxiliary_loss_clip": 0.01183561,
      "auxiliary_loss_mlp": 0.01028779,
      "balance_loss_clip": 1.0550487,
      "balance_loss_mlp": 1.01996911,
      "epoch": 0.5321950339686166,
      "flos": 18588081870720.0,
      "grad_norm": 1.9969909427776842,
      "language_loss": 0.85482025,
      "learning_rate": 1.8886646374581463e-06,
      "loss": 0.87694365,
      "num_input_tokens_seen": 95567770,
      "step": 4426,
      "time_per_iteration": 2.5969796180725098
    },
    {
      "auxiliary_loss_clip": 0.01181055,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.05239427,
      "balance_loss_mlp": 1.01829648,
      "epoch": 0.5323152768592557,
      "flos": 22856818999680.0,
      "grad_norm": 1.7138819403740795,
      "language_loss": 0.70992291,
      "learning_rate": 1.8878868821857795e-06,
      "loss": 0.73200405,
      "num_input_tokens_seen": 95587420,
      "step": 4427,
      "time_per_iteration": 2.688051462173462
    },
    {
      "auxiliary_loss_clip": 0.01183108,
      "auxiliary_loss_mlp": 0.01027481,
      "balance_loss_clip": 1.05207896,
      "balance_loss_mlp": 1.01879656,
      "epoch": 0.5324355197498948,
      "flos": 33948690998400.0,
      "grad_norm": 2.5560020111454227,
      "language_loss": 0.7525965,
      "learning_rate": 1.8871091439208838e-06,
      "loss": 0.77470243,
      "num_input_tokens_seen": 95609030,
      "step": 4428,
      "time_per_iteration": 2.717373847961426
    },
    {
      "auxiliary_loss_clip": 0.01184383,
      "auxiliary_loss_mlp": 0.01030888,
      "balance_loss_clip": 1.05429363,
      "balance_loss_mlp": 1.02174509,
      "epoch": 0.5325557626405338,
      "flos": 23256720092160.0,
      "grad_norm": 2.5859612651476964,
      "language_loss": 0.77237368,
      "learning_rate": 1.8863314227814414e-06,
      "loss": 0.79452634,
      "num_input_tokens_seen": 95627340,
      "step": 4429,
      "time_per_iteration": 2.641789674758911
    },
    {
      "auxiliary_loss_clip": 0.0118446,
      "auxiliary_loss_mlp": 0.01029149,
      "balance_loss_clip": 1.05422616,
      "balance_loss_mlp": 1.02074444,
      "epoch": 0.532676005531173,
      "flos": 26718687797760.0,
      "grad_norm": 2.814320910080408,
      "language_loss": 0.48872769,
      "learning_rate": 1.8855537188854313e-06,
      "loss": 0.51086378,
      "num_input_tokens_seen": 95646315,
      "step": 4430,
      "time_per_iteration": 2.6316394805908203
    },
    {
      "auxiliary_loss_clip": 0.01181932,
      "auxiliary_loss_mlp": 0.01029598,
      "balance_loss_clip": 1.05226922,
      "balance_loss_mlp": 1.02122927,
      "epoch": 0.5327962484218121,
      "flos": 17894610921600.0,
      "grad_norm": 2.282349409077055,
      "language_loss": 0.78418827,
      "learning_rate": 1.8847760323508315e-06,
      "loss": 0.8063035,
      "num_input_tokens_seen": 95665220,
      "step": 4431,
      "time_per_iteration": 2.6965839862823486
    },
    {
      "auxiliary_loss_clip": 0.01178116,
      "auxiliary_loss_mlp": 0.01027141,
      "balance_loss_clip": 1.05291033,
      "balance_loss_mlp": 1.01920199,
      "epoch": 0.5329164913124511,
      "flos": 17925385898880.0,
      "grad_norm": 2.030960759715977,
      "language_loss": 0.75633436,
      "learning_rate": 1.883998363295616e-06,
      "loss": 0.77838689,
      "num_input_tokens_seen": 95682700,
      "step": 4432,
      "time_per_iteration": 2.6258645057678223
    },
    {
      "auxiliary_loss_clip": 0.01080907,
      "auxiliary_loss_mlp": 0.01001133,
      "balance_loss_clip": 1.01710212,
      "balance_loss_mlp": 0.99989361,
      "epoch": 0.5330367342030903,
      "flos": 57254178781440.0,
      "grad_norm": 0.8823682010078814,
      "language_loss": 0.62560254,
      "learning_rate": 1.8832207118377565e-06,
      "loss": 0.64642292,
      "num_input_tokens_seen": 95738070,
      "step": 4433,
      "time_per_iteration": 3.0995590686798096
    },
    {
      "auxiliary_loss_clip": 0.01178353,
      "auxiliary_loss_mlp": 0.01028499,
      "balance_loss_clip": 1.05152869,
      "balance_loss_mlp": 1.02052987,
      "epoch": 0.5331569770937293,
      "flos": 17420518287360.0,
      "grad_norm": 3.4564263016936128,
      "language_loss": 0.6918422,
      "learning_rate": 1.882443078095222e-06,
      "loss": 0.7139107,
      "num_input_tokens_seen": 95756950,
      "step": 4434,
      "time_per_iteration": 2.559812068939209
    },
    {
      "auxiliary_loss_clip": 0.01080861,
      "auxiliary_loss_mlp": 0.01002147,
      "balance_loss_clip": 1.01726604,
      "balance_loss_mlp": 1.00090742,
      "epoch": 0.5332772199843684,
      "flos": 56750783627520.0,
      "grad_norm": 0.8673938868703408,
      "language_loss": 0.66799378,
      "learning_rate": 1.8816654621859794e-06,
      "loss": 0.68882382,
      "num_input_tokens_seen": 95816615,
      "step": 4435,
      "time_per_iteration": 3.015383005142212
    },
    {
      "auxiliary_loss_clip": 0.01180321,
      "auxiliary_loss_mlp": 0.01024487,
      "balance_loss_clip": 1.05279946,
      "balance_loss_mlp": 1.01652396,
      "epoch": 0.5333974628750076,
      "flos": 18697753071360.0,
      "grad_norm": 2.4852087115328105,
      "language_loss": 0.7275784,
      "learning_rate": 1.8808878642279915e-06,
      "loss": 0.74962646,
      "num_input_tokens_seen": 95832020,
      "step": 4436,
      "time_per_iteration": 2.576103687286377
    },
    {
      "auxiliary_loss_clip": 0.01185096,
      "auxiliary_loss_mlp": 0.01020108,
      "balance_loss_clip": 1.05403614,
      "balance_loss_mlp": 1.01192439,
      "epoch": 0.5335177057656466,
      "flos": 23805507058560.0,
      "grad_norm": 3.0391783215988624,
      "language_loss": 0.65005332,
      "learning_rate": 1.8801102843392209e-06,
      "loss": 0.67210531,
      "num_input_tokens_seen": 95851425,
      "step": 4437,
      "time_per_iteration": 2.5235207080841064
    },
    {
      "auxiliary_loss_clip": 0.01179893,
      "auxiliary_loss_mlp": 0.01021883,
      "balance_loss_clip": 1.05202544,
      "balance_loss_mlp": 1.01397324,
      "epoch": 0.5336379486562857,
      "flos": 25078683605760.0,
      "grad_norm": 5.13491278499298,
      "language_loss": 0.85522389,
      "learning_rate": 1.8793327226376238e-06,
      "loss": 0.87724161,
      "num_input_tokens_seen": 95870745,
      "step": 4438,
      "time_per_iteration": 2.539158344268799
    },
    {
      "auxiliary_loss_clip": 0.01182274,
      "auxiliary_loss_mlp": 0.01031826,
      "balance_loss_clip": 1.05337763,
      "balance_loss_mlp": 1.02305222,
      "epoch": 0.5337581915469248,
      "flos": 21396691140480.0,
      "grad_norm": 1.901277531103919,
      "language_loss": 0.79872233,
      "learning_rate": 1.8785551792411569e-06,
      "loss": 0.82086337,
      "num_input_tokens_seen": 95889755,
      "step": 4439,
      "time_per_iteration": 2.606600284576416
    },
    {
      "auxiliary_loss_clip": 0.01181459,
      "auxiliary_loss_mlp": 0.01021047,
      "balance_loss_clip": 1.05257642,
      "balance_loss_mlp": 1.01313126,
      "epoch": 0.5338784344375639,
      "flos": 14865905064960.0,
      "grad_norm": 2.2693038337288876,
      "language_loss": 0.8259511,
      "learning_rate": 1.8777776542677733e-06,
      "loss": 0.84797615,
      "num_input_tokens_seen": 95907805,
      "step": 4440,
      "time_per_iteration": 3.483142614364624
    },
    {
      "auxiliary_loss_clip": 0.01181733,
      "auxiliary_loss_mlp": 0.01031358,
      "balance_loss_clip": 1.05135918,
      "balance_loss_mlp": 1.0226264,
      "epoch": 0.5339986773282029,
      "flos": 20813501923200.0,
      "grad_norm": 1.9042154561584004,
      "language_loss": 0.7343086,
      "learning_rate": 1.8770001478354216e-06,
      "loss": 0.75643957,
      "num_input_tokens_seen": 95927480,
      "step": 4441,
      "time_per_iteration": 2.571894645690918
    },
    {
      "auxiliary_loss_clip": 0.01182371,
      "auxiliary_loss_mlp": 0.01030089,
      "balance_loss_clip": 1.05288482,
      "balance_loss_mlp": 1.02069557,
      "epoch": 0.5341189202188421,
      "flos": 17969089772160.0,
      "grad_norm": 2.8532372155248824,
      "language_loss": 0.83272803,
      "learning_rate": 1.8762226600620504e-06,
      "loss": 0.85485256,
      "num_input_tokens_seen": 95946095,
      "step": 4442,
      "time_per_iteration": 2.592940092086792
    },
    {
      "auxiliary_loss_clip": 0.01185386,
      "auxiliary_loss_mlp": 0.01030055,
      "balance_loss_clip": 1.05504107,
      "balance_loss_mlp": 1.02128744,
      "epoch": 0.5342391631094812,
      "flos": 11031866328960.0,
      "grad_norm": 2.4390667333548675,
      "language_loss": 0.5891667,
      "learning_rate": 1.8754451910656031e-06,
      "loss": 0.61132115,
      "num_input_tokens_seen": 95959995,
      "step": 4443,
      "time_per_iteration": 3.445103645324707
    },
    {
      "auxiliary_loss_clip": 0.01183848,
      "auxiliary_loss_mlp": 0.01030215,
      "balance_loss_clip": 1.05415154,
      "balance_loss_mlp": 1.02144718,
      "epoch": 0.5343594060001202,
      "flos": 15339135772800.0,
      "grad_norm": 2.032327568083211,
      "language_loss": 0.82585251,
      "learning_rate": 1.8746677409640212e-06,
      "loss": 0.84799314,
      "num_input_tokens_seen": 95977095,
      "step": 4444,
      "time_per_iteration": 2.534350633621216
    },
    {
      "auxiliary_loss_clip": 0.0118368,
      "auxiliary_loss_mlp": 0.01028747,
      "balance_loss_clip": 1.05408907,
      "balance_loss_mlp": 1.01987731,
      "epoch": 0.5344796488907594,
      "flos": 26900898514560.0,
      "grad_norm": 1.7655816098175154,
      "language_loss": 0.84493244,
      "learning_rate": 1.8738903098752432e-06,
      "loss": 0.86705667,
      "num_input_tokens_seen": 95996225,
      "step": 4445,
      "time_per_iteration": 2.6140551567077637
    },
    {
      "auxiliary_loss_clip": 0.01182852,
      "auxiliary_loss_mlp": 0.01027782,
      "balance_loss_clip": 1.05489635,
      "balance_loss_mlp": 1.02006602,
      "epoch": 0.5345998917813984,
      "flos": 25411216740480.0,
      "grad_norm": 2.4348109774893087,
      "language_loss": 0.7322827,
      "learning_rate": 1.8731128979172052e-06,
      "loss": 0.75438905,
      "num_input_tokens_seen": 96015425,
      "step": 4446,
      "time_per_iteration": 2.622483253479004
    },
    {
      "auxiliary_loss_clip": 0.01177863,
      "auxiliary_loss_mlp": 0.01022454,
      "balance_loss_clip": 1.04942036,
      "balance_loss_mlp": 1.01495564,
      "epoch": 0.5347201346720375,
      "flos": 32853379622400.0,
      "grad_norm": 2.5001513000804123,
      "language_loss": 0.67005312,
      "learning_rate": 1.8723355052078394e-06,
      "loss": 0.6920563,
      "num_input_tokens_seen": 96035460,
      "step": 4447,
      "time_per_iteration": 4.516971588134766
    },
    {
      "auxiliary_loss_clip": 0.01184913,
      "auxiliary_loss_mlp": 0.01030517,
      "balance_loss_clip": 1.05397773,
      "balance_loss_mlp": 1.02126074,
      "epoch": 0.5348403775626767,
      "flos": 17967940536960.0,
      "grad_norm": 2.175870656191087,
      "language_loss": 0.77207434,
      "learning_rate": 1.8715581318650765e-06,
      "loss": 0.79422867,
      "num_input_tokens_seen": 96054515,
      "step": 4448,
      "time_per_iteration": 2.7002668380737305
    },
    {
      "auxiliary_loss_clip": 0.01185981,
      "auxiliary_loss_mlp": 0.01026517,
      "balance_loss_clip": 1.05481625,
      "balance_loss_mlp": 1.01733756,
      "epoch": 0.5349606204533157,
      "flos": 17603339535360.0,
      "grad_norm": 2.4013342513912996,
      "language_loss": 0.82103121,
      "learning_rate": 1.8707807780068422e-06,
      "loss": 0.84315616,
      "num_input_tokens_seen": 96072330,
      "step": 4449,
      "time_per_iteration": 2.59263014793396
    },
    {
      "auxiliary_loss_clip": 0.01180826,
      "auxiliary_loss_mlp": 0.01023705,
      "balance_loss_clip": 1.05149674,
      "balance_loss_mlp": 1.01530647,
      "epoch": 0.5350808633439548,
      "flos": 29167831710720.0,
      "grad_norm": 2.723602312472371,
      "language_loss": 0.66452515,
      "learning_rate": 1.8700034437510611e-06,
      "loss": 0.68657053,
      "num_input_tokens_seen": 96092425,
      "step": 4450,
      "time_per_iteration": 2.648155689239502
    },
    {
      "auxiliary_loss_clip": 0.0117987,
      "auxiliary_loss_mlp": 0.01029459,
      "balance_loss_clip": 1.05234981,
      "balance_loss_mlp": 1.02079272,
      "epoch": 0.5352011062345938,
      "flos": 19499997381120.0,
      "grad_norm": 6.662137053933195,
      "language_loss": 0.81541306,
      "learning_rate": 1.8692261292156549e-06,
      "loss": 0.83750629,
      "num_input_tokens_seen": 96111660,
      "step": 4451,
      "time_per_iteration": 2.5731067657470703
    },
    {
      "auxiliary_loss_clip": 0.01178468,
      "auxiliary_loss_mlp": 0.01023851,
      "balance_loss_clip": 1.05265713,
      "balance_loss_mlp": 1.01579833,
      "epoch": 0.535321349125233,
      "flos": 23477642691840.0,
      "grad_norm": 1.9656610598023458,
      "language_loss": 0.80984986,
      "learning_rate": 1.8684488345185401e-06,
      "loss": 0.83187306,
      "num_input_tokens_seen": 96131835,
      "step": 4452,
      "time_per_iteration": 2.6855690479278564
    },
    {
      "auxiliary_loss_clip": 0.01181902,
      "auxiliary_loss_mlp": 0.01030252,
      "balance_loss_clip": 1.0534364,
      "balance_loss_mlp": 1.02219355,
      "epoch": 0.535441592015872,
      "flos": 20478059786880.0,
      "grad_norm": 2.274030109072816,
      "language_loss": 0.78536499,
      "learning_rate": 1.8676715597776332e-06,
      "loss": 0.80748653,
      "num_input_tokens_seen": 96150180,
      "step": 4453,
      "time_per_iteration": 2.5680387020111084
    },
    {
      "auxiliary_loss_clip": 0.01177864,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.05180001,
      "balance_loss_mlp": 1.01766777,
      "epoch": 0.5355618349065111,
      "flos": 19573147428480.0,
      "grad_norm": 2.1025852746087494,
      "language_loss": 0.76148701,
      "learning_rate": 1.8668943051108455e-06,
      "loss": 0.78352022,
      "num_input_tokens_seen": 96167485,
      "step": 4454,
      "time_per_iteration": 2.751899242401123
    },
    {
      "auxiliary_loss_clip": 0.01185619,
      "auxiliary_loss_mlp": 0.01022439,
      "balance_loss_clip": 1.05490088,
      "balance_loss_mlp": 1.01426697,
      "epoch": 0.5356820777971503,
      "flos": 24024633978240.0,
      "grad_norm": 1.7616520345494313,
      "language_loss": 0.76011455,
      "learning_rate": 1.8661170706360856e-06,
      "loss": 0.78219509,
      "num_input_tokens_seen": 96186650,
      "step": 4455,
      "time_per_iteration": 2.585953712463379
    },
    {
      "auxiliary_loss_clip": 0.01180003,
      "auxiliary_loss_mlp": 0.01028968,
      "balance_loss_clip": 1.05277181,
      "balance_loss_mlp": 1.02073073,
      "epoch": 0.5358023206877893,
      "flos": 20884676722560.0,
      "grad_norm": 1.8000311009793262,
      "language_loss": 0.81584311,
      "learning_rate": 1.8653398564712594e-06,
      "loss": 0.83793283,
      "num_input_tokens_seen": 96205595,
      "step": 4456,
      "time_per_iteration": 2.605487823486328
    },
    {
      "auxiliary_loss_clip": 0.01179347,
      "auxiliary_loss_mlp": 0.01024878,
      "balance_loss_clip": 1.05218828,
      "balance_loss_mlp": 1.01708734,
      "epoch": 0.5359225635784284,
      "flos": 22418996123520.0,
      "grad_norm": 2.0400923911252105,
      "language_loss": 0.82243216,
      "learning_rate": 1.8645626627342704e-06,
      "loss": 0.84447438,
      "num_input_tokens_seen": 96226360,
      "step": 4457,
      "time_per_iteration": 2.648430824279785
    },
    {
      "auxiliary_loss_clip": 0.01179348,
      "auxiliary_loss_mlp": 0.01029234,
      "balance_loss_clip": 1.05184317,
      "balance_loss_mlp": 1.02084732,
      "epoch": 0.5360428064690675,
      "flos": 24097784025600.0,
      "grad_norm": 2.4633117384373255,
      "language_loss": 0.80882275,
      "learning_rate": 1.8637854895430172e-06,
      "loss": 0.8309086,
      "num_input_tokens_seen": 96245625,
      "step": 4458,
      "time_per_iteration": 2.6166257858276367
    },
    {
      "auxiliary_loss_clip": 0.01180872,
      "auxiliary_loss_mlp": 0.01023931,
      "balance_loss_clip": 1.05115235,
      "balance_loss_mlp": 1.01573539,
      "epoch": 0.5361630493597066,
      "flos": 21434505183360.0,
      "grad_norm": 2.4376637090663134,
      "language_loss": 0.69468427,
      "learning_rate": 1.8630083370153978e-06,
      "loss": 0.71673226,
      "num_input_tokens_seen": 96265265,
      "step": 4459,
      "time_per_iteration": 2.6317074298858643
    },
    {
      "auxiliary_loss_clip": 0.01083616,
      "auxiliary_loss_mlp": 0.01005364,
      "balance_loss_clip": 1.01917112,
      "balance_loss_mlp": 1.00407696,
      "epoch": 0.5362832922503457,
      "flos": 68888696520960.0,
      "grad_norm": 0.7454377952449902,
      "language_loss": 0.55356395,
      "learning_rate": 1.8622312052693041e-06,
      "loss": 0.57445371,
      "num_input_tokens_seen": 96326445,
      "step": 4460,
      "time_per_iteration": 3.2720959186553955
    },
    {
      "auxiliary_loss_clip": 0.01177971,
      "auxiliary_loss_mlp": 0.01026983,
      "balance_loss_clip": 1.04893148,
      "balance_loss_mlp": 1.01914489,
      "epoch": 0.5364035351409848,
      "flos": 9793702563840.0,
      "grad_norm": 2.7776298495862344,
      "language_loss": 0.71534359,
      "learning_rate": 1.8614540944226267e-06,
      "loss": 0.73739314,
      "num_input_tokens_seen": 96343115,
      "step": 4461,
      "time_per_iteration": 2.584951162338257
    },
    {
      "auxiliary_loss_clip": 0.01178381,
      "auxiliary_loss_mlp": 0.01023971,
      "balance_loss_clip": 1.05320764,
      "balance_loss_mlp": 1.01628196,
      "epoch": 0.5365237780316239,
      "flos": 23290080848640.0,
      "grad_norm": 2.6941403800251007,
      "language_loss": 0.68456775,
      "learning_rate": 1.8606770045932537e-06,
      "loss": 0.70659125,
      "num_input_tokens_seen": 96362230,
      "step": 4462,
      "time_per_iteration": 2.6229605674743652
    },
    {
      "auxiliary_loss_clip": 0.011847,
      "auxiliary_loss_mlp": 0.01029351,
      "balance_loss_clip": 1.05431342,
      "balance_loss_mlp": 1.02065516,
      "epoch": 0.5366440209222629,
      "flos": 26578133879040.0,
      "grad_norm": 2.568437402467268,
      "language_loss": 0.81688327,
      "learning_rate": 1.859899935899068e-06,
      "loss": 0.83902383,
      "num_input_tokens_seen": 96382085,
      "step": 4463,
      "time_per_iteration": 2.6860415935516357
    },
    {
      "auxiliary_loss_clip": 0.01181394,
      "auxiliary_loss_mlp": 0.01025727,
      "balance_loss_clip": 1.05272055,
      "balance_loss_mlp": 1.01799083,
      "epoch": 0.5367642638129021,
      "flos": 19608052469760.0,
      "grad_norm": 1.8744225260550846,
      "language_loss": 0.79126269,
      "learning_rate": 1.8591228884579506e-06,
      "loss": 0.81333387,
      "num_input_tokens_seen": 96400580,
      "step": 4464,
      "time_per_iteration": 2.603572130203247
    },
    {
      "auxiliary_loss_clip": 0.01179695,
      "auxiliary_loss_mlp": 0.01025763,
      "balance_loss_clip": 1.05106735,
      "balance_loss_mlp": 1.01747155,
      "epoch": 0.5368845067035412,
      "flos": 23915214172800.0,
      "grad_norm": 2.8319694982978065,
      "language_loss": 0.81914473,
      "learning_rate": 1.8583458623877795e-06,
      "loss": 0.84119928,
      "num_input_tokens_seen": 96419680,
      "step": 4465,
      "time_per_iteration": 2.6604695320129395
    },
    {
      "auxiliary_loss_clip": 0.01180051,
      "auxiliary_loss_mlp": 0.01024178,
      "balance_loss_clip": 1.05250287,
      "balance_loss_mlp": 1.01584494,
      "epoch": 0.5370047495941802,
      "flos": 16873131951360.0,
      "grad_norm": 2.2064706211094567,
      "language_loss": 0.74302542,
      "learning_rate": 1.8575688578064281e-06,
      "loss": 0.76506776,
      "num_input_tokens_seen": 96437805,
      "step": 4466,
      "time_per_iteration": 2.547294855117798
    },
    {
      "auxiliary_loss_clip": 0.01182212,
      "auxiliary_loss_mlp": 0.01030258,
      "balance_loss_clip": 1.05479503,
      "balance_loss_mlp": 1.02221096,
      "epoch": 0.5371249924848194,
      "flos": 20740926493440.0,
      "grad_norm": 1.643969361024476,
      "language_loss": 0.76788735,
      "learning_rate": 1.8567918748317674e-06,
      "loss": 0.79001206,
      "num_input_tokens_seen": 96457155,
      "step": 4467,
      "time_per_iteration": 3.515181541442871
    },
    {
      "auxiliary_loss_clip": 0.01181734,
      "auxiliary_loss_mlp": 0.01030429,
      "balance_loss_clip": 1.05336905,
      "balance_loss_mlp": 1.02250779,
      "epoch": 0.5372452353754584,
      "flos": 17968120104960.0,
      "grad_norm": 2.105024600114082,
      "language_loss": 0.83237052,
      "learning_rate": 1.8560149135816659e-06,
      "loss": 0.85449219,
      "num_input_tokens_seen": 96473990,
      "step": 4468,
      "time_per_iteration": 2.6543118953704834
    },
    {
      "auxiliary_loss_clip": 0.01176399,
      "auxiliary_loss_mlp": 0.01026223,
      "balance_loss_clip": 1.05074441,
      "balance_loss_mlp": 1.01860607,
      "epoch": 0.5373654782660975,
      "flos": 15377021642880.0,
      "grad_norm": 3.0666815621290695,
      "language_loss": 0.8474623,
      "learning_rate": 1.8552379741739873e-06,
      "loss": 0.86948848,
      "num_input_tokens_seen": 96491335,
      "step": 4469,
      "time_per_iteration": 2.579596996307373
    },
    {
      "auxiliary_loss_clip": 0.0108038,
      "auxiliary_loss_mlp": 0.01001319,
      "balance_loss_clip": 1.01684022,
      "balance_loss_mlp": 1.00010288,
      "epoch": 0.5374857211567367,
      "flos": 69000091574400.0,
      "grad_norm": 0.8917970803512016,
      "language_loss": 0.55686039,
      "learning_rate": 1.8544610567265935e-06,
      "loss": 0.57767737,
      "num_input_tokens_seen": 96545275,
      "step": 4470,
      "time_per_iteration": 4.07967472076416
    },
    {
      "auxiliary_loss_clip": 0.01181357,
      "auxiliary_loss_mlp": 0.01025333,
      "balance_loss_clip": 1.05473864,
      "balance_loss_mlp": 1.01708925,
      "epoch": 0.5376059640473757,
      "flos": 15085355207040.0,
      "grad_norm": 2.366093092253901,
      "language_loss": 0.82897455,
      "learning_rate": 1.853684161357341e-06,
      "loss": 0.8510415,
      "num_input_tokens_seen": 96562935,
      "step": 4471,
      "time_per_iteration": 2.636235475540161
    },
    {
      "auxiliary_loss_clip": 0.01178087,
      "auxiliary_loss_mlp": 0.01028759,
      "balance_loss_clip": 1.05127406,
      "balance_loss_mlp": 1.02037907,
      "epoch": 0.5377262069380148,
      "flos": 19792597570560.0,
      "grad_norm": 1.9041286895504799,
      "language_loss": 0.76652336,
      "learning_rate": 1.852907288184085e-06,
      "loss": 0.7885918,
      "num_input_tokens_seen": 96581820,
      "step": 4472,
      "time_per_iteration": 2.6310489177703857
    },
    {
      "auxiliary_loss_clip": 0.01183118,
      "auxiliary_loss_mlp": 0.0102611,
      "balance_loss_clip": 1.05149698,
      "balance_loss_mlp": 1.0176456,
      "epoch": 0.5378464498286539,
      "flos": 30003077640960.0,
      "grad_norm": 1.985291437298924,
      "language_loss": 0.70313978,
      "learning_rate": 1.8521304373246762e-06,
      "loss": 0.72523212,
      "num_input_tokens_seen": 96602865,
      "step": 4473,
      "time_per_iteration": 3.5609264373779297
    },
    {
      "auxiliary_loss_clip": 0.01187417,
      "auxiliary_loss_mlp": 0.01025565,
      "balance_loss_clip": 1.05674386,
      "balance_loss_mlp": 1.01624906,
      "epoch": 0.537966692719293,
      "flos": 21251217058560.0,
      "grad_norm": 2.260110755661092,
      "language_loss": 0.88778591,
      "learning_rate": 1.8513536088969626e-06,
      "loss": 0.90991569,
      "num_input_tokens_seen": 96620530,
      "step": 4474,
      "time_per_iteration": 3.495338201522827
    },
    {
      "auxiliary_loss_clip": 0.01181862,
      "auxiliary_loss_mlp": 0.01031395,
      "balance_loss_clip": 1.05335152,
      "balance_loss_mlp": 1.0227704,
      "epoch": 0.538086935609932,
      "flos": 21543170803200.0,
      "grad_norm": 4.097221496351495,
      "language_loss": 0.80104047,
      "learning_rate": 1.8505768030187884e-06,
      "loss": 0.82317305,
      "num_input_tokens_seen": 96640660,
      "step": 4475,
      "time_per_iteration": 2.6491622924804688
    },
    {
      "auxiliary_loss_clip": 0.01181937,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.05420685,
      "balance_loss_mlp": 1.01911676,
      "epoch": 0.5382071785005712,
      "flos": 22747219626240.0,
      "grad_norm": 1.6376301349562288,
      "language_loss": 0.79974687,
      "learning_rate": 1.849800019807995e-06,
      "loss": 0.82183886,
      "num_input_tokens_seen": 96661885,
      "step": 4476,
      "time_per_iteration": 2.624868631362915
    },
    {
      "auxiliary_loss_clip": 0.01182962,
      "auxiliary_loss_mlp": 0.01027416,
      "balance_loss_clip": 1.05261648,
      "balance_loss_mlp": 1.01929784,
      "epoch": 0.5383274213912103,
      "flos": 24934574240640.0,
      "grad_norm": 23.33872185698518,
      "language_loss": 0.71263921,
      "learning_rate": 1.8490232593824186e-06,
      "loss": 0.734743,
      "num_input_tokens_seen": 96678340,
      "step": 4477,
      "time_per_iteration": 2.648407220840454
    },
    {
      "auxiliary_loss_clip": 0.01180113,
      "auxiliary_loss_mlp": 0.01025389,
      "balance_loss_clip": 1.05368721,
      "balance_loss_mlp": 1.01776528,
      "epoch": 0.5384476642818493,
      "flos": 22310186849280.0,
      "grad_norm": 2.9873222596725566,
      "language_loss": 0.84984422,
      "learning_rate": 1.8482465218598935e-06,
      "loss": 0.87189925,
      "num_input_tokens_seen": 96698285,
      "step": 4478,
      "time_per_iteration": 2.6455447673797607
    },
    {
      "auxiliary_loss_clip": 0.01182548,
      "auxiliary_loss_mlp": 0.0103469,
      "balance_loss_clip": 1.05425072,
      "balance_loss_mlp": 1.02554035,
      "epoch": 0.5385679071724885,
      "flos": 22711021695360.0,
      "grad_norm": 2.266639163474921,
      "language_loss": 0.83104134,
      "learning_rate": 1.8474698073582508e-06,
      "loss": 0.85321367,
      "num_input_tokens_seen": 96719655,
      "step": 4479,
      "time_per_iteration": 2.6453442573547363
    },
    {
      "auxiliary_loss_clip": 0.01182197,
      "auxiliary_loss_mlp": 0.01025573,
      "balance_loss_clip": 1.05361557,
      "balance_loss_mlp": 1.01752043,
      "epoch": 0.5386881500631275,
      "flos": 15953746412160.0,
      "grad_norm": 2.8244522720168384,
      "language_loss": 0.87503326,
      "learning_rate": 1.8466931159953166e-06,
      "loss": 0.89711094,
      "num_input_tokens_seen": 96736290,
      "step": 4480,
      "time_per_iteration": 2.6020381450653076
    },
    {
      "auxiliary_loss_clip": 0.01183794,
      "auxiliary_loss_mlp": 0.01027232,
      "balance_loss_clip": 1.05507088,
      "balance_loss_mlp": 1.01886344,
      "epoch": 0.5388083929537666,
      "flos": 24060041809920.0,
      "grad_norm": 6.229079857777933,
      "language_loss": 0.84328318,
      "learning_rate": 1.8459164478889158e-06,
      "loss": 0.8653934,
      "num_input_tokens_seen": 96757685,
      "step": 4481,
      "time_per_iteration": 2.563732862472534
    },
    {
      "auxiliary_loss_clip": 0.01177025,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.05078459,
      "balance_loss_mlp": 1.01859236,
      "epoch": 0.5389286358444056,
      "flos": 22236893147520.0,
      "grad_norm": 1.8464258597055088,
      "language_loss": 0.76320052,
      "learning_rate": 1.8451398031568663e-06,
      "loss": 0.78523356,
      "num_input_tokens_seen": 96777310,
      "step": 4482,
      "time_per_iteration": 2.654705286026001
    },
    {
      "auxiliary_loss_clip": 0.01183038,
      "auxiliary_loss_mlp": 0.01027094,
      "balance_loss_clip": 1.05387044,
      "balance_loss_mlp": 1.01857066,
      "epoch": 0.5390488787350448,
      "flos": 24281718595200.0,
      "grad_norm": 3.8553223420332263,
      "language_loss": 0.7481612,
      "learning_rate": 1.844363181916986e-06,
      "loss": 0.77026254,
      "num_input_tokens_seen": 96798035,
      "step": 4483,
      "time_per_iteration": 2.6419215202331543
    },
    {
      "auxiliary_loss_clip": 0.01184041,
      "auxiliary_loss_mlp": 0.01026601,
      "balance_loss_clip": 1.05336785,
      "balance_loss_mlp": 1.01780915,
      "epoch": 0.5391691216256839,
      "flos": 16581393688320.0,
      "grad_norm": 2.8333038568817948,
      "language_loss": 0.83392,
      "learning_rate": 1.8435865842870868e-06,
      "loss": 0.85602641,
      "num_input_tokens_seen": 96815975,
      "step": 4484,
      "time_per_iteration": 2.627570390701294
    },
    {
      "auxiliary_loss_clip": 0.01178438,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.05211127,
      "balance_loss_mlp": 1.01626754,
      "epoch": 0.5392893645163229,
      "flos": 23330049707520.0,
      "grad_norm": 2.5115148655935484,
      "language_loss": 0.718997,
      "learning_rate": 1.8428100103849787e-06,
      "loss": 0.74101901,
      "num_input_tokens_seen": 96835770,
      "step": 4485,
      "time_per_iteration": 2.6180479526519775
    },
    {
      "auxiliary_loss_clip": 0.01183245,
      "auxiliary_loss_mlp": 0.01023371,
      "balance_loss_clip": 1.05579245,
      "balance_loss_mlp": 1.01512146,
      "epoch": 0.5394096074069621,
      "flos": 15669801400320.0,
      "grad_norm": 2.2348522382761753,
      "language_loss": 0.73183,
      "learning_rate": 1.842033460328467e-06,
      "loss": 0.75389612,
      "num_input_tokens_seen": 96854490,
      "step": 4486,
      "time_per_iteration": 2.5767202377319336
    },
    {
      "auxiliary_loss_clip": 0.01179744,
      "auxiliary_loss_mlp": 0.01025001,
      "balance_loss_clip": 1.05388904,
      "balance_loss_mlp": 1.01713276,
      "epoch": 0.5395298502976011,
      "flos": 22893447893760.0,
      "grad_norm": 2.2603161315993625,
      "language_loss": 0.75048488,
      "learning_rate": 1.8412569342353541e-06,
      "loss": 0.77253234,
      "num_input_tokens_seen": 96874645,
      "step": 4487,
      "time_per_iteration": 2.705782651901245
    },
    {
      "auxiliary_loss_clip": 0.01186294,
      "auxiliary_loss_mlp": 0.01029686,
      "balance_loss_clip": 1.05500829,
      "balance_loss_mlp": 1.0208112,
      "epoch": 0.5396500931882402,
      "flos": 23842135952640.0,
      "grad_norm": 1.8090655059822385,
      "language_loss": 0.84911215,
      "learning_rate": 1.840480432223438e-06,
      "loss": 0.87127197,
      "num_input_tokens_seen": 96893650,
      "step": 4488,
      "time_per_iteration": 2.638397693634033
    },
    {
      "auxiliary_loss_clip": 0.01178601,
      "auxiliary_loss_mlp": 0.01023676,
      "balance_loss_clip": 1.05206549,
      "balance_loss_mlp": 1.01550972,
      "epoch": 0.5397703360788794,
      "flos": 26322988596480.0,
      "grad_norm": 2.3661726020574894,
      "language_loss": 0.77670515,
      "learning_rate": 1.8397039544105131e-06,
      "loss": 0.79872793,
      "num_input_tokens_seen": 96912735,
      "step": 4489,
      "time_per_iteration": 2.608595848083496
    },
    {
      "auxiliary_loss_clip": 0.0118198,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.05320024,
      "balance_loss_mlp": 1.02100623,
      "epoch": 0.5398905789695184,
      "flos": 21214588164480.0,
      "grad_norm": 2.2660355740708367,
      "language_loss": 0.69938296,
      "learning_rate": 1.8389275009143711e-06,
      "loss": 0.72149009,
      "num_input_tokens_seen": 96932475,
      "step": 4490,
      "time_per_iteration": 2.6059038639068604
    },
    {
      "auxiliary_loss_clip": 0.01178031,
      "auxiliary_loss_mlp": 0.01022133,
      "balance_loss_clip": 1.05118132,
      "balance_loss_mlp": 1.01409864,
      "epoch": 0.5400108218601575,
      "flos": 25080335631360.0,
      "grad_norm": 1.6907637977454084,
      "language_loss": 0.7334764,
      "learning_rate": 1.8381510718527988e-06,
      "loss": 0.75547802,
      "num_input_tokens_seen": 96952085,
      "step": 4491,
      "time_per_iteration": 2.634897232055664
    },
    {
      "auxiliary_loss_clip": 0.01181313,
      "auxiliary_loss_mlp": 0.01027816,
      "balance_loss_clip": 1.05296779,
      "balance_loss_mlp": 1.01882148,
      "epoch": 0.5401310647507966,
      "flos": 26357498588160.0,
      "grad_norm": 2.123529472048964,
      "language_loss": 0.63874924,
      "learning_rate": 1.8373746673435812e-06,
      "loss": 0.66084051,
      "num_input_tokens_seen": 96973110,
      "step": 4492,
      "time_per_iteration": 2.6553702354431152
    },
    {
      "auxiliary_loss_clip": 0.01182398,
      "auxiliary_loss_mlp": 0.01025814,
      "balance_loss_clip": 1.05425191,
      "balance_loss_mlp": 1.01739216,
      "epoch": 0.5402513076414357,
      "flos": 27855332749440.0,
      "grad_norm": 1.8804739170674418,
      "language_loss": 0.79441208,
      "learning_rate": 1.8365982875044964e-06,
      "loss": 0.81649423,
      "num_input_tokens_seen": 96993420,
      "step": 4493,
      "time_per_iteration": 3.632643222808838
    },
    {
      "auxiliary_loss_clip": 0.01183931,
      "auxiliary_loss_mlp": 0.01030156,
      "balance_loss_clip": 1.05396461,
      "balance_loss_mlp": 1.02162647,
      "epoch": 0.5403715505320748,
      "flos": 22893771116160.0,
      "grad_norm": 2.118352310981962,
      "language_loss": 0.75931931,
      "learning_rate": 1.8358219324533217e-06,
      "loss": 0.78146023,
      "num_input_tokens_seen": 97013685,
      "step": 4494,
      "time_per_iteration": 2.6725807189941406
    },
    {
      "auxiliary_loss_clip": 0.01175435,
      "auxiliary_loss_mlp": 0.0102288,
      "balance_loss_clip": 1.05213404,
      "balance_loss_mlp": 1.01598418,
      "epoch": 0.5404917934227139,
      "flos": 30224143895040.0,
      "grad_norm": 1.9638057202931622,
      "language_loss": 0.70267224,
      "learning_rate": 1.8350456023078292e-06,
      "loss": 0.72465539,
      "num_input_tokens_seen": 97036060,
      "step": 4495,
      "time_per_iteration": 2.6967720985412598
    },
    {
      "auxiliary_loss_clip": 0.01186081,
      "auxiliary_loss_mlp": 0.01028377,
      "balance_loss_clip": 1.05481339,
      "balance_loss_mlp": 1.02049756,
      "epoch": 0.540612036313353,
      "flos": 19938502615680.0,
      "grad_norm": 5.953774691294458,
      "language_loss": 0.78047693,
      "learning_rate": 1.8342692971857874e-06,
      "loss": 0.80262154,
      "num_input_tokens_seen": 97055260,
      "step": 4496,
      "time_per_iteration": 2.637911081314087
    },
    {
      "auxiliary_loss_clip": 0.01179081,
      "auxiliary_loss_mlp": 0.01030101,
      "balance_loss_clip": 1.05130863,
      "balance_loss_mlp": 1.02190578,
      "epoch": 0.540732279203992,
      "flos": 24279599692800.0,
      "grad_norm": 2.7625845806590394,
      "language_loss": 0.71364832,
      "learning_rate": 1.833493017204962e-06,
      "loss": 0.73574007,
      "num_input_tokens_seen": 97075365,
      "step": 4497,
      "time_per_iteration": 3.5614306926727295
    },
    {
      "auxiliary_loss_clip": 0.01180073,
      "auxiliary_loss_mlp": 0.01028257,
      "balance_loss_clip": 1.05191684,
      "balance_loss_mlp": 1.02032375,
      "epoch": 0.5408525220946312,
      "flos": 20193216935040.0,
      "grad_norm": 2.080787799931111,
      "language_loss": 0.77939773,
      "learning_rate": 1.8327167624831134e-06,
      "loss": 0.80148107,
      "num_input_tokens_seen": 97093095,
      "step": 4498,
      "time_per_iteration": 2.640489101409912
    },
    {
      "auxiliary_loss_clip": 0.01180369,
      "auxiliary_loss_mlp": 0.01022962,
      "balance_loss_clip": 1.05329776,
      "balance_loss_mlp": 1.01455164,
      "epoch": 0.5409727649852702,
      "flos": 24134448833280.0,
      "grad_norm": 1.965066794913054,
      "language_loss": 0.70883501,
      "learning_rate": 1.831940533137999e-06,
      "loss": 0.73086834,
      "num_input_tokens_seen": 97112000,
      "step": 4499,
      "time_per_iteration": 3.565523386001587
    },
    {
      "auxiliary_loss_clip": 0.01179113,
      "auxiliary_loss_mlp": 0.0102659,
      "balance_loss_clip": 1.05186248,
      "balance_loss_mlp": 1.01800728,
      "epoch": 0.5410930078759093,
      "flos": 23912700220800.0,
      "grad_norm": 2.5573737860513357,
      "language_loss": 0.72474337,
      "learning_rate": 1.8311643292873718e-06,
      "loss": 0.74680042,
      "num_input_tokens_seen": 97130820,
      "step": 4500,
      "time_per_iteration": 3.5080292224884033
    },
    {
      "auxiliary_loss_clip": 0.01176792,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.05135894,
      "balance_loss_mlp": 1.01945961,
      "epoch": 0.5412132507665485,
      "flos": 21105132445440.0,
      "grad_norm": 1.862816413457725,
      "language_loss": 0.88182747,
      "learning_rate": 1.8303881510489818e-06,
      "loss": 0.90386629,
      "num_input_tokens_seen": 97149210,
      "step": 4501,
      "time_per_iteration": 2.5692214965820312
    },
    {
      "auxiliary_loss_clip": 0.01184029,
      "auxiliary_loss_mlp": 0.01026217,
      "balance_loss_clip": 1.05359519,
      "balance_loss_mlp": 1.01829004,
      "epoch": 0.5413334936571875,
      "flos": 30227340205440.0,
      "grad_norm": 32.394941702082235,
      "language_loss": 0.69425642,
      "learning_rate": 1.829611998540574e-06,
      "loss": 0.7163589,
      "num_input_tokens_seen": 97170415,
      "step": 4502,
      "time_per_iteration": 2.6254162788391113
    },
    {
      "auxiliary_loss_clip": 0.01181133,
      "auxiliary_loss_mlp": 0.01024901,
      "balance_loss_clip": 1.05093563,
      "balance_loss_mlp": 1.01662767,
      "epoch": 0.5414537365478266,
      "flos": 24279635606400.0,
      "grad_norm": 15.2837986695116,
      "language_loss": 0.79890597,
      "learning_rate": 1.8288358718798914e-06,
      "loss": 0.82096624,
      "num_input_tokens_seen": 97189605,
      "step": 4503,
      "time_per_iteration": 2.62978458404541
    },
    {
      "auxiliary_loss_clip": 0.01181149,
      "auxiliary_loss_mlp": 0.0103308,
      "balance_loss_clip": 1.05402994,
      "balance_loss_mlp": 1.0249114,
      "epoch": 0.5415739794384657,
      "flos": 16654543735680.0,
      "grad_norm": 2.3324301105564254,
      "language_loss": 0.72716153,
      "learning_rate": 1.8280597711846703e-06,
      "loss": 0.74930382,
      "num_input_tokens_seen": 97207845,
      "step": 4504,
      "time_per_iteration": 2.661670446395874
    },
    {
      "auxiliary_loss_clip": 0.01178564,
      "auxiliary_loss_mlp": 0.01023881,
      "balance_loss_clip": 1.05172491,
      "balance_loss_mlp": 1.01666319,
      "epoch": 0.5416942223291048,
      "flos": 23185724860800.0,
      "grad_norm": 2.0953724670339633,
      "language_loss": 0.83545542,
      "learning_rate": 1.8272836965726455e-06,
      "loss": 0.85747981,
      "num_input_tokens_seen": 97226780,
      "step": 4505,
      "time_per_iteration": 2.6646969318389893
    },
    {
      "auxiliary_loss_clip": 0.01181153,
      "auxiliary_loss_mlp": 0.01031831,
      "balance_loss_clip": 1.05366206,
      "balance_loss_mlp": 1.02365613,
      "epoch": 0.5418144652197439,
      "flos": 20303247271680.0,
      "grad_norm": 2.1959261833413026,
      "language_loss": 0.78688538,
      "learning_rate": 1.8265076481615461e-06,
      "loss": 0.80901515,
      "num_input_tokens_seen": 97246695,
      "step": 4506,
      "time_per_iteration": 2.6455578804016113
    },
    {
      "auxiliary_loss_clip": 0.01186165,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.05599451,
      "balance_loss_mlp": 1.01793635,
      "epoch": 0.541934708110383,
      "flos": 12458633431680.0,
      "grad_norm": 2.655470196583913,
      "language_loss": 0.87349951,
      "learning_rate": 1.8257316260690987e-06,
      "loss": 0.89562303,
      "num_input_tokens_seen": 97264480,
      "step": 4507,
      "time_per_iteration": 2.5680341720581055
    },
    {
      "auxiliary_loss_clip": 0.01177305,
      "auxiliary_loss_mlp": 0.01026481,
      "balance_loss_clip": 1.05238342,
      "balance_loss_mlp": 1.01856565,
      "epoch": 0.5420549510010221,
      "flos": 21253802837760.0,
      "grad_norm": 1.6484258574152733,
      "language_loss": 0.76124632,
      "learning_rate": 1.8249556304130254e-06,
      "loss": 0.78328419,
      "num_input_tokens_seen": 97285760,
      "step": 4508,
      "time_per_iteration": 2.6002676486968994
    },
    {
      "auxiliary_loss_clip": 0.01179261,
      "auxiliary_loss_mlp": 0.01026286,
      "balance_loss_clip": 1.05251145,
      "balance_loss_mlp": 1.01843345,
      "epoch": 0.5421751938916611,
      "flos": 29490524519040.0,
      "grad_norm": 1.9960943669072275,
      "language_loss": 0.68286514,
      "learning_rate": 1.824179661311044e-06,
      "loss": 0.70492059,
      "num_input_tokens_seen": 97304510,
      "step": 4509,
      "time_per_iteration": 2.6797430515289307
    },
    {
      "auxiliary_loss_clip": 0.01177636,
      "auxiliary_loss_mlp": 0.01024355,
      "balance_loss_clip": 1.05103374,
      "balance_loss_mlp": 1.01667786,
      "epoch": 0.5422954367823003,
      "flos": 18734238311040.0,
      "grad_norm": 5.857982665462605,
      "language_loss": 0.7985003,
      "learning_rate": 1.823403718880868e-06,
      "loss": 0.82052028,
      "num_input_tokens_seen": 97323270,
      "step": 4510,
      "time_per_iteration": 2.5733585357666016
    },
    {
      "auxiliary_loss_clip": 0.01182549,
      "auxiliary_loss_mlp": 0.01027163,
      "balance_loss_clip": 1.05315173,
      "balance_loss_mlp": 1.01896191,
      "epoch": 0.5424156796729394,
      "flos": 39969006940800.0,
      "grad_norm": 6.149163753643927,
      "language_loss": 0.66896355,
      "learning_rate": 1.822627803240207e-06,
      "loss": 0.69106066,
      "num_input_tokens_seen": 97345600,
      "step": 4511,
      "time_per_iteration": 2.799945592880249
    },
    {
      "auxiliary_loss_clip": 0.01182979,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.05407047,
      "balance_loss_mlp": 1.01755714,
      "epoch": 0.5425359225635784,
      "flos": 11546538353280.0,
      "grad_norm": 2.576539645355899,
      "language_loss": 0.85223407,
      "learning_rate": 1.8218519145067675e-06,
      "loss": 0.87431985,
      "num_input_tokens_seen": 97361220,
      "step": 4512,
      "time_per_iteration": 2.5605082511901855
    },
    {
      "auxiliary_loss_clip": 0.01180224,
      "auxiliary_loss_mlp": 0.01025239,
      "balance_loss_clip": 1.05198979,
      "balance_loss_mlp": 1.01691198,
      "epoch": 0.5426561654542175,
      "flos": 20229702174720.0,
      "grad_norm": 1.8977612681079192,
      "language_loss": 0.89611876,
      "learning_rate": 1.8210760527982508e-06,
      "loss": 0.91817343,
      "num_input_tokens_seen": 97381505,
      "step": 4513,
      "time_per_iteration": 2.5918526649475098
    },
    {
      "auxiliary_loss_clip": 0.01182488,
      "auxiliary_loss_mlp": 0.01026044,
      "balance_loss_clip": 1.05625391,
      "balance_loss_mlp": 1.01827145,
      "epoch": 0.5427764083448566,
      "flos": 21871681614720.0,
      "grad_norm": 1.9245763386987365,
      "language_loss": 0.75051415,
      "learning_rate": 1.8203002182323552e-06,
      "loss": 0.77259946,
      "num_input_tokens_seen": 97399060,
      "step": 4514,
      "time_per_iteration": 2.5754168033599854
    },
    {
      "auxiliary_loss_clip": 0.01181788,
      "auxiliary_loss_mlp": 0.01025949,
      "balance_loss_clip": 1.05304849,
      "balance_loss_mlp": 1.01776576,
      "epoch": 0.5428966512354957,
      "flos": 19640946349440.0,
      "grad_norm": 2.4253238891942948,
      "language_loss": 0.75946635,
      "learning_rate": 1.819524410926773e-06,
      "loss": 0.78154373,
      "num_input_tokens_seen": 97416740,
      "step": 4515,
      "time_per_iteration": 2.622857093811035
    },
    {
      "auxiliary_loss_clip": 0.01179835,
      "auxiliary_loss_mlp": 0.01030943,
      "balance_loss_clip": 1.05356991,
      "balance_loss_mlp": 1.0220201,
      "epoch": 0.5430168941261347,
      "flos": 22382187661440.0,
      "grad_norm": 1.5246847869220654,
      "language_loss": 0.76872873,
      "learning_rate": 1.8187486309991944e-06,
      "loss": 0.79083651,
      "num_input_tokens_seen": 97437620,
      "step": 4516,
      "time_per_iteration": 2.595642328262329
    },
    {
      "auxiliary_loss_clip": 0.01182175,
      "auxiliary_loss_mlp": 0.01025137,
      "balance_loss_clip": 1.05386615,
      "balance_loss_mlp": 1.01713264,
      "epoch": 0.5431371370167739,
      "flos": 18764187275520.0,
      "grad_norm": 2.1447485404065483,
      "language_loss": 0.77804923,
      "learning_rate": 1.817972878567304e-06,
      "loss": 0.80012238,
      "num_input_tokens_seen": 97456275,
      "step": 4517,
      "time_per_iteration": 2.709555149078369
    },
    {
      "auxiliary_loss_clip": 0.01179527,
      "auxiliary_loss_mlp": 0.01024517,
      "balance_loss_clip": 1.05347729,
      "balance_loss_mlp": 1.01712632,
      "epoch": 0.543257379907413,
      "flos": 18806023641600.0,
      "grad_norm": 2.0216858307830763,
      "language_loss": 0.7649796,
      "learning_rate": 1.8171971537487834e-06,
      "loss": 0.78702009,
      "num_input_tokens_seen": 97474925,
      "step": 4518,
      "time_per_iteration": 2.607149839401245
    },
    {
      "auxiliary_loss_clip": 0.01181446,
      "auxiliary_loss_mlp": 0.0102292,
      "balance_loss_clip": 1.05239439,
      "balance_loss_mlp": 1.01495075,
      "epoch": 0.543377622798052,
      "flos": 17493381025920.0,
      "grad_norm": 2.472605118264161,
      "language_loss": 0.80976474,
      "learning_rate": 1.8164214566613093e-06,
      "loss": 0.83180833,
      "num_input_tokens_seen": 97493550,
      "step": 4519,
      "time_per_iteration": 2.5625979900360107
    },
    {
      "auxiliary_loss_clip": 0.01178741,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.05224514,
      "balance_loss_mlp": 1.01690221,
      "epoch": 0.5434978656886912,
      "flos": 18989311766400.0,
      "grad_norm": 2.734913847818426,
      "language_loss": 0.65479463,
      "learning_rate": 1.8156457874225547e-06,
      "loss": 0.67683238,
      "num_input_tokens_seen": 97512010,
      "step": 4520,
      "time_per_iteration": 3.5435547828674316
    },
    {
      "auxiliary_loss_clip": 0.01178569,
      "auxiliary_loss_mlp": 0.01025951,
      "balance_loss_clip": 1.05235898,
      "balance_loss_mlp": 1.01810074,
      "epoch": 0.5436181085793302,
      "flos": 17274936464640.0,
      "grad_norm": 2.210188703833873,
      "language_loss": 0.80729502,
      "learning_rate": 1.814870146150187e-06,
      "loss": 0.82934022,
      "num_input_tokens_seen": 97530120,
      "step": 4521,
      "time_per_iteration": 2.564466714859009
    },
    {
      "auxiliary_loss_clip": 0.01181911,
      "auxiliary_loss_mlp": 0.01025987,
      "balance_loss_clip": 1.05258286,
      "balance_loss_mlp": 1.01802373,
      "epoch": 0.5437383514699693,
      "flos": 19098587917440.0,
      "grad_norm": 3.837352379170568,
      "language_loss": 0.78877574,
      "learning_rate": 1.814094532961871e-06,
      "loss": 0.81085473,
      "num_input_tokens_seen": 97548695,
      "step": 4522,
      "time_per_iteration": 2.6310551166534424
    },
    {
      "auxiliary_loss_clip": 0.01180487,
      "auxiliary_loss_mlp": 0.01025028,
      "balance_loss_clip": 1.05298519,
      "balance_loss_mlp": 1.01711226,
      "epoch": 0.5438585943606085,
      "flos": 22602715211520.0,
      "grad_norm": 2.0353219985900166,
      "language_loss": 0.83667636,
      "learning_rate": 1.8133189479752666e-06,
      "loss": 0.85873151,
      "num_input_tokens_seen": 97567625,
      "step": 4523,
      "time_per_iteration": 2.618086099624634
    },
    {
      "auxiliary_loss_clip": 0.0118106,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.05408466,
      "balance_loss_mlp": 1.01802659,
      "epoch": 0.5439788372512475,
      "flos": 21798495653760.0,
      "grad_norm": 2.30137959227894,
      "language_loss": 0.81907743,
      "learning_rate": 1.8125433913080292e-06,
      "loss": 0.84114414,
      "num_input_tokens_seen": 97585325,
      "step": 4524,
      "time_per_iteration": 3.5671093463897705
    },
    {
      "auxiliary_loss_clip": 0.01179422,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.05304527,
      "balance_loss_mlp": 1.02007103,
      "epoch": 0.5440990801418866,
      "flos": 16399362539520.0,
      "grad_norm": 2.257048953643444,
      "language_loss": 0.83005643,
      "learning_rate": 1.811767863077811e-06,
      "loss": 0.85212362,
      "num_input_tokens_seen": 97604275,
      "step": 4525,
      "time_per_iteration": 2.5591204166412354
    },
    {
      "auxiliary_loss_clip": 0.01175557,
      "auxiliary_loss_mlp": 0.01025486,
      "balance_loss_clip": 1.05148721,
      "balance_loss_mlp": 1.01783299,
      "epoch": 0.5442193230325257,
      "flos": 21615638492160.0,
      "grad_norm": 5.6365440474112205,
      "language_loss": 0.78257573,
      "learning_rate": 1.8109923634022577e-06,
      "loss": 0.80458605,
      "num_input_tokens_seen": 97624300,
      "step": 4526,
      "time_per_iteration": 3.536116123199463
    },
    {
      "auxiliary_loss_clip": 0.01184497,
      "auxiliary_loss_mlp": 0.01026536,
      "balance_loss_clip": 1.05538213,
      "balance_loss_mlp": 1.01814997,
      "epoch": 0.5443395659231648,
      "flos": 15481198062720.0,
      "grad_norm": 4.830767013287441,
      "language_loss": 0.86524034,
      "learning_rate": 1.8102168923990128e-06,
      "loss": 0.88735074,
      "num_input_tokens_seen": 97637845,
      "step": 4527,
      "time_per_iteration": 3.540945291519165
    },
    {
      "auxiliary_loss_clip": 0.011827,
      "auxiliary_loss_mlp": 0.01030328,
      "balance_loss_clip": 1.05533493,
      "balance_loss_mlp": 1.02239525,
      "epoch": 0.5444598088138038,
      "flos": 18770436241920.0,
      "grad_norm": 2.4455580505214005,
      "language_loss": 0.80043101,
      "learning_rate": 1.809441450185714e-06,
      "loss": 0.82256138,
      "num_input_tokens_seen": 97656330,
      "step": 4528,
      "time_per_iteration": 2.6238319873809814
    },
    {
      "auxiliary_loss_clip": 0.01179382,
      "auxiliary_loss_mlp": 0.01022364,
      "balance_loss_clip": 1.05197167,
      "balance_loss_mlp": 1.01477051,
      "epoch": 0.544580051704443,
      "flos": 21142335957120.0,
      "grad_norm": 2.6546400699157253,
      "language_loss": 0.73081756,
      "learning_rate": 1.8086660368799958e-06,
      "loss": 0.75283504,
      "num_input_tokens_seen": 97674380,
      "step": 4529,
      "time_per_iteration": 2.6416471004486084
    },
    {
      "auxiliary_loss_clip": 0.01183811,
      "auxiliary_loss_mlp": 0.01026142,
      "balance_loss_clip": 1.0541923,
      "balance_loss_mlp": 1.01754677,
      "epoch": 0.5447002945950821,
      "flos": 32491508054400.0,
      "grad_norm": 1.834502334627215,
      "language_loss": 0.77360362,
      "learning_rate": 1.807890652599488e-06,
      "loss": 0.79570317,
      "num_input_tokens_seen": 97698765,
      "step": 4530,
      "time_per_iteration": 2.683349609375
    },
    {
      "auxiliary_loss_clip": 0.01176386,
      "auxiliary_loss_mlp": 0.01024931,
      "balance_loss_clip": 1.05201375,
      "balance_loss_mlp": 1.01710546,
      "epoch": 0.5448205374857211,
      "flos": 11798307757440.0,
      "grad_norm": 2.20538539861503,
      "language_loss": 0.82636607,
      "learning_rate": 1.8071152974618156e-06,
      "loss": 0.84837919,
      "num_input_tokens_seen": 97716565,
      "step": 4531,
      "time_per_iteration": 2.6194546222686768
    },
    {
      "auxiliary_loss_clip": 0.01176884,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.05123758,
      "balance_loss_mlp": 1.02059722,
      "epoch": 0.5449407803763603,
      "flos": 24133766474880.0,
      "grad_norm": 4.176957666845627,
      "language_loss": 0.78411669,
      "learning_rate": 1.806339971584599e-06,
      "loss": 0.80616724,
      "num_input_tokens_seen": 97733225,
      "step": 4532,
      "time_per_iteration": 2.6311192512512207
    },
    {
      "auxiliary_loss_clip": 0.01178485,
      "auxiliary_loss_mlp": 0.01028305,
      "balance_loss_clip": 1.05183578,
      "balance_loss_mlp": 1.02077138,
      "epoch": 0.5450610232669993,
      "flos": 23258551685760.0,
      "grad_norm": 2.1637312223184932,
      "language_loss": 0.85139203,
      "learning_rate": 1.8055646750854546e-06,
      "loss": 0.87345994,
      "num_input_tokens_seen": 97752735,
      "step": 4533,
      "time_per_iteration": 2.6145615577697754
    },
    {
      "auxiliary_loss_clip": 0.01181359,
      "auxiliary_loss_mlp": 0.01028075,
      "balance_loss_clip": 1.05391669,
      "balance_loss_mlp": 1.02022481,
      "epoch": 0.5451812661576384,
      "flos": 17785083375360.0,
      "grad_norm": 3.7861003466298495,
      "language_loss": 0.81610942,
      "learning_rate": 1.8047894080819945e-06,
      "loss": 0.83820379,
      "num_input_tokens_seen": 97769985,
      "step": 4534,
      "time_per_iteration": 2.6179161071777344
    },
    {
      "auxiliary_loss_clip": 0.01084828,
      "auxiliary_loss_mlp": 0.01005309,
      "balance_loss_clip": 1.01992536,
      "balance_loss_mlp": 1.00415218,
      "epoch": 0.5453015090482776,
      "flos": 71062586513280.0,
      "grad_norm": 0.722209905073873,
      "language_loss": 0.63166666,
      "learning_rate": 1.8040141706918258e-06,
      "loss": 0.65256798,
      "num_input_tokens_seen": 97831225,
      "step": 4535,
      "time_per_iteration": 3.3095104694366455
    },
    {
      "auxiliary_loss_clip": 0.01182782,
      "auxiliary_loss_mlp": 0.01023222,
      "balance_loss_clip": 1.05588484,
      "balance_loss_mlp": 1.01524734,
      "epoch": 0.5454217519389166,
      "flos": 25552201622400.0,
      "grad_norm": 1.9820503236240887,
      "language_loss": 0.77187437,
      "learning_rate": 1.8032389630325525e-06,
      "loss": 0.7939344,
      "num_input_tokens_seen": 97849975,
      "step": 4536,
      "time_per_iteration": 2.6821558475494385
    },
    {
      "auxiliary_loss_clip": 0.01179985,
      "auxiliary_loss_mlp": 0.01022115,
      "balance_loss_clip": 1.05212545,
      "balance_loss_mlp": 1.01458371,
      "epoch": 0.5455419948295557,
      "flos": 23658345037440.0,
      "grad_norm": 17.39755546175868,
      "language_loss": 0.75658274,
      "learning_rate": 1.8024637852217707e-06,
      "loss": 0.77860373,
      "num_input_tokens_seen": 97869700,
      "step": 4537,
      "time_per_iteration": 2.6289095878601074
    },
    {
      "auxiliary_loss_clip": 0.01179374,
      "auxiliary_loss_mlp": 0.01024567,
      "balance_loss_clip": 1.05295396,
      "balance_loss_mlp": 1.01713419,
      "epoch": 0.5456622377201948,
      "flos": 23403989854080.0,
      "grad_norm": 2.0686625579318862,
      "language_loss": 0.85029221,
      "learning_rate": 1.8016886373770766e-06,
      "loss": 0.87233156,
      "num_input_tokens_seen": 97888215,
      "step": 4538,
      "time_per_iteration": 2.652163028717041
    },
    {
      "auxiliary_loss_clip": 0.01182414,
      "auxiliary_loss_mlp": 0.01030156,
      "balance_loss_clip": 1.0543859,
      "balance_loss_mlp": 1.02234221,
      "epoch": 0.5457824806108339,
      "flos": 23988040997760.0,
      "grad_norm": 1.9532282588851753,
      "language_loss": 0.78740501,
      "learning_rate": 1.8009135196160579e-06,
      "loss": 0.80953074,
      "num_input_tokens_seen": 97907090,
      "step": 4539,
      "time_per_iteration": 2.6533658504486084
    },
    {
      "auxiliary_loss_clip": 0.01181372,
      "auxiliary_loss_mlp": 0.01023765,
      "balance_loss_clip": 1.05392933,
      "balance_loss_mlp": 1.01634467,
      "epoch": 0.545902723501473,
      "flos": 22565870835840.0,
      "grad_norm": 6.3942409775998135,
      "language_loss": 0.84419978,
      "learning_rate": 1.8001384320563e-06,
      "loss": 0.86625111,
      "num_input_tokens_seen": 97927345,
      "step": 4540,
      "time_per_iteration": 2.6909117698669434
    },
    {
      "auxiliary_loss_clip": 0.01082958,
      "auxiliary_loss_mlp": 0.01003258,
      "balance_loss_clip": 1.01864707,
      "balance_loss_mlp": 1.00213742,
      "epoch": 0.5460229663921121,
      "flos": 55198399685760.0,
      "grad_norm": 0.7699411555417301,
      "language_loss": 0.57719421,
      "learning_rate": 1.7993633748153833e-06,
      "loss": 0.59805638,
      "num_input_tokens_seen": 97981950,
      "step": 4541,
      "time_per_iteration": 3.050992012023926
    },
    {
      "auxiliary_loss_clip": 0.0118038,
      "auxiliary_loss_mlp": 0.01027636,
      "balance_loss_clip": 1.05283177,
      "balance_loss_mlp": 1.01948237,
      "epoch": 0.5461432092827512,
      "flos": 15413866018560.0,
      "grad_norm": 2.6334681414048866,
      "language_loss": 0.72554934,
      "learning_rate": 1.7985883480108834e-06,
      "loss": 0.74762952,
      "num_input_tokens_seen": 97999585,
      "step": 4542,
      "time_per_iteration": 2.5839831829071045
    },
    {
      "auxiliary_loss_clip": 0.01180795,
      "auxiliary_loss_mlp": 0.01024772,
      "balance_loss_clip": 1.05314374,
      "balance_loss_mlp": 1.01679075,
      "epoch": 0.5462634521733902,
      "flos": 24024921287040.0,
      "grad_norm": 2.025607674772619,
      "language_loss": 0.72294879,
      "learning_rate": 1.797813351760371e-06,
      "loss": 0.74500448,
      "num_input_tokens_seen": 98021290,
      "step": 4543,
      "time_per_iteration": 2.601039171218872
    },
    {
      "auxiliary_loss_clip": 0.01181945,
      "auxiliary_loss_mlp": 0.0102851,
      "balance_loss_clip": 1.05329204,
      "balance_loss_mlp": 1.02063024,
      "epoch": 0.5463836950640293,
      "flos": 22820944291200.0,
      "grad_norm": 4.110592566816512,
      "language_loss": 0.78066492,
      "learning_rate": 1.7970383861814116e-06,
      "loss": 0.80276948,
      "num_input_tokens_seen": 98041060,
      "step": 4544,
      "time_per_iteration": 2.607407569885254
    },
    {
      "auxiliary_loss_clip": 0.01183562,
      "auxiliary_loss_mlp": 0.01027032,
      "balance_loss_clip": 1.05441368,
      "balance_loss_mlp": 1.01865149,
      "epoch": 0.5465039379546685,
      "flos": 20448290390400.0,
      "grad_norm": 3.366918894050565,
      "language_loss": 0.74194252,
      "learning_rate": 1.7962634513915684e-06,
      "loss": 0.76404846,
      "num_input_tokens_seen": 98058410,
      "step": 4545,
      "time_per_iteration": 2.562394380569458
    },
    {
      "auxiliary_loss_clip": 0.01180678,
      "auxiliary_loss_mlp": 0.01026693,
      "balance_loss_clip": 1.05379689,
      "balance_loss_mlp": 1.01861322,
      "epoch": 0.5466241808453075,
      "flos": 17343310003200.0,
      "grad_norm": 2.1091749488487763,
      "language_loss": 0.79543257,
      "learning_rate": 1.7954885475083969e-06,
      "loss": 0.81750625,
      "num_input_tokens_seen": 98076080,
      "step": 4546,
      "time_per_iteration": 3.5317373275756836
    },
    {
      "auxiliary_loss_clip": 0.01185737,
      "auxiliary_loss_mlp": 0.01028944,
      "balance_loss_clip": 1.05564857,
      "balance_loss_mlp": 1.02018821,
      "epoch": 0.5467444237359466,
      "flos": 21617039122560.0,
      "grad_norm": 2.1131929617699687,
      "language_loss": 0.72949356,
      "learning_rate": 1.7947136746494513e-06,
      "loss": 0.75164038,
      "num_input_tokens_seen": 98096995,
      "step": 4547,
      "time_per_iteration": 2.6684393882751465
    },
    {
      "auxiliary_loss_clip": 0.0118035,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.05237198,
      "balance_loss_mlp": 1.01905847,
      "epoch": 0.5468646666265857,
      "flos": 24170467196160.0,
      "grad_norm": 3.1955388407951935,
      "language_loss": 0.88052976,
      "learning_rate": 1.793938832932277e-06,
      "loss": 0.90260792,
      "num_input_tokens_seen": 98115105,
      "step": 4548,
      "time_per_iteration": 2.6167197227478027
    },
    {
      "auxiliary_loss_clip": 0.01182148,
      "auxiliary_loss_mlp": 0.01029144,
      "balance_loss_clip": 1.0542717,
      "balance_loss_mlp": 1.02079916,
      "epoch": 0.5469849095172248,
      "flos": 27527001505920.0,
      "grad_norm": 2.172001657990452,
      "language_loss": 0.70569807,
      "learning_rate": 1.7931640224744185e-06,
      "loss": 0.72781098,
      "num_input_tokens_seen": 98135655,
      "step": 4549,
      "time_per_iteration": 2.6206300258636475
    },
    {
      "auxiliary_loss_clip": 0.01179356,
      "auxiliary_loss_mlp": 0.01027001,
      "balance_loss_clip": 1.05235708,
      "balance_loss_mlp": 1.01894498,
      "epoch": 0.5471051524078638,
      "flos": 27964680727680.0,
      "grad_norm": 1.6774430278557364,
      "language_loss": 0.73417526,
      "learning_rate": 1.7923892433934127e-06,
      "loss": 0.75623882,
      "num_input_tokens_seen": 98156730,
      "step": 4550,
      "time_per_iteration": 3.489222764968872
    },
    {
      "auxiliary_loss_clip": 0.01180096,
      "auxiliary_loss_mlp": 0.01024553,
      "balance_loss_clip": 1.05357707,
      "balance_loss_mlp": 1.0163095,
      "epoch": 0.547225395298503,
      "flos": 18150510389760.0,
      "grad_norm": 1.6909445212331018,
      "language_loss": 0.78866541,
      "learning_rate": 1.7916144958067939e-06,
      "loss": 0.81071198,
      "num_input_tokens_seen": 98174590,
      "step": 4551,
      "time_per_iteration": 2.71915864944458
    },
    {
      "auxiliary_loss_clip": 0.01179728,
      "auxiliary_loss_mlp": 0.01028918,
      "balance_loss_clip": 1.05305099,
      "balance_loss_mlp": 1.02064192,
      "epoch": 0.5473456381891421,
      "flos": 21361498790400.0,
      "grad_norm": 2.5345360835190385,
      "language_loss": 0.79054201,
      "learning_rate": 1.7908397798320905e-06,
      "loss": 0.81262851,
      "num_input_tokens_seen": 98194325,
      "step": 4552,
      "time_per_iteration": 2.6466596126556396
    },
    {
      "auxiliary_loss_clip": 0.01186036,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.05688763,
      "balance_loss_mlp": 1.01911163,
      "epoch": 0.5474658810797811,
      "flos": 19932145908480.0,
      "grad_norm": 5.056244879390911,
      "language_loss": 0.7489984,
      "learning_rate": 1.7900650955868265e-06,
      "loss": 0.77113163,
      "num_input_tokens_seen": 98213970,
      "step": 4553,
      "time_per_iteration": 3.5766701698303223
    },
    {
      "auxiliary_loss_clip": 0.01181855,
      "auxiliary_loss_mlp": 0.01032161,
      "balance_loss_clip": 1.05481648,
      "balance_loss_mlp": 1.02410865,
      "epoch": 0.5475861239704203,
      "flos": 50476217264640.0,
      "grad_norm": 2.420930713680161,
      "language_loss": 0.76758254,
      "learning_rate": 1.7892904431885202e-06,
      "loss": 0.78972268,
      "num_input_tokens_seen": 98241145,
      "step": 4554,
      "time_per_iteration": 2.937497615814209
    },
    {
      "auxiliary_loss_clip": 0.01180075,
      "auxiliary_loss_mlp": 0.01024162,
      "balance_loss_clip": 1.0539062,
      "balance_loss_mlp": 1.01678324,
      "epoch": 0.5477063668610593,
      "flos": 20705123612160.0,
      "grad_norm": 2.5004601824184363,
      "language_loss": 0.75384915,
      "learning_rate": 1.788515822754686e-06,
      "loss": 0.77589142,
      "num_input_tokens_seen": 98261565,
      "step": 4555,
      "time_per_iteration": 2.5915844440460205
    },
    {
      "auxiliary_loss_clip": 0.0118193,
      "auxiliary_loss_mlp": 0.01031468,
      "balance_loss_clip": 1.05433393,
      "balance_loss_mlp": 1.02246737,
      "epoch": 0.5478266097516984,
      "flos": 19609740408960.0,
      "grad_norm": 3.875171006129674,
      "language_loss": 0.78092557,
      "learning_rate": 1.7877412344028335e-06,
      "loss": 0.80305958,
      "num_input_tokens_seen": 98281370,
      "step": 4556,
      "time_per_iteration": 2.6376733779907227
    },
    {
      "auxiliary_loss_clip": 0.0118084,
      "auxiliary_loss_mlp": 0.01029922,
      "balance_loss_clip": 1.05418229,
      "balance_loss_mlp": 1.02221537,
      "epoch": 0.5479468526423376,
      "flos": 12896599962240.0,
      "grad_norm": 2.266868642168892,
      "language_loss": 0.76840901,
      "learning_rate": 1.7869666782504668e-06,
      "loss": 0.79051661,
      "num_input_tokens_seen": 98297950,
      "step": 4557,
      "time_per_iteration": 2.6053338050842285
    },
    {
      "auxiliary_loss_clip": 0.01176123,
      "auxiliary_loss_mlp": 0.01027371,
      "balance_loss_clip": 1.05116022,
      "balance_loss_mlp": 1.01979518,
      "epoch": 0.5480670955329766,
      "flos": 18588800142720.0,
      "grad_norm": 3.8226509026304916,
      "language_loss": 0.69142234,
      "learning_rate": 1.7861921544150867e-06,
      "loss": 0.71345729,
      "num_input_tokens_seen": 98316800,
      "step": 4558,
      "time_per_iteration": 2.6109654903411865
    },
    {
      "auxiliary_loss_clip": 0.01183175,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.05542731,
      "balance_loss_mlp": 1.02045298,
      "epoch": 0.5481873384236157,
      "flos": 15954608338560.0,
      "grad_norm": 1.8903617211573187,
      "language_loss": 0.76371509,
      "learning_rate": 1.7854176630141856e-06,
      "loss": 0.7858296,
      "num_input_tokens_seen": 98333935,
      "step": 4559,
      "time_per_iteration": 2.5789246559143066
    },
    {
      "auxiliary_loss_clip": 0.0118571,
      "auxiliary_loss_mlp": 0.01025999,
      "balance_loss_clip": 1.05635464,
      "balance_loss_mlp": 1.01805377,
      "epoch": 0.5483075813142548,
      "flos": 22783812606720.0,
      "grad_norm": 3.0668341819552087,
      "language_loss": 0.84358388,
      "learning_rate": 1.784643204165255e-06,
      "loss": 0.86570102,
      "num_input_tokens_seen": 98353255,
      "step": 4560,
      "time_per_iteration": 2.638740301132202
    },
    {
      "auxiliary_loss_clip": 0.01179483,
      "auxiliary_loss_mlp": 0.01024647,
      "balance_loss_clip": 1.05446291,
      "balance_loss_mlp": 1.01673186,
      "epoch": 0.5484278242048939,
      "flos": 19317212046720.0,
      "grad_norm": 2.1740044967547933,
      "language_loss": 0.7743023,
      "learning_rate": 1.7838687779857783e-06,
      "loss": 0.79634368,
      "num_input_tokens_seen": 98371130,
      "step": 4561,
      "time_per_iteration": 2.6117215156555176
    },
    {
      "auxiliary_loss_clip": 0.01176454,
      "auxiliary_loss_mlp": 0.01027963,
      "balance_loss_clip": 1.05060327,
      "balance_loss_mlp": 1.01997566,
      "epoch": 0.5485480670955329,
      "flos": 22816024128000.0,
      "grad_norm": 2.300484714789922,
      "language_loss": 0.64255536,
      "learning_rate": 1.7830943845932366e-06,
      "loss": 0.66459954,
      "num_input_tokens_seen": 98390455,
      "step": 4562,
      "time_per_iteration": 2.69650936126709
    },
    {
      "auxiliary_loss_clip": 0.01177509,
      "auxiliary_loss_mlp": 0.01026634,
      "balance_loss_clip": 1.05253482,
      "balance_loss_mlp": 1.01811683,
      "epoch": 0.5486683099861721,
      "flos": 22671304231680.0,
      "grad_norm": 1.8237191266544956,
      "language_loss": 0.75351101,
      "learning_rate": 1.7823200241051044e-06,
      "loss": 0.77555245,
      "num_input_tokens_seen": 98409370,
      "step": 4563,
      "time_per_iteration": 2.610060691833496
    },
    {
      "auxiliary_loss_clip": 0.01180739,
      "auxiliary_loss_mlp": 0.01028369,
      "balance_loss_clip": 1.05375028,
      "balance_loss_mlp": 1.02018225,
      "epoch": 0.5487885528768112,
      "flos": 23149383275520.0,
      "grad_norm": 2.0812896295225563,
      "language_loss": 0.80974448,
      "learning_rate": 1.7815456966388513e-06,
      "loss": 0.83183557,
      "num_input_tokens_seen": 98428465,
      "step": 4564,
      "time_per_iteration": 2.6797664165496826
    },
    {
      "auxiliary_loss_clip": 0.01182246,
      "auxiliary_loss_mlp": 0.01024348,
      "balance_loss_clip": 1.0537703,
      "balance_loss_mlp": 1.01578879,
      "epoch": 0.5489087957674502,
      "flos": 22053928245120.0,
      "grad_norm": 3.3405165347418366,
      "language_loss": 0.80817556,
      "learning_rate": 1.780771402311943e-06,
      "loss": 0.8302415,
      "num_input_tokens_seen": 98447300,
      "step": 4565,
      "time_per_iteration": 2.623054265975952
    },
    {
      "auxiliary_loss_clip": 0.01183742,
      "auxiliary_loss_mlp": 0.01028019,
      "balance_loss_clip": 1.05529118,
      "balance_loss_mlp": 1.01948977,
      "epoch": 0.5490290386580894,
      "flos": 24315977191680.0,
      "grad_norm": 1.9690880727741316,
      "language_loss": 0.78647572,
      "learning_rate": 1.7799971412418374e-06,
      "loss": 0.80859333,
      "num_input_tokens_seen": 98468695,
      "step": 4566,
      "time_per_iteration": 2.63851261138916
    },
    {
      "auxiliary_loss_clip": 0.01184447,
      "auxiliary_loss_mlp": 0.01029866,
      "balance_loss_clip": 1.05527711,
      "balance_loss_mlp": 1.02093172,
      "epoch": 0.5491492815487284,
      "flos": 18294942977280.0,
      "grad_norm": 2.50142144573411,
      "language_loss": 0.73973167,
      "learning_rate": 1.7792229135459918e-06,
      "loss": 0.76187479,
      "num_input_tokens_seen": 98485345,
      "step": 4567,
      "time_per_iteration": 2.6740305423736572
    },
    {
      "auxiliary_loss_clip": 0.01078904,
      "auxiliary_loss_mlp": 0.01001786,
      "balance_loss_clip": 1.01554072,
      "balance_loss_mlp": 1.00051689,
      "epoch": 0.5492695244393675,
      "flos": 64550257050240.0,
      "grad_norm": 0.7371881657798522,
      "language_loss": 0.61617571,
      "learning_rate": 1.7784487193418538e-06,
      "loss": 0.63698262,
      "num_input_tokens_seen": 98543195,
      "step": 4568,
      "time_per_iteration": 3.070449113845825
    },
    {
      "auxiliary_loss_clip": 0.01181852,
      "auxiliary_loss_mlp": 0.01022079,
      "balance_loss_clip": 1.05158067,
      "balance_loss_mlp": 1.01348984,
      "epoch": 0.5493897673300067,
      "flos": 17379579761280.0,
      "grad_norm": 2.012189495944074,
      "language_loss": 0.60607982,
      "learning_rate": 1.7776745587468698e-06,
      "loss": 0.62811911,
      "num_input_tokens_seen": 98560620,
      "step": 4569,
      "time_per_iteration": 2.5645885467529297
    },
    {
      "auxiliary_loss_clip": 0.01178472,
      "auxiliary_loss_mlp": 0.01024622,
      "balance_loss_clip": 1.05020308,
      "balance_loss_mlp": 1.01654589,
      "epoch": 0.5495100102206457,
      "flos": 19901765980800.0,
      "grad_norm": 5.3907675730784135,
      "language_loss": 0.81555688,
      "learning_rate": 1.7769004318784776e-06,
      "loss": 0.83758783,
      "num_input_tokens_seen": 98578265,
      "step": 4570,
      "time_per_iteration": 2.5597989559173584
    },
    {
      "auxiliary_loss_clip": 0.01178726,
      "auxiliary_loss_mlp": 0.01025196,
      "balance_loss_clip": 1.05120802,
      "balance_loss_mlp": 1.0169822,
      "epoch": 0.5496302531112848,
      "flos": 16727190992640.0,
      "grad_norm": 2.9145410910436023,
      "language_loss": 0.805336,
      "learning_rate": 1.776126338854113e-06,
      "loss": 0.82737517,
      "num_input_tokens_seen": 98596055,
      "step": 4571,
      "time_per_iteration": 2.5858702659606934
    },
    {
      "auxiliary_loss_clip": 0.01177098,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.05139017,
      "balance_loss_mlp": 1.01877284,
      "epoch": 0.5497504960019239,
      "flos": 24572343536640.0,
      "grad_norm": 2.7740392100968503,
      "language_loss": 0.84721112,
      "learning_rate": 1.7753522797912044e-06,
      "loss": 0.8692503,
      "num_input_tokens_seen": 98616140,
      "step": 4572,
      "time_per_iteration": 2.624049663543701
    },
    {
      "auxiliary_loss_clip": 0.01183085,
      "auxiliary_loss_mlp": 0.01029565,
      "balance_loss_clip": 1.05377078,
      "balance_loss_mlp": 1.02089286,
      "epoch": 0.549870738892563,
      "flos": 15450494912640.0,
      "grad_norm": 3.678734601306235,
      "language_loss": 0.70203656,
      "learning_rate": 1.7745782548071765e-06,
      "loss": 0.72416306,
      "num_input_tokens_seen": 98633035,
      "step": 4573,
      "time_per_iteration": 3.4874043464660645
    },
    {
      "auxiliary_loss_clip": 0.01179172,
      "auxiliary_loss_mlp": 0.01029017,
      "balance_loss_clip": 1.05245316,
      "balance_loss_mlp": 1.02049947,
      "epoch": 0.549990981783202,
      "flos": 21069114082560.0,
      "grad_norm": 2.4412023399455602,
      "language_loss": 0.74290442,
      "learning_rate": 1.7738042640194482e-06,
      "loss": 0.76498634,
      "num_input_tokens_seen": 98652700,
      "step": 4574,
      "time_per_iteration": 2.6548640727996826
    },
    {
      "auxiliary_loss_clip": 0.01180075,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.05153179,
      "balance_loss_mlp": 1.01896238,
      "epoch": 0.5501112246738411,
      "flos": 21395901041280.0,
      "grad_norm": 1.7726546784829358,
      "language_loss": 0.70655918,
      "learning_rate": 1.7730303075454335e-06,
      "loss": 0.72863829,
      "num_input_tokens_seen": 98671590,
      "step": 4575,
      "time_per_iteration": 2.609867572784424
    },
    {
      "auxiliary_loss_clip": 0.01179728,
      "auxiliary_loss_mlp": 0.01030283,
      "balance_loss_clip": 1.05202174,
      "balance_loss_mlp": 1.02145553,
      "epoch": 0.5502314675644803,
      "flos": 17456931699840.0,
      "grad_norm": 2.663925648304356,
      "language_loss": 0.849612,
      "learning_rate": 1.7722563855025402e-06,
      "loss": 0.87171209,
      "num_input_tokens_seen": 98689620,
      "step": 4576,
      "time_per_iteration": 3.5064401626586914
    },
    {
      "auxiliary_loss_clip": 0.01179902,
      "auxiliary_loss_mlp": 0.01028834,
      "balance_loss_clip": 1.0510236,
      "balance_loss_mlp": 1.02035844,
      "epoch": 0.5503517104551193,
      "flos": 24310410583680.0,
      "grad_norm": 3.1314322027844095,
      "language_loss": 0.71219647,
      "learning_rate": 1.7714824980081721e-06,
      "loss": 0.7342838,
      "num_input_tokens_seen": 98708915,
      "step": 4577,
      "time_per_iteration": 2.6038973331451416
    },
    {
      "auxiliary_loss_clip": 0.0117781,
      "auxiliary_loss_mlp": 0.01025746,
      "balance_loss_clip": 1.05313635,
      "balance_loss_mlp": 1.01757431,
      "epoch": 0.5504719533457584,
      "flos": 22419427086720.0,
      "grad_norm": 1.8373633486432779,
      "language_loss": 0.73606741,
      "learning_rate": 1.7707086451797276e-06,
      "loss": 0.75810301,
      "num_input_tokens_seen": 98729790,
      "step": 4578,
      "time_per_iteration": 2.650418281555176
    },
    {
      "auxiliary_loss_clip": 0.01077695,
      "auxiliary_loss_mlp": 0.01001571,
      "balance_loss_clip": 1.01542044,
      "balance_loss_mlp": 1.00019383,
      "epoch": 0.5505921962363975,
      "flos": 67294155968640.0,
      "grad_norm": 0.7024141218879334,
      "language_loss": 0.52330679,
      "learning_rate": 1.7699348271345993e-06,
      "loss": 0.54409945,
      "num_input_tokens_seen": 98792415,
      "step": 4579,
      "time_per_iteration": 3.9899094104766846
    },
    {
      "auxiliary_loss_clip": 0.01077775,
      "auxiliary_loss_mlp": 0.0100133,
      "balance_loss_clip": 1.01521277,
      "balance_loss_mlp": 0.99993479,
      "epoch": 0.5507124391270366,
      "flos": 45685125578880.0,
      "grad_norm": 0.7122906251583974,
      "language_loss": 0.54426658,
      "learning_rate": 1.7691610439901753e-06,
      "loss": 0.56505764,
      "num_input_tokens_seen": 98855350,
      "step": 4580,
      "time_per_iteration": 4.115381240844727
    },
    {
      "auxiliary_loss_clip": 0.01181948,
      "auxiliary_loss_mlp": 0.01029865,
      "balance_loss_clip": 1.0533309,
      "balance_loss_mlp": 1.02129388,
      "epoch": 0.5508326820176757,
      "flos": 22273845264000.0,
      "grad_norm": 2.426781335076476,
      "language_loss": 0.75429392,
      "learning_rate": 1.7683872958638367e-06,
      "loss": 0.77641207,
      "num_input_tokens_seen": 98874230,
      "step": 4581,
      "time_per_iteration": 2.685415506362915
    },
    {
      "auxiliary_loss_clip": 0.01176156,
      "auxiliary_loss_mlp": 0.01024748,
      "balance_loss_clip": 1.04926085,
      "balance_loss_mlp": 1.01686811,
      "epoch": 0.5509529249083148,
      "flos": 20012442762240.0,
      "grad_norm": 2.1783425727961787,
      "language_loss": 0.8424949,
      "learning_rate": 1.7676135828729614e-06,
      "loss": 0.86450392,
      "num_input_tokens_seen": 98893940,
      "step": 4582,
      "time_per_iteration": 2.6577069759368896
    },
    {
      "auxiliary_loss_clip": 0.01179702,
      "auxiliary_loss_mlp": 0.01023329,
      "balance_loss_clip": 1.05334401,
      "balance_loss_mlp": 1.0151099,
      "epoch": 0.5510731677989539,
      "flos": 21834801325440.0,
      "grad_norm": 2.021680396431873,
      "language_loss": 0.83199036,
      "learning_rate": 1.7668399051349205e-06,
      "loss": 0.85402066,
      "num_input_tokens_seen": 98913620,
      "step": 4583,
      "time_per_iteration": 2.6373538970947266
    },
    {
      "auxiliary_loss_clip": 0.01181226,
      "auxiliary_loss_mlp": 0.01034897,
      "balance_loss_clip": 1.05295002,
      "balance_loss_mlp": 1.0264926,
      "epoch": 0.5511934106895929,
      "flos": 21467901853440.0,
      "grad_norm": 2.2084825247820246,
      "language_loss": 0.83365464,
      "learning_rate": 1.766066262767081e-06,
      "loss": 0.85581583,
      "num_input_tokens_seen": 98931460,
      "step": 4584,
      "time_per_iteration": 2.6551353931427
    },
    {
      "auxiliary_loss_clip": 0.01175756,
      "auxiliary_loss_mlp": 0.01024962,
      "balance_loss_clip": 1.05089378,
      "balance_loss_mlp": 1.01648653,
      "epoch": 0.5513136535802321,
      "flos": 21068934514560.0,
      "grad_norm": 3.451456126107924,
      "language_loss": 0.76915944,
      "learning_rate": 1.765292655886803e-06,
      "loss": 0.79116666,
      "num_input_tokens_seen": 98950105,
      "step": 4585,
      "time_per_iteration": 2.6107661724090576
    },
    {
      "auxiliary_loss_clip": 0.01179513,
      "auxiliary_loss_mlp": 0.01030578,
      "balance_loss_clip": 1.05196404,
      "balance_loss_mlp": 1.02261496,
      "epoch": 0.5514338964708712,
      "flos": 27815004754560.0,
      "grad_norm": 2.016052734104686,
      "language_loss": 0.70674407,
      "learning_rate": 1.764519084611443e-06,
      "loss": 0.728845,
      "num_input_tokens_seen": 98970560,
      "step": 4586,
      "time_per_iteration": 2.683950662612915
    },
    {
      "auxiliary_loss_clip": 0.01181754,
      "auxiliary_loss_mlp": 0.01030242,
      "balance_loss_clip": 1.05151629,
      "balance_loss_mlp": 1.02065718,
      "epoch": 0.5515541393615102,
      "flos": 21908525990400.0,
      "grad_norm": 1.9913179243504868,
      "language_loss": 0.77937353,
      "learning_rate": 1.7637455490583505e-06,
      "loss": 0.80149353,
      "num_input_tokens_seen": 98989885,
      "step": 4587,
      "time_per_iteration": 2.6072161197662354
    },
    {
      "auxiliary_loss_clip": 0.01177859,
      "auxiliary_loss_mlp": 0.01027789,
      "balance_loss_clip": 1.0514524,
      "balance_loss_mlp": 1.01973021,
      "epoch": 0.5516743822521494,
      "flos": 20485422074880.0,
      "grad_norm": 2.0023723341136273,
      "language_loss": 0.77271914,
      "learning_rate": 1.7629720493448701e-06,
      "loss": 0.79477561,
      "num_input_tokens_seen": 99007180,
      "step": 4588,
      "time_per_iteration": 2.6522316932678223
    },
    {
      "auxiliary_loss_clip": 0.01179899,
      "auxiliary_loss_mlp": 0.01027263,
      "balance_loss_clip": 1.05242288,
      "balance_loss_mlp": 1.01934791,
      "epoch": 0.5517946251427884,
      "flos": 14940383915520.0,
      "grad_norm": 1.962928115122331,
      "language_loss": 0.85470474,
      "learning_rate": 1.7621985855883418e-06,
      "loss": 0.87677634,
      "num_input_tokens_seen": 99023880,
      "step": 4589,
      "time_per_iteration": 2.5688726902008057
    },
    {
      "auxiliary_loss_clip": 0.0117639,
      "auxiliary_loss_mlp": 0.01025227,
      "balance_loss_clip": 1.04898262,
      "balance_loss_mlp": 1.01689398,
      "epoch": 0.5519148680334275,
      "flos": 18404865573120.0,
      "grad_norm": 1.9389773543724955,
      "language_loss": 0.72493351,
      "learning_rate": 1.7614251579060983e-06,
      "loss": 0.74694967,
      "num_input_tokens_seen": 99042475,
      "step": 4590,
      "time_per_iteration": 2.6089463233947754
    },
    {
      "auxiliary_loss_clip": 0.01179788,
      "auxiliary_loss_mlp": 0.01022975,
      "balance_loss_clip": 1.05116963,
      "balance_loss_mlp": 1.0149703,
      "epoch": 0.5520351109240667,
      "flos": 25113337251840.0,
      "grad_norm": 2.0073819520704568,
      "language_loss": 0.84781176,
      "learning_rate": 1.76065176641547e-06,
      "loss": 0.86983943,
      "num_input_tokens_seen": 99065185,
      "step": 4591,
      "time_per_iteration": 2.6735711097717285
    },
    {
      "auxiliary_loss_clip": 0.01173947,
      "auxiliary_loss_mlp": 0.01025263,
      "balance_loss_clip": 1.04859447,
      "balance_loss_mlp": 1.0176332,
      "epoch": 0.5521553538147057,
      "flos": 21069545045760.0,
      "grad_norm": 2.7259936439356265,
      "language_loss": 0.78165388,
      "learning_rate": 1.759878411233777e-06,
      "loss": 0.80364597,
      "num_input_tokens_seen": 99083645,
      "step": 4592,
      "time_per_iteration": 2.629815101623535
    },
    {
      "auxiliary_loss_clip": 0.01179918,
      "auxiliary_loss_mlp": 0.01028071,
      "balance_loss_clip": 1.05237782,
      "balance_loss_mlp": 1.01936853,
      "epoch": 0.5522755967053448,
      "flos": 18879999701760.0,
      "grad_norm": 2.409323766547056,
      "language_loss": 0.75681937,
      "learning_rate": 1.7591050924783388e-06,
      "loss": 0.77889931,
      "num_input_tokens_seen": 99100835,
      "step": 4593,
      "time_per_iteration": 2.591099977493286
    },
    {
      "auxiliary_loss_clip": 0.01076355,
      "auxiliary_loss_mlp": 0.01001127,
      "balance_loss_clip": 1.01434946,
      "balance_loss_mlp": 0.99982178,
      "epoch": 0.5523958395959839,
      "flos": 64675622494080.0,
      "grad_norm": 0.8357826793543633,
      "language_loss": 0.57886702,
      "learning_rate": 1.7583318102664661e-06,
      "loss": 0.5996418,
      "num_input_tokens_seen": 99168400,
      "step": 4594,
      "time_per_iteration": 3.235248327255249
    },
    {
      "auxiliary_loss_clip": 0.01175598,
      "auxiliary_loss_mlp": 0.01027089,
      "balance_loss_clip": 1.0477742,
      "balance_loss_mlp": 1.01876783,
      "epoch": 0.552516082486623,
      "flos": 10889732211840.0,
      "grad_norm": 2.172814930300996,
      "language_loss": 0.7919054,
      "learning_rate": 1.757558564715466e-06,
      "loss": 0.81393224,
      "num_input_tokens_seen": 99186475,
      "step": 4595,
      "time_per_iteration": 2.5882580280303955
    },
    {
      "auxiliary_loss_clip": 0.01181302,
      "auxiliary_loss_mlp": 0.01030954,
      "balance_loss_clip": 1.05125546,
      "balance_loss_mlp": 1.02251446,
      "epoch": 0.552636325377262,
      "flos": 22199797376640.0,
      "grad_norm": 4.354390565630429,
      "language_loss": 0.73694491,
      "learning_rate": 1.7567853559426386e-06,
      "loss": 0.75906754,
      "num_input_tokens_seen": 99203525,
      "step": 4596,
      "time_per_iteration": 2.64361572265625
    },
    {
      "auxiliary_loss_clip": 0.01179465,
      "auxiliary_loss_mlp": 0.01025029,
      "balance_loss_clip": 1.05165899,
      "balance_loss_mlp": 1.01677966,
      "epoch": 0.5527565682679012,
      "flos": 23988184652160.0,
      "grad_norm": 3.960546565395502,
      "language_loss": 0.74919069,
      "learning_rate": 1.7560121840652797e-06,
      "loss": 0.77123564,
      "num_input_tokens_seen": 99222910,
      "step": 4597,
      "time_per_iteration": 2.748950958251953
    },
    {
      "auxiliary_loss_clip": 0.01178852,
      "auxiliary_loss_mlp": 0.01027516,
      "balance_loss_clip": 1.05206275,
      "balance_loss_mlp": 1.01941001,
      "epoch": 0.5528768111585403,
      "flos": 19719267955200.0,
      "grad_norm": 2.2731772173138105,
      "language_loss": 0.69087648,
      "learning_rate": 1.7552390492006782e-06,
      "loss": 0.71294022,
      "num_input_tokens_seen": 99241230,
      "step": 4598,
      "time_per_iteration": 2.652763605117798
    },
    {
      "auxiliary_loss_clip": 0.01181828,
      "auxiliary_loss_mlp": 0.01027289,
      "balance_loss_clip": 1.05347133,
      "balance_loss_mlp": 1.01809192,
      "epoch": 0.5529970540491793,
      "flos": 26215975002240.0,
      "grad_norm": 1.745820274123674,
      "language_loss": 0.6426636,
      "learning_rate": 1.7544659514661184e-06,
      "loss": 0.66475475,
      "num_input_tokens_seen": 99264320,
      "step": 4599,
      "time_per_iteration": 3.671363115310669
    },
    {
      "auxiliary_loss_clip": 0.01181741,
      "auxiliary_loss_mlp": 0.01020145,
      "balance_loss_clip": 1.05428147,
      "balance_loss_mlp": 1.01239014,
      "epoch": 0.5531172969398185,
      "flos": 24425971614720.0,
      "grad_norm": 2.385013245250679,
      "language_loss": 0.79916263,
      "learning_rate": 1.7536928909788786e-06,
      "loss": 0.82118148,
      "num_input_tokens_seen": 99283625,
      "step": 4600,
      "time_per_iteration": 2.6317992210388184
    },
    {
      "auxiliary_loss_clip": 0.01075253,
      "auxiliary_loss_mlp": 0.01001493,
      "balance_loss_clip": 1.01323712,
      "balance_loss_mlp": 1.0002358,
      "epoch": 0.5532375398304575,
      "flos": 64907316195840.0,
      "grad_norm": 0.880855344097166,
      "language_loss": 0.61950088,
      "learning_rate": 1.752919867856231e-06,
      "loss": 0.64026833,
      "num_input_tokens_seen": 99335270,
      "step": 4601,
      "time_per_iteration": 3.0259580612182617
    },
    {
      "auxiliary_loss_clip": 0.0117884,
      "auxiliary_loss_mlp": 0.01026452,
      "balance_loss_clip": 1.05119634,
      "balance_loss_mlp": 1.01900482,
      "epoch": 0.5533577827210966,
      "flos": 19683105937920.0,
      "grad_norm": 2.0261092584919824,
      "language_loss": 0.78778517,
      "learning_rate": 1.7521468822154436e-06,
      "loss": 0.80983806,
      "num_input_tokens_seen": 99354185,
      "step": 4602,
      "time_per_iteration": 3.5648980140686035
    },
    {
      "auxiliary_loss_clip": 0.01175577,
      "auxiliary_loss_mlp": 0.0102791,
      "balance_loss_clip": 1.0509795,
      "balance_loss_mlp": 1.01982796,
      "epoch": 0.5534780256117358,
      "flos": 32306496076800.0,
      "grad_norm": 1.8408221598379648,
      "language_loss": 0.75269276,
      "learning_rate": 1.751373934173777e-06,
      "loss": 0.77472758,
      "num_input_tokens_seen": 99376930,
      "step": 4603,
      "time_per_iteration": 2.667177200317383
    },
    {
      "auxiliary_loss_clip": 0.01180815,
      "auxiliary_loss_mlp": 0.01028166,
      "balance_loss_clip": 1.05136359,
      "balance_loss_mlp": 1.02011323,
      "epoch": 0.5535982685023748,
      "flos": 23222425582080.0,
      "grad_norm": 2.3910903311231224,
      "language_loss": 0.73191476,
      "learning_rate": 1.750601023848487e-06,
      "loss": 0.75400454,
      "num_input_tokens_seen": 99397655,
      "step": 4604,
      "time_per_iteration": 2.609220504760742
    },
    {
      "auxiliary_loss_clip": 0.01176943,
      "auxiliary_loss_mlp": 0.01025537,
      "balance_loss_clip": 1.05061078,
      "balance_loss_mlp": 1.01745439,
      "epoch": 0.5537185113930139,
      "flos": 24352534258560.0,
      "grad_norm": 9.339354701821433,
      "language_loss": 0.74090934,
      "learning_rate": 1.749828151356823e-06,
      "loss": 0.76293409,
      "num_input_tokens_seen": 99417850,
      "step": 4605,
      "time_per_iteration": 2.704530954360962
    },
    {
      "auxiliary_loss_clip": 0.01178997,
      "auxiliary_loss_mlp": 0.0102328,
      "balance_loss_clip": 1.05204773,
      "balance_loss_mlp": 1.01547158,
      "epoch": 0.553838754283653,
      "flos": 23549068886400.0,
      "grad_norm": 2.2309486382076975,
      "language_loss": 0.75630987,
      "learning_rate": 1.7490553168160297e-06,
      "loss": 0.77833271,
      "num_input_tokens_seen": 99438920,
      "step": 4606,
      "time_per_iteration": 3.5581045150756836
    },
    {
      "auxiliary_loss_clip": 0.01177793,
      "auxiliary_loss_mlp": 0.01026231,
      "balance_loss_clip": 1.05091035,
      "balance_loss_mlp": 1.01786804,
      "epoch": 0.5539589971742921,
      "flos": 17275044205440.0,
      "grad_norm": 2.5621539800478432,
      "language_loss": 0.76067674,
      "learning_rate": 1.748282520343345e-06,
      "loss": 0.78271699,
      "num_input_tokens_seen": 99457950,
      "step": 4607,
      "time_per_iteration": 3.592829704284668
    },
    {
      "auxiliary_loss_clip": 0.01183107,
      "auxiliary_loss_mlp": 0.01028927,
      "balance_loss_clip": 1.05223894,
      "balance_loss_mlp": 1.02037954,
      "epoch": 0.5540792400649311,
      "flos": 27564169104000.0,
      "grad_norm": 3.0022209171171266,
      "language_loss": 0.7866683,
      "learning_rate": 1.7475097620560023e-06,
      "loss": 0.80878866,
      "num_input_tokens_seen": 99478015,
      "step": 4608,
      "time_per_iteration": 2.6190123558044434
    },
    {
      "auxiliary_loss_clip": 0.01179328,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.05235267,
      "balance_loss_mlp": 1.01542449,
      "epoch": 0.5541994829555702,
      "flos": 23878657105920.0,
      "grad_norm": 1.8069040672669878,
      "language_loss": 0.71187431,
      "learning_rate": 1.746737042071228e-06,
      "loss": 0.73389661,
      "num_input_tokens_seen": 99496520,
      "step": 4609,
      "time_per_iteration": 2.680734157562256
    },
    {
      "auxiliary_loss_clip": 0.01176522,
      "auxiliary_loss_mlp": 0.01023451,
      "balance_loss_clip": 1.05077004,
      "balance_loss_mlp": 1.01565492,
      "epoch": 0.5543197258462094,
      "flos": 20115721342080.0,
      "grad_norm": 2.4460401592797933,
      "language_loss": 0.78725457,
      "learning_rate": 1.7459643605062424e-06,
      "loss": 0.80925423,
      "num_input_tokens_seen": 99513780,
      "step": 4610,
      "time_per_iteration": 2.644242525100708
    },
    {
      "auxiliary_loss_clip": 0.01178095,
      "auxiliary_loss_mlp": 0.01028793,
      "balance_loss_clip": 1.0500989,
      "balance_loss_mlp": 1.02037692,
      "epoch": 0.5544399687368484,
      "flos": 20916565021440.0,
      "grad_norm": 1.6774802755015066,
      "language_loss": 0.81022578,
      "learning_rate": 1.745191717478262e-06,
      "loss": 0.8322947,
      "num_input_tokens_seen": 99532360,
      "step": 4611,
      "time_per_iteration": 2.64969801902771
    },
    {
      "auxiliary_loss_clip": 0.01177239,
      "auxiliary_loss_mlp": 0.0102641,
      "balance_loss_clip": 1.05152488,
      "balance_loss_mlp": 1.01848817,
      "epoch": 0.5545602116274875,
      "flos": 25518661297920.0,
      "grad_norm": 3.5857384966264187,
      "language_loss": 0.79211724,
      "learning_rate": 1.7444191131044948e-06,
      "loss": 0.81415379,
      "num_input_tokens_seen": 99552635,
      "step": 4612,
      "time_per_iteration": 2.626281261444092
    },
    {
      "auxiliary_loss_clip": 0.01179244,
      "auxiliary_loss_mlp": 0.01026833,
      "balance_loss_clip": 1.0526371,
      "balance_loss_mlp": 1.01903701,
      "epoch": 0.5546804545181266,
      "flos": 20995568985600.0,
      "grad_norm": 1.832416337982212,
      "language_loss": 0.73103237,
      "learning_rate": 1.7436465475021456e-06,
      "loss": 0.75309312,
      "num_input_tokens_seen": 99572685,
      "step": 4613,
      "time_per_iteration": 2.64077091217041
    },
    {
      "auxiliary_loss_clip": 0.01177234,
      "auxiliary_loss_mlp": 0.01029058,
      "balance_loss_clip": 1.05117226,
      "balance_loss_mlp": 1.02099919,
      "epoch": 0.5548006974087657,
      "flos": 26833638297600.0,
      "grad_norm": 2.813212939166635,
      "language_loss": 0.7132113,
      "learning_rate": 1.7428740207884111e-06,
      "loss": 0.73527426,
      "num_input_tokens_seen": 99593565,
      "step": 4614,
      "time_per_iteration": 2.6462652683258057
    },
    {
      "auxiliary_loss_clip": 0.01179666,
      "auxiliary_loss_mlp": 0.01025126,
      "balance_loss_clip": 1.05128121,
      "balance_loss_mlp": 1.016945,
      "epoch": 0.5549209402994048,
      "flos": 33656414031360.0,
      "grad_norm": 2.55073612521725,
      "language_loss": 0.60746139,
      "learning_rate": 1.7421015330804833e-06,
      "loss": 0.62950933,
      "num_input_tokens_seen": 99613485,
      "step": 4615,
      "time_per_iteration": 2.699553966522217
    },
    {
      "auxiliary_loss_clip": 0.01181184,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.05355215,
      "balance_loss_mlp": 1.01903737,
      "epoch": 0.5550411831900439,
      "flos": 23769524609280.0,
      "grad_norm": 2.635586039304268,
      "language_loss": 0.7242468,
      "learning_rate": 1.7413290844955475e-06,
      "loss": 0.74633336,
      "num_input_tokens_seen": 99633515,
      "step": 4616,
      "time_per_iteration": 2.5754623413085938
    },
    {
      "auxiliary_loss_clip": 0.01179813,
      "auxiliary_loss_mlp": 0.01038959,
      "balance_loss_clip": 1.05394077,
      "balance_loss_mlp": 1.03097832,
      "epoch": 0.555161426080683,
      "flos": 21651189978240.0,
      "grad_norm": 2.5440652347987025,
      "language_loss": 0.78040636,
      "learning_rate": 1.7405566751507843e-06,
      "loss": 0.80259407,
      "num_input_tokens_seen": 99651560,
      "step": 4617,
      "time_per_iteration": 2.58821439743042
    },
    {
      "auxiliary_loss_clip": 0.01179397,
      "auxiliary_loss_mlp": 0.01028665,
      "balance_loss_clip": 1.05281281,
      "balance_loss_mlp": 1.02024865,
      "epoch": 0.555281668971322,
      "flos": 49563116605440.0,
      "grad_norm": 1.5828188695189354,
      "language_loss": 0.67596149,
      "learning_rate": 1.7397843051633668e-06,
      "loss": 0.69804204,
      "num_input_tokens_seen": 99674255,
      "step": 4618,
      "time_per_iteration": 2.8796937465667725
    },
    {
      "auxiliary_loss_clip": 0.01175242,
      "auxiliary_loss_mlp": 0.01026578,
      "balance_loss_clip": 1.04878223,
      "balance_loss_mlp": 1.01791704,
      "epoch": 0.5554019118619612,
      "flos": 20741608851840.0,
      "grad_norm": 1.7086963858032829,
      "language_loss": 0.71610028,
      "learning_rate": 1.739011974650464e-06,
      "loss": 0.73811853,
      "num_input_tokens_seen": 99693585,
      "step": 4619,
      "time_per_iteration": 2.67661190032959
    },
    {
      "auxiliary_loss_clip": 0.0118125,
      "auxiliary_loss_mlp": 0.01028101,
      "balance_loss_clip": 1.05017948,
      "balance_loss_mlp": 1.01889205,
      "epoch": 0.5555221547526003,
      "flos": 25483217552640.0,
      "grad_norm": 3.2124550231085203,
      "language_loss": 0.76702315,
      "learning_rate": 1.7382396837292365e-06,
      "loss": 0.78911668,
      "num_input_tokens_seen": 99714045,
      "step": 4620,
      "time_per_iteration": 2.664591073989868
    },
    {
      "auxiliary_loss_clip": 0.0118195,
      "auxiliary_loss_mlp": 0.01026376,
      "balance_loss_clip": 1.05405831,
      "balance_loss_mlp": 1.01815677,
      "epoch": 0.5556423976432393,
      "flos": 21762513204480.0,
      "grad_norm": 1.9528956342774912,
      "language_loss": 0.73156101,
      "learning_rate": 1.737467432516841e-06,
      "loss": 0.75364423,
      "num_input_tokens_seen": 99734145,
      "step": 4621,
      "time_per_iteration": 2.623392343521118
    },
    {
      "auxiliary_loss_clip": 0.01180577,
      "auxiliary_loss_mlp": 0.01026479,
      "balance_loss_clip": 1.05288029,
      "balance_loss_mlp": 1.01835465,
      "epoch": 0.5557626405338785,
      "flos": 24900171989760.0,
      "grad_norm": 2.710681637654291,
      "language_loss": 0.74438429,
      "learning_rate": 1.7366952211304274e-06,
      "loss": 0.76645494,
      "num_input_tokens_seen": 99751990,
      "step": 4622,
      "time_per_iteration": 2.6744141578674316
    },
    {
      "auxiliary_loss_clip": 0.01175517,
      "auxiliary_loss_mlp": 0.01027192,
      "balance_loss_clip": 1.04874492,
      "balance_loss_mlp": 1.01974154,
      "epoch": 0.5558828834245175,
      "flos": 18697501676160.0,
      "grad_norm": 3.2595127978401166,
      "language_loss": 0.8333931,
      "learning_rate": 1.735923049687139e-06,
      "loss": 0.85542011,
      "num_input_tokens_seen": 99768565,
      "step": 4623,
      "time_per_iteration": 2.5725135803222656
    },
    {
      "auxiliary_loss_clip": 0.01178127,
      "auxiliary_loss_mlp": 0.0103222,
      "balance_loss_clip": 1.05166626,
      "balance_loss_mlp": 1.02478743,
      "epoch": 0.5560031263151566,
      "flos": 27272179445760.0,
      "grad_norm": 1.685112008473315,
      "language_loss": 0.73853195,
      "learning_rate": 1.7351509183041144e-06,
      "loss": 0.76063544,
      "num_input_tokens_seen": 99788895,
      "step": 4624,
      "time_per_iteration": 2.667473554611206
    },
    {
      "auxiliary_loss_clip": 0.01180939,
      "auxiliary_loss_mlp": 0.01023525,
      "balance_loss_clip": 1.05322492,
      "balance_loss_mlp": 1.01556146,
      "epoch": 0.5561233692057957,
      "flos": 23403738458880.0,
      "grad_norm": 2.4649030674379477,
      "language_loss": 0.71550339,
      "learning_rate": 1.7343788270984852e-06,
      "loss": 0.73754799,
      "num_input_tokens_seen": 99808035,
      "step": 4625,
      "time_per_iteration": 2.625012159347534
    },
    {
      "auxiliary_loss_clip": 0.0117984,
      "auxiliary_loss_mlp": 0.01031887,
      "balance_loss_clip": 1.05339622,
      "balance_loss_mlp": 1.024019,
      "epoch": 0.5562436120964348,
      "flos": 37670867804160.0,
      "grad_norm": 2.1197558688269162,
      "language_loss": 0.74916798,
      "learning_rate": 1.7336067761873764e-06,
      "loss": 0.77128518,
      "num_input_tokens_seen": 99830460,
      "step": 4626,
      "time_per_iteration": 3.7960383892059326
    },
    {
      "auxiliary_loss_clip": 0.01178885,
      "auxiliary_loss_mlp": 0.01030385,
      "balance_loss_clip": 1.04992485,
      "balance_loss_mlp": 1.02184331,
      "epoch": 0.5563638549870739,
      "flos": 25155245445120.0,
      "grad_norm": 4.024917481513719,
      "language_loss": 0.76736522,
      "learning_rate": 1.7328347656879076e-06,
      "loss": 0.78945792,
      "num_input_tokens_seen": 99850320,
      "step": 4627,
      "time_per_iteration": 2.6948678493499756
    },
    {
      "auxiliary_loss_clip": 0.01179489,
      "auxiliary_loss_mlp": 0.01032315,
      "balance_loss_clip": 1.05313182,
      "balance_loss_mlp": 1.0244174,
      "epoch": 0.556484097877713,
      "flos": 13581810783360.0,
      "grad_norm": 2.708604776822569,
      "language_loss": 0.68452907,
      "learning_rate": 1.7320627957171927e-06,
      "loss": 0.70664704,
      "num_input_tokens_seen": 99864980,
      "step": 4628,
      "time_per_iteration": 3.568455457687378
    },
    {
      "auxiliary_loss_clip": 0.01178914,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 1.05276203,
      "balance_loss_mlp": 1.0168817,
      "epoch": 0.5566043407683521,
      "flos": 24681368292480.0,
      "grad_norm": 1.7277377251242525,
      "language_loss": 0.8157267,
      "learning_rate": 1.7312908663923382e-06,
      "loss": 0.83776391,
      "num_input_tokens_seen": 99881155,
      "step": 4629,
      "time_per_iteration": 2.68827748298645
    },
    {
      "auxiliary_loss_clip": 0.01177717,
      "auxiliary_loss_mlp": 0.01026192,
      "balance_loss_clip": 1.05037045,
      "balance_loss_mlp": 1.01782393,
      "epoch": 0.5567245836589911,
      "flos": 20588161950720.0,
      "grad_norm": 2.3862198615783243,
      "language_loss": 0.67255741,
      "learning_rate": 1.7305189778304463e-06,
      "loss": 0.69459647,
      "num_input_tokens_seen": 99899330,
      "step": 4630,
      "time_per_iteration": 2.669309377670288
    },
    {
      "auxiliary_loss_clip": 0.01179013,
      "auxiliary_loss_mlp": 0.01026385,
      "balance_loss_clip": 1.05273533,
      "balance_loss_mlp": 1.01879787,
      "epoch": 0.5568448265496303,
      "flos": 20704189858560.0,
      "grad_norm": 2.0182957065438005,
      "language_loss": 0.79905868,
      "learning_rate": 1.729747130148611e-06,
      "loss": 0.82111269,
      "num_input_tokens_seen": 99918525,
      "step": 4631,
      "time_per_iteration": 2.590848445892334
    },
    {
      "auxiliary_loss_clip": 0.0117882,
      "auxiliary_loss_mlp": 0.010347,
      "balance_loss_clip": 1.05101109,
      "balance_loss_mlp": 1.02579796,
      "epoch": 0.5569650694402694,
      "flos": 25302910256640.0,
      "grad_norm": 2.1710740738561713,
      "language_loss": 0.76838356,
      "learning_rate": 1.7289753234639208e-06,
      "loss": 0.79051876,
      "num_input_tokens_seen": 99937500,
      "step": 4632,
      "time_per_iteration": 3.490478277206421
    },
    {
      "auxiliary_loss_clip": 0.01179904,
      "auxiliary_loss_mlp": 0.01027884,
      "balance_loss_clip": 1.05163026,
      "balance_loss_mlp": 1.01931882,
      "epoch": 0.5570853123309084,
      "flos": 19712623939200.0,
      "grad_norm": 2.1346025185546256,
      "language_loss": 0.76303232,
      "learning_rate": 1.7282035578934592e-06,
      "loss": 0.78511024,
      "num_input_tokens_seen": 99955665,
      "step": 4633,
      "time_per_iteration": 3.489924430847168
    },
    {
      "auxiliary_loss_clip": 0.0117819,
      "auxiliary_loss_mlp": 0.01026022,
      "balance_loss_clip": 1.0518328,
      "balance_loss_mlp": 1.01880956,
      "epoch": 0.5572055552215476,
      "flos": 16108091153280.0,
      "grad_norm": 5.522231974275183,
      "language_loss": 0.78825474,
      "learning_rate": 1.727431833554301e-06,
      "loss": 0.81029689,
      "num_input_tokens_seen": 99974140,
      "step": 4634,
      "time_per_iteration": 2.716066837310791
    },
    {
      "auxiliary_loss_clip": 0.01175694,
      "auxiliary_loss_mlp": 0.01025873,
      "balance_loss_clip": 1.04808009,
      "balance_loss_mlp": 1.01818967,
      "epoch": 0.5573257981121866,
      "flos": 17128815937920.0,
      "grad_norm": 2.066652020846546,
      "language_loss": 0.77408123,
      "learning_rate": 1.7266601505635175e-06,
      "loss": 0.79609692,
      "num_input_tokens_seen": 99991480,
      "step": 4635,
      "time_per_iteration": 2.6073570251464844
    },
    {
      "auxiliary_loss_clip": 0.01178923,
      "auxiliary_loss_mlp": 0.01028833,
      "balance_loss_clip": 1.05289054,
      "balance_loss_mlp": 1.02064323,
      "epoch": 0.5574460410028257,
      "flos": 18807029222400.0,
      "grad_norm": 3.47845435153774,
      "language_loss": 0.75958049,
      "learning_rate": 1.7258885090381717e-06,
      "loss": 0.78165799,
      "num_input_tokens_seen": 100009520,
      "step": 4636,
      "time_per_iteration": 2.582484483718872
    },
    {
      "auxiliary_loss_clip": 0.01173114,
      "auxiliary_loss_mlp": 0.01028037,
      "balance_loss_clip": 1.04838288,
      "balance_loss_mlp": 1.02072096,
      "epoch": 0.5575662838934649,
      "flos": 29642678530560.0,
      "grad_norm": 2.33786174145464,
      "language_loss": 0.78879446,
      "learning_rate": 1.7251169090953213e-06,
      "loss": 0.81080598,
      "num_input_tokens_seen": 100029995,
      "step": 4637,
      "time_per_iteration": 2.6052591800689697
    },
    {
      "auxiliary_loss_clip": 0.01177428,
      "auxiliary_loss_mlp": 0.01024251,
      "balance_loss_clip": 1.04897499,
      "balance_loss_mlp": 1.01585317,
      "epoch": 0.5576865267841039,
      "flos": 22054466949120.0,
      "grad_norm": 2.591291826425808,
      "language_loss": 0.77138793,
      "learning_rate": 1.7243453508520168e-06,
      "loss": 0.7934047,
      "num_input_tokens_seen": 100046980,
      "step": 4638,
      "time_per_iteration": 2.603048324584961
    },
    {
      "auxiliary_loss_clip": 0.01180023,
      "auxiliary_loss_mlp": 0.01028693,
      "balance_loss_clip": 1.05140042,
      "balance_loss_mlp": 1.02025914,
      "epoch": 0.557806769674743,
      "flos": 17196040241280.0,
      "grad_norm": 2.470087021911036,
      "language_loss": 0.84697872,
      "learning_rate": 1.7235738344253038e-06,
      "loss": 0.86906588,
      "num_input_tokens_seen": 100060610,
      "step": 4639,
      "time_per_iteration": 2.5163145065307617
    },
    {
      "auxiliary_loss_clip": 0.01177544,
      "auxiliary_loss_mlp": 0.01025768,
      "balance_loss_clip": 1.0509752,
      "balance_loss_mlp": 1.01767397,
      "epoch": 0.557927012565382,
      "flos": 24712717887360.0,
      "grad_norm": 1.9261949828234521,
      "language_loss": 0.82740057,
      "learning_rate": 1.72280235993222e-06,
      "loss": 0.84943372,
      "num_input_tokens_seen": 100078915,
      "step": 4640,
      "time_per_iteration": 2.6930556297302246
    },
    {
      "auxiliary_loss_clip": 0.01175337,
      "auxiliary_loss_mlp": 0.01029275,
      "balance_loss_clip": 1.04902852,
      "balance_loss_mlp": 1.02065063,
      "epoch": 0.5580472554560212,
      "flos": 16983090460800.0,
      "grad_norm": 2.504941406843802,
      "language_loss": 0.69270098,
      "learning_rate": 1.722030927489798e-06,
      "loss": 0.71474707,
      "num_input_tokens_seen": 100096195,
      "step": 4641,
      "time_per_iteration": 2.623795986175537
    },
    {
      "auxiliary_loss_clip": 0.01177416,
      "auxiliary_loss_mlp": 0.01027806,
      "balance_loss_clip": 1.0507071,
      "balance_loss_mlp": 1.01976275,
      "epoch": 0.5581674983466602,
      "flos": 23509100027520.0,
      "grad_norm": 13.762675736330271,
      "language_loss": 0.74517334,
      "learning_rate": 1.7212595372150634e-06,
      "loss": 0.7672255,
      "num_input_tokens_seen": 100116175,
      "step": 4642,
      "time_per_iteration": 2.640151262283325
    },
    {
      "auxiliary_loss_clip": 0.01177552,
      "auxiliary_loss_mlp": 0.01020858,
      "balance_loss_clip": 1.05087113,
      "balance_loss_mlp": 1.01292491,
      "epoch": 0.5582877412372993,
      "flos": 13480291969920.0,
      "grad_norm": 2.10874144523756,
      "language_loss": 0.72425771,
      "learning_rate": 1.720488189225035e-06,
      "loss": 0.74624181,
      "num_input_tokens_seen": 100133875,
      "step": 4643,
      "time_per_iteration": 2.6250832080841064
    },
    {
      "auxiliary_loss_clip": 0.0117995,
      "auxiliary_loss_mlp": 0.01024591,
      "balance_loss_clip": 1.05199838,
      "balance_loss_mlp": 1.01631761,
      "epoch": 0.5584079841279385,
      "flos": 21903605827200.0,
      "grad_norm": 2.7149287309698975,
      "language_loss": 0.79206759,
      "learning_rate": 1.7197168836367265e-06,
      "loss": 0.81411302,
      "num_input_tokens_seen": 100150685,
      "step": 4644,
      "time_per_iteration": 2.5996594429016113
    },
    {
      "auxiliary_loss_clip": 0.01174416,
      "auxiliary_loss_mlp": 0.01023486,
      "balance_loss_clip": 1.0492295,
      "balance_loss_mlp": 1.01544487,
      "epoch": 0.5585282270185775,
      "flos": 18843550375680.0,
      "grad_norm": 3.662192397070406,
      "language_loss": 0.81582403,
      "learning_rate": 1.7189456205671433e-06,
      "loss": 0.83780313,
      "num_input_tokens_seen": 100169530,
      "step": 4645,
      "time_per_iteration": 2.6495187282562256
    },
    {
      "auxiliary_loss_clip": 0.01183427,
      "auxiliary_loss_mlp": 0.01028505,
      "balance_loss_clip": 1.05421698,
      "balance_loss_mlp": 1.02007079,
      "epoch": 0.5586484699092166,
      "flos": 21868449390720.0,
      "grad_norm": 2.0594978714577468,
      "language_loss": 0.82335812,
      "learning_rate": 1.7181744001332866e-06,
      "loss": 0.84547746,
      "num_input_tokens_seen": 100188140,
      "step": 4646,
      "time_per_iteration": 2.6009321212768555
    },
    {
      "auxiliary_loss_clip": 0.01174901,
      "auxiliary_loss_mlp": 0.01024469,
      "balance_loss_clip": 1.05107248,
      "balance_loss_mlp": 1.01663125,
      "epoch": 0.5587687127998557,
      "flos": 22893232412160.0,
      "grad_norm": 2.3670995637847843,
      "language_loss": 0.63920712,
      "learning_rate": 1.7174032224521493e-06,
      "loss": 0.66120082,
      "num_input_tokens_seen": 100206850,
      "step": 4647,
      "time_per_iteration": 2.7001001834869385
    },
    {
      "auxiliary_loss_clip": 0.01173827,
      "auxiliary_loss_mlp": 0.01026221,
      "balance_loss_clip": 1.04912186,
      "balance_loss_mlp": 1.01818323,
      "epoch": 0.5588889556904948,
      "flos": 20303067703680.0,
      "grad_norm": 1.844402676687923,
      "language_loss": 0.69865507,
      "learning_rate": 1.7166320876407184e-06,
      "loss": 0.72065556,
      "num_input_tokens_seen": 100226270,
      "step": 4648,
      "time_per_iteration": 2.6027567386627197
    },
    {
      "auxiliary_loss_clip": 0.01181818,
      "auxiliary_loss_mlp": 0.01030547,
      "balance_loss_clip": 1.05345583,
      "balance_loss_mlp": 1.02304268,
      "epoch": 0.5590091985811338,
      "flos": 16472153450880.0,
      "grad_norm": 2.1094519539544603,
      "language_loss": 0.67943245,
      "learning_rate": 1.7158609958159742e-06,
      "loss": 0.70155609,
      "num_input_tokens_seen": 100243675,
      "step": 4649,
      "time_per_iteration": 2.66231369972229
    },
    {
      "auxiliary_loss_clip": 0.01183559,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.0547533,
      "balance_loss_mlp": 1.01840496,
      "epoch": 0.559129441471773,
      "flos": 14532186781440.0,
      "grad_norm": 1.9572803476878964,
      "language_loss": 0.77962589,
      "learning_rate": 1.7150899470948911e-06,
      "loss": 0.80173218,
      "num_input_tokens_seen": 100258940,
      "step": 4650,
      "time_per_iteration": 2.587399959564209
    },
    {
      "auxiliary_loss_clip": 0.01088753,
      "auxiliary_loss_mlp": 0.01008786,
      "balance_loss_clip": 1.02269077,
      "balance_loss_mlp": 1.00745094,
      "epoch": 0.5592496843624121,
      "flos": 60521009852160.0,
      "grad_norm": 0.8053309295681225,
      "language_loss": 0.56575054,
      "learning_rate": 1.7143189415944365e-06,
      "loss": 0.58672589,
      "num_input_tokens_seen": 100323400,
      "step": 4651,
      "time_per_iteration": 3.2182130813598633
    },
    {
      "auxiliary_loss_clip": 0.01177996,
      "auxiliary_loss_mlp": 0.01023989,
      "balance_loss_clip": 1.05202937,
      "balance_loss_mlp": 1.01612091,
      "epoch": 0.5593699272530511,
      "flos": 20886256920960.0,
      "grad_norm": 2.670489499434219,
      "language_loss": 0.76313812,
      "learning_rate": 1.7135479794315714e-06,
      "loss": 0.78515792,
      "num_input_tokens_seen": 100340355,
      "step": 4652,
      "time_per_iteration": 2.605419397354126
    },
    {
      "auxiliary_loss_clip": 0.01176006,
      "auxiliary_loss_mlp": 0.01023709,
      "balance_loss_clip": 1.05022979,
      "balance_loss_mlp": 1.01646757,
      "epoch": 0.5594901701436903,
      "flos": 12896743616640.0,
      "grad_norm": 2.068712790577787,
      "language_loss": 0.78965884,
      "learning_rate": 1.7127770607232502e-06,
      "loss": 0.811656,
      "num_input_tokens_seen": 100358900,
      "step": 4653,
      "time_per_iteration": 3.570021152496338
    },
    {
      "auxiliary_loss_clip": 0.01175552,
      "auxiliary_loss_mlp": 0.01029188,
      "balance_loss_clip": 1.05106378,
      "balance_loss_mlp": 1.02095699,
      "epoch": 0.5596104130343293,
      "flos": 23112107936640.0,
      "grad_norm": 3.1862004714104635,
      "language_loss": 0.79938644,
      "learning_rate": 1.7120061855864204e-06,
      "loss": 0.82143384,
      "num_input_tokens_seen": 100378910,
      "step": 4654,
      "time_per_iteration": 3.5546364784240723
    },
    {
      "auxiliary_loss_clip": 0.0118074,
      "auxiliary_loss_mlp": 0.01029345,
      "balance_loss_clip": 1.05478895,
      "balance_loss_mlp": 1.02126884,
      "epoch": 0.5597306559249684,
      "flos": 25957812977280.0,
      "grad_norm": 2.364078838767678,
      "language_loss": 0.7144832,
      "learning_rate": 1.7112353541380233e-06,
      "loss": 0.73658407,
      "num_input_tokens_seen": 100398770,
      "step": 4655,
      "time_per_iteration": 2.642577648162842
    },
    {
      "auxiliary_loss_clip": 0.01178989,
      "auxiliary_loss_mlp": 0.0103091,
      "balance_loss_clip": 1.05474973,
      "balance_loss_mlp": 1.02261877,
      "epoch": 0.5598508988156076,
      "flos": 22492289825280.0,
      "grad_norm": 1.842665451730733,
      "language_loss": 0.72042257,
      "learning_rate": 1.7104645664949931e-06,
      "loss": 0.74252152,
      "num_input_tokens_seen": 100421240,
      "step": 4656,
      "time_per_iteration": 2.6467936038970947
    },
    {
      "auxiliary_loss_clip": 0.01176223,
      "auxiliary_loss_mlp": 0.01024421,
      "balance_loss_clip": 1.04944873,
      "balance_loss_mlp": 1.01721787,
      "epoch": 0.5599711417062466,
      "flos": 23112538899840.0,
      "grad_norm": 2.774235063254046,
      "language_loss": 0.71624637,
      "learning_rate": 1.7096938227742584e-06,
      "loss": 0.73825276,
      "num_input_tokens_seen": 100442370,
      "step": 4657,
      "time_per_iteration": 2.6834073066711426
    },
    {
      "auxiliary_loss_clip": 0.01177666,
      "auxiliary_loss_mlp": 0.01024984,
      "balance_loss_clip": 1.05253959,
      "balance_loss_mlp": 1.01714611,
      "epoch": 0.5600913845968857,
      "flos": 22339345714560.0,
      "grad_norm": 2.441887289425429,
      "language_loss": 0.83923095,
      "learning_rate": 1.70892312309274e-06,
      "loss": 0.86125743,
      "num_input_tokens_seen": 100460260,
      "step": 4658,
      "time_per_iteration": 2.6512677669525146
    },
    {
      "auxiliary_loss_clip": 0.0117728,
      "auxiliary_loss_mlp": 0.01024449,
      "balance_loss_clip": 1.04917359,
      "balance_loss_mlp": 1.01660514,
      "epoch": 0.5602116274875248,
      "flos": 17633791290240.0,
      "grad_norm": 2.1179025307572497,
      "language_loss": 0.67922956,
      "learning_rate": 1.7081524675673523e-06,
      "loss": 0.70124686,
      "num_input_tokens_seen": 100475750,
      "step": 4659,
      "time_per_iteration": 3.4260432720184326
    },
    {
      "auxiliary_loss_clip": 0.01083442,
      "auxiliary_loss_mlp": 0.01001767,
      "balance_loss_clip": 1.01940012,
      "balance_loss_mlp": 1.00059271,
      "epoch": 0.5603318703781639,
      "flos": 70115945529600.0,
      "grad_norm": 0.7676900530196901,
      "language_loss": 0.59625721,
      "learning_rate": 1.7073818563150026e-06,
      "loss": 0.6171093,
      "num_input_tokens_seen": 100537830,
      "step": 4660,
      "time_per_iteration": 4.1349523067474365
    },
    {
      "auxiliary_loss_clip": 0.01178852,
      "auxiliary_loss_mlp": 0.01029877,
      "balance_loss_clip": 1.05196333,
      "balance_loss_mlp": 1.02173543,
      "epoch": 0.560452113268803,
      "flos": 18545850455040.0,
      "grad_norm": 2.4676765886671403,
      "language_loss": 0.86118805,
      "learning_rate": 1.7066112894525935e-06,
      "loss": 0.88327533,
      "num_input_tokens_seen": 100555910,
      "step": 4661,
      "time_per_iteration": 2.619441032409668
    },
    {
      "auxiliary_loss_clip": 0.01176077,
      "auxiliary_loss_mlp": 0.01029995,
      "balance_loss_clip": 1.05074716,
      "balance_loss_mlp": 1.02159119,
      "epoch": 0.5605723561594421,
      "flos": 25264665250560.0,
      "grad_norm": 2.169747242312628,
      "language_loss": 0.72442389,
      "learning_rate": 1.7058407670970177e-06,
      "loss": 0.74648452,
      "num_input_tokens_seen": 100577385,
      "step": 4662,
      "time_per_iteration": 2.6397430896759033
    },
    {
      "auxiliary_loss_clip": 0.01178652,
      "auxiliary_loss_mlp": 0.01024607,
      "balance_loss_clip": 1.05228031,
      "balance_loss_mlp": 1.01706672,
      "epoch": 0.5606925990500812,
      "flos": 20594949621120.0,
      "grad_norm": 2.008243638909978,
      "language_loss": 0.61495245,
      "learning_rate": 1.7050702893651643e-06,
      "loss": 0.63698506,
      "num_input_tokens_seen": 100596965,
      "step": 4663,
      "time_per_iteration": 2.6456098556518555
    },
    {
      "auxiliary_loss_clip": 0.01179078,
      "auxiliary_loss_mlp": 0.01026097,
      "balance_loss_clip": 1.05232263,
      "balance_loss_mlp": 1.01760364,
      "epoch": 0.5608128419407202,
      "flos": 35006044677120.0,
      "grad_norm": 2.3823170719991924,
      "language_loss": 0.75899351,
      "learning_rate": 1.7042998563739134e-06,
      "loss": 0.78104526,
      "num_input_tokens_seen": 100615315,
      "step": 4664,
      "time_per_iteration": 2.6973016262054443
    },
    {
      "auxiliary_loss_clip": 0.0118344,
      "auxiliary_loss_mlp": 0.01033551,
      "balance_loss_clip": 1.05287087,
      "balance_loss_mlp": 1.0248189,
      "epoch": 0.5609330848313594,
      "flos": 24639819235200.0,
      "grad_norm": 3.2127128055400727,
      "language_loss": 0.71667975,
      "learning_rate": 1.703529468240139e-06,
      "loss": 0.73884964,
      "num_input_tokens_seen": 100634185,
      "step": 4665,
      "time_per_iteration": 2.5867676734924316
    },
    {
      "auxiliary_loss_clip": 0.0117793,
      "auxiliary_loss_mlp": 0.01022959,
      "balance_loss_clip": 1.05182648,
      "balance_loss_mlp": 1.01518607,
      "epoch": 0.5610533277219985,
      "flos": 18762894385920.0,
      "grad_norm": 2.5432082481791403,
      "language_loss": 0.73771346,
      "learning_rate": 1.7027591250807088e-06,
      "loss": 0.75972235,
      "num_input_tokens_seen": 100651360,
      "step": 4666,
      "time_per_iteration": 2.6043848991394043
    },
    {
      "auxiliary_loss_clip": 0.01181495,
      "auxiliary_loss_mlp": 0.01034241,
      "balance_loss_clip": 1.05453324,
      "balance_loss_mlp": 1.02644444,
      "epoch": 0.5611735706126375,
      "flos": 15012384727680.0,
      "grad_norm": 3.3133507910378706,
      "language_loss": 0.84467733,
      "learning_rate": 1.7019888270124825e-06,
      "loss": 0.86683464,
      "num_input_tokens_seen": 100668525,
      "step": 4667,
      "time_per_iteration": 2.564495801925659
    },
    {
      "auxiliary_loss_clip": 0.01180433,
      "auxiliary_loss_mlp": 0.01031224,
      "balance_loss_clip": 1.05563176,
      "balance_loss_mlp": 1.02357078,
      "epoch": 0.5612938135032767,
      "flos": 16468167041280.0,
      "grad_norm": 2.2995249921818797,
      "language_loss": 0.81738216,
      "learning_rate": 1.7012185741523147e-06,
      "loss": 0.83949876,
      "num_input_tokens_seen": 100684850,
      "step": 4668,
      "time_per_iteration": 2.5840933322906494
    },
    {
      "auxiliary_loss_clip": 0.01182025,
      "auxiliary_loss_mlp": 0.01027159,
      "balance_loss_clip": 1.05596495,
      "balance_loss_mlp": 1.01907682,
      "epoch": 0.5614140563939157,
      "flos": 25666433850240.0,
      "grad_norm": 2.971870006711208,
      "language_loss": 0.6267013,
      "learning_rate": 1.7004483666170514e-06,
      "loss": 0.6487931,
      "num_input_tokens_seen": 100705345,
      "step": 4669,
      "time_per_iteration": 2.6098010540008545
    },
    {
      "auxiliary_loss_clip": 0.01176487,
      "auxiliary_loss_mlp": 0.01023035,
      "balance_loss_clip": 1.05124438,
      "balance_loss_mlp": 1.01561403,
      "epoch": 0.5615342992845548,
      "flos": 24717566223360.0,
      "grad_norm": 2.4117788714086945,
      "language_loss": 0.80892599,
      "learning_rate": 1.699678204523533e-06,
      "loss": 0.83092117,
      "num_input_tokens_seen": 100725210,
      "step": 4670,
      "time_per_iteration": 2.651094436645508
    },
    {
      "auxiliary_loss_clip": 0.01182419,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 1.05448413,
      "balance_loss_mlp": 1.01957202,
      "epoch": 0.5616545421751938,
      "flos": 22015934634240.0,
      "grad_norm": 2.531721314260555,
      "language_loss": 0.68431568,
      "learning_rate": 1.6989080879885918e-06,
      "loss": 0.70642573,
      "num_input_tokens_seen": 100743070,
      "step": 4671,
      "time_per_iteration": 2.552835464477539
    },
    {
      "auxiliary_loss_clip": 0.01079366,
      "auxiliary_loss_mlp": 0.01002082,
      "balance_loss_clip": 1.01638234,
      "balance_loss_mlp": 1.00103307,
      "epoch": 0.561774785065833,
      "flos": 53760358690560.0,
      "grad_norm": 0.904950183442447,
      "language_loss": 0.60945266,
      "learning_rate": 1.6981380171290544e-06,
      "loss": 0.63026714,
      "num_input_tokens_seen": 100804095,
      "step": 4672,
      "time_per_iteration": 3.219851016998291
    },
    {
      "auxiliary_loss_clip": 0.0117863,
      "auxiliary_loss_mlp": 0.01026,
      "balance_loss_clip": 1.05092645,
      "balance_loss_mlp": 1.01803088,
      "epoch": 0.5618950279564721,
      "flos": 19750007018880.0,
      "grad_norm": 1.9994038610191291,
      "language_loss": 0.74636221,
      "learning_rate": 1.6973679920617396e-06,
      "loss": 0.76840854,
      "num_input_tokens_seen": 100821630,
      "step": 4673,
      "time_per_iteration": 2.5770909786224365
    },
    {
      "auxiliary_loss_clip": 0.01178867,
      "auxiliary_loss_mlp": 0.01030113,
      "balance_loss_clip": 1.05338788,
      "balance_loss_mlp": 1.02190542,
      "epoch": 0.5620152708471111,
      "flos": 16800592435200.0,
      "grad_norm": 2.0982386204089467,
      "language_loss": 0.85411125,
      "learning_rate": 1.6965980129034603e-06,
      "loss": 0.87620103,
      "num_input_tokens_seen": 100839015,
      "step": 4674,
      "time_per_iteration": 2.6134886741638184
    },
    {
      "auxiliary_loss_clip": 0.01178358,
      "auxiliary_loss_mlp": 0.01028593,
      "balance_loss_clip": 1.05313897,
      "balance_loss_mlp": 1.02006078,
      "epoch": 0.5621355137377503,
      "flos": 26797799502720.0,
      "grad_norm": 1.6243320486423545,
      "language_loss": 0.76632357,
      "learning_rate": 1.6958280797710209e-06,
      "loss": 0.78839308,
      "num_input_tokens_seen": 100860940,
      "step": 4675,
      "time_per_iteration": 2.6745455265045166
    },
    {
      "auxiliary_loss_clip": 0.01077893,
      "auxiliary_loss_mlp": 0.01000648,
      "balance_loss_clip": 1.01526451,
      "balance_loss_mlp": 0.99961728,
      "epoch": 0.5622557566283893,
      "flos": 61207046686080.0,
      "grad_norm": 0.7304043910488193,
      "language_loss": 0.54709923,
      "learning_rate": 1.6950581927812198e-06,
      "loss": 0.56788468,
      "num_input_tokens_seen": 100920510,
      "step": 4676,
      "time_per_iteration": 3.0546886920928955
    },
    {
      "auxiliary_loss_clip": 0.01176054,
      "auxiliary_loss_mlp": 0.01024483,
      "balance_loss_clip": 1.05047321,
      "balance_loss_mlp": 1.01683605,
      "epoch": 0.5623759995190284,
      "flos": 26468534505600.0,
      "grad_norm": 2.1636041610710457,
      "language_loss": 0.7893545,
      "learning_rate": 1.6942883520508486e-06,
      "loss": 0.81135988,
      "num_input_tokens_seen": 100939245,
      "step": 4677,
      "time_per_iteration": 2.6837313175201416
    },
    {
      "auxiliary_loss_clip": 0.01174888,
      "auxiliary_loss_mlp": 0.01023493,
      "balance_loss_clip": 1.05172873,
      "balance_loss_mlp": 1.0158664,
      "epoch": 0.5624962424096676,
      "flos": 19390900798080.0,
      "grad_norm": 3.4136201361999703,
      "language_loss": 0.77176833,
      "learning_rate": 1.693518557696691e-06,
      "loss": 0.79375207,
      "num_input_tokens_seen": 100958385,
      "step": 4678,
      "time_per_iteration": 2.5860414505004883
    },
    {
      "auxiliary_loss_clip": 0.01175882,
      "auxiliary_loss_mlp": 0.01029309,
      "balance_loss_clip": 1.05078852,
      "balance_loss_mlp": 1.02104759,
      "epoch": 0.5626164853003066,
      "flos": 20667345482880.0,
      "grad_norm": 2.491759499736943,
      "language_loss": 0.88703191,
      "learning_rate": 1.6927488098355252e-06,
      "loss": 0.90908378,
      "num_input_tokens_seen": 100976015,
      "step": 4679,
      "time_per_iteration": 3.5807390213012695
    },
    {
      "auxiliary_loss_clip": 0.01076218,
      "auxiliary_loss_mlp": 0.01000944,
      "balance_loss_clip": 1.01406121,
      "balance_loss_mlp": 0.99991286,
      "epoch": 0.5627367281909457,
      "flos": 62766071665920.0,
      "grad_norm": 0.9094462063118469,
      "language_loss": 0.63138032,
      "learning_rate": 1.6919791085841201e-06,
      "loss": 0.65215194,
      "num_input_tokens_seen": 101033425,
      "step": 4680,
      "time_per_iteration": 4.039405584335327
    },
    {
      "auxiliary_loss_clip": 0.01177371,
      "auxiliary_loss_mlp": 0.01024727,
      "balance_loss_clip": 1.04953003,
      "balance_loss_mlp": 1.01631129,
      "epoch": 0.5628569710815848,
      "flos": 12787144243200.0,
      "grad_norm": 2.3295286362495955,
      "language_loss": 0.78884816,
      "learning_rate": 1.6912094540592396e-06,
      "loss": 0.81086916,
      "num_input_tokens_seen": 101048945,
      "step": 4681,
      "time_per_iteration": 2.551227569580078
    },
    {
      "auxiliary_loss_clip": 0.01178752,
      "auxiliary_loss_mlp": 0.01029642,
      "balance_loss_clip": 1.05106509,
      "balance_loss_mlp": 1.02201259,
      "epoch": 0.5629772139722239,
      "flos": 13762082165760.0,
      "grad_norm": 4.33387915358743,
      "language_loss": 0.81096834,
      "learning_rate": 1.6904398463776393e-06,
      "loss": 0.83305228,
      "num_input_tokens_seen": 101062745,
      "step": 4682,
      "time_per_iteration": 2.616177558898926
    },
    {
      "auxiliary_loss_clip": 0.01178415,
      "auxiliary_loss_mlp": 0.01023008,
      "balance_loss_clip": 1.05244756,
      "balance_loss_mlp": 1.01474142,
      "epoch": 0.5630974568628629,
      "flos": 21467830026240.0,
      "grad_norm": 1.8729123335365725,
      "language_loss": 0.72733462,
      "learning_rate": 1.6896702856560683e-06,
      "loss": 0.74934888,
      "num_input_tokens_seen": 101081840,
      "step": 4683,
      "time_per_iteration": 2.6438169479370117
    },
    {
      "auxiliary_loss_clip": 0.01177063,
      "auxiliary_loss_mlp": 0.01026272,
      "balance_loss_clip": 1.05049777,
      "balance_loss_mlp": 1.01859486,
      "epoch": 0.5632176997535021,
      "flos": 14245907385600.0,
      "grad_norm": 3.192499431589036,
      "language_loss": 0.69790572,
      "learning_rate": 1.6889007720112677e-06,
      "loss": 0.71993911,
      "num_input_tokens_seen": 101099585,
      "step": 4684,
      "time_per_iteration": 2.6143624782562256
    },
    {
      "auxiliary_loss_clip": 0.01179637,
      "auxiliary_loss_mlp": 0.01025979,
      "balance_loss_clip": 1.05309296,
      "balance_loss_mlp": 1.01813495,
      "epoch": 0.5633379426441412,
      "flos": 20812244947200.0,
      "grad_norm": 2.546445951102888,
      "language_loss": 0.77466702,
      "learning_rate": 1.6881313055599734e-06,
      "loss": 0.79672319,
      "num_input_tokens_seen": 101119515,
      "step": 4685,
      "time_per_iteration": 3.5438570976257324
    },
    {
      "auxiliary_loss_clip": 0.01181847,
      "auxiliary_loss_mlp": 0.01024951,
      "balance_loss_clip": 1.05295277,
      "balance_loss_mlp": 1.01681197,
      "epoch": 0.5634581855347802,
      "flos": 22600883617920.0,
      "grad_norm": 2.629409670337191,
      "language_loss": 0.82228422,
      "learning_rate": 1.6873618864189117e-06,
      "loss": 0.84435225,
      "num_input_tokens_seen": 101135285,
      "step": 4686,
      "time_per_iteration": 2.716526746749878
    },
    {
      "auxiliary_loss_clip": 0.01179042,
      "auxiliary_loss_mlp": 0.01028619,
      "balance_loss_clip": 1.04993582,
      "balance_loss_mlp": 1.0200119,
      "epoch": 0.5635784284254194,
      "flos": 21506972872320.0,
      "grad_norm": 2.313575802537954,
      "language_loss": 0.77629972,
      "learning_rate": 1.686592514704803e-06,
      "loss": 0.79837638,
      "num_input_tokens_seen": 101152680,
      "step": 4687,
      "time_per_iteration": 3.6188807487487793
    },
    {
      "auxiliary_loss_clip": 0.0117728,
      "auxiliary_loss_mlp": 0.01027977,
      "balance_loss_clip": 1.0535183,
      "balance_loss_mlp": 1.02015674,
      "epoch": 0.5636986713160584,
      "flos": 19827466698240.0,
      "grad_norm": 2.3342022600107946,
      "language_loss": 0.70321012,
      "learning_rate": 1.685823190534361e-06,
      "loss": 0.72526264,
      "num_input_tokens_seen": 101170920,
      "step": 4688,
      "time_per_iteration": 2.5735161304473877
    },
    {
      "auxiliary_loss_clip": 0.01180765,
      "auxiliary_loss_mlp": 0.01024631,
      "balance_loss_clip": 1.05236101,
      "balance_loss_mlp": 1.01641178,
      "epoch": 0.5638189142066975,
      "flos": 19792453916160.0,
      "grad_norm": 2.552488156083785,
      "language_loss": 0.84227753,
      "learning_rate": 1.6850539140242907e-06,
      "loss": 0.86433148,
      "num_input_tokens_seen": 101190180,
      "step": 4689,
      "time_per_iteration": 2.6028201580047607
    },
    {
      "auxiliary_loss_clip": 0.01175547,
      "auxiliary_loss_mlp": 0.0102407,
      "balance_loss_clip": 1.04990149,
      "balance_loss_mlp": 1.0162859,
      "epoch": 0.5639391570973367,
      "flos": 22893771116160.0,
      "grad_norm": 1.7972720552196988,
      "language_loss": 0.82165414,
      "learning_rate": 1.684284685291292e-06,
      "loss": 0.84365028,
      "num_input_tokens_seen": 101211825,
      "step": 4690,
      "time_per_iteration": 2.6123156547546387
    },
    {
      "auxiliary_loss_clip": 0.01179049,
      "auxiliary_loss_mlp": 0.01029772,
      "balance_loss_clip": 1.05197966,
      "balance_loss_mlp": 1.02190435,
      "epoch": 0.5640593999879757,
      "flos": 23727077712000.0,
      "grad_norm": 4.842142237836239,
      "language_loss": 0.81234878,
      "learning_rate": 1.683515504452055e-06,
      "loss": 0.83443701,
      "num_input_tokens_seen": 101229200,
      "step": 4691,
      "time_per_iteration": 2.6163110733032227
    },
    {
      "auxiliary_loss_clip": 0.01180453,
      "auxiliary_loss_mlp": 0.01026464,
      "balance_loss_clip": 1.05119979,
      "balance_loss_mlp": 1.01864362,
      "epoch": 0.5641796428786148,
      "flos": 22710123855360.0,
      "grad_norm": 1.6230346986574995,
      "language_loss": 0.6659745,
      "learning_rate": 1.6827463716232648e-06,
      "loss": 0.68804371,
      "num_input_tokens_seen": 101249860,
      "step": 4692,
      "time_per_iteration": 2.6470084190368652
    },
    {
      "auxiliary_loss_clip": 0.01179876,
      "auxiliary_loss_mlp": 0.01026637,
      "balance_loss_clip": 1.05292821,
      "balance_loss_mlp": 1.01868606,
      "epoch": 0.5642998857692539,
      "flos": 19791987039360.0,
      "grad_norm": 1.8204722997476144,
      "language_loss": 0.75504851,
      "learning_rate": 1.6819772869215972e-06,
      "loss": 0.77711368,
      "num_input_tokens_seen": 101268940,
      "step": 4693,
      "time_per_iteration": 2.6183571815490723
    },
    {
      "auxiliary_loss_clip": 0.0117727,
      "auxiliary_loss_mlp": 0.01025717,
      "balance_loss_clip": 1.05184007,
      "balance_loss_mlp": 1.01757514,
      "epoch": 0.564420128659893,
      "flos": 23185904428800.0,
      "grad_norm": 2.311162835248548,
      "language_loss": 0.8230204,
      "learning_rate": 1.6812082504637228e-06,
      "loss": 0.84505022,
      "num_input_tokens_seen": 101290260,
      "step": 4694,
      "time_per_iteration": 2.604752540588379
    },
    {
      "auxiliary_loss_clip": 0.01176482,
      "auxiliary_loss_mlp": 0.01025442,
      "balance_loss_clip": 1.05190706,
      "balance_loss_mlp": 1.01701427,
      "epoch": 0.564540371550532,
      "flos": 23258264376960.0,
      "grad_norm": 1.8786960109089266,
      "language_loss": 0.7426157,
      "learning_rate": 1.6804392623663025e-06,
      "loss": 0.76463497,
      "num_input_tokens_seen": 101311465,
      "step": 4695,
      "time_per_iteration": 2.6318907737731934
    },
    {
      "auxiliary_loss_clip": 0.01175675,
      "auxiliary_loss_mlp": 0.01025019,
      "balance_loss_clip": 1.05189013,
      "balance_loss_mlp": 1.01774156,
      "epoch": 0.5646606144411712,
      "flos": 25010058672000.0,
      "grad_norm": 2.012712729550645,
      "language_loss": 0.78339887,
      "learning_rate": 1.6796703227459935e-06,
      "loss": 0.80540574,
      "num_input_tokens_seen": 101329420,
      "step": 4696,
      "time_per_iteration": 2.6251747608184814
    },
    {
      "auxiliary_loss_clip": 0.01178073,
      "auxiliary_loss_mlp": 0.01025896,
      "balance_loss_clip": 1.05155861,
      "balance_loss_mlp": 1.0184958,
      "epoch": 0.5647808573318103,
      "flos": 36539645806080.0,
      "grad_norm": 2.6968687499848536,
      "language_loss": 0.75970674,
      "learning_rate": 1.6789014317194407e-06,
      "loss": 0.78174645,
      "num_input_tokens_seen": 101350900,
      "step": 4697,
      "time_per_iteration": 2.7109735012054443
    },
    {
      "auxiliary_loss_clip": 0.01183682,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.05504036,
      "balance_loss_mlp": 1.01912808,
      "epoch": 0.5649011002224493,
      "flos": 22528451842560.0,
      "grad_norm": 16.376111454265622,
      "language_loss": 0.73274523,
      "learning_rate": 1.6781325894032853e-06,
      "loss": 0.75485432,
      "num_input_tokens_seen": 101369860,
      "step": 4698,
      "time_per_iteration": 2.6875033378601074
    },
    {
      "auxiliary_loss_clip": 0.01176194,
      "auxiliary_loss_mlp": 0.01024385,
      "balance_loss_clip": 1.05022669,
      "balance_loss_mlp": 1.01658869,
      "epoch": 0.5650213431130885,
      "flos": 18515147304960.0,
      "grad_norm": 2.877177253563892,
      "language_loss": 0.91800988,
      "learning_rate": 1.6773637959141608e-06,
      "loss": 0.94001561,
      "num_input_tokens_seen": 101386835,
      "step": 4699,
      "time_per_iteration": 2.5776612758636475
    },
    {
      "auxiliary_loss_clip": 0.01174659,
      "auxiliary_loss_mlp": 0.01026282,
      "balance_loss_clip": 1.05048943,
      "balance_loss_mlp": 1.01841426,
      "epoch": 0.5651415860037275,
      "flos": 17526310819200.0,
      "grad_norm": 2.593053172135211,
      "language_loss": 0.66334939,
      "learning_rate": 1.6765950513686915e-06,
      "loss": 0.68535882,
      "num_input_tokens_seen": 101404945,
      "step": 4700,
      "time_per_iteration": 2.633531332015991
    },
    {
      "auxiliary_loss_clip": 0.01177346,
      "auxiliary_loss_mlp": 0.01029439,
      "balance_loss_clip": 1.05117106,
      "balance_loss_mlp": 1.02149951,
      "epoch": 0.5652618288943666,
      "flos": 25520026014720.0,
      "grad_norm": 1.9256336990044225,
      "language_loss": 0.76057065,
      "learning_rate": 1.675826355883496e-06,
      "loss": 0.78263855,
      "num_input_tokens_seen": 101424160,
      "step": 4701,
      "time_per_iteration": 2.700900077819824
    },
    {
      "auxiliary_loss_clip": 0.01172041,
      "auxiliary_loss_mlp": 0.01029907,
      "balance_loss_clip": 1.04904127,
      "balance_loss_mlp": 1.0221647,
      "epoch": 0.5653820717850057,
      "flos": 19683105937920.0,
      "grad_norm": 2.259743224304098,
      "language_loss": 0.79154682,
      "learning_rate": 1.6750577095751848e-06,
      "loss": 0.81356621,
      "num_input_tokens_seen": 101443270,
      "step": 4702,
      "time_per_iteration": 2.5915071964263916
    },
    {
      "auxiliary_loss_clip": 0.01174868,
      "auxiliary_loss_mlp": 0.01022467,
      "balance_loss_clip": 1.05143702,
      "balance_loss_mlp": 1.01483715,
      "epoch": 0.5655023146756448,
      "flos": 26979722910720.0,
      "grad_norm": 1.7209420552004442,
      "language_loss": 0.72722948,
      "learning_rate": 1.6742891125603605e-06,
      "loss": 0.74920279,
      "num_input_tokens_seen": 101464175,
      "step": 4703,
      "time_per_iteration": 2.638094663619995
    },
    {
      "auxiliary_loss_clip": 0.01178197,
      "auxiliary_loss_mlp": 0.01028872,
      "balance_loss_clip": 1.05201709,
      "balance_loss_mlp": 1.02086663,
      "epoch": 0.5656225575662839,
      "flos": 27669351104640.0,
      "grad_norm": 1.9420381829232627,
      "language_loss": 0.72272539,
      "learning_rate": 1.6735205649556185e-06,
      "loss": 0.74479598,
      "num_input_tokens_seen": 101484045,
      "step": 4704,
      "time_per_iteration": 2.7526304721832275
    },
    {
      "auxiliary_loss_clip": 0.01177806,
      "auxiliary_loss_mlp": 0.01029185,
      "balance_loss_clip": 1.05245066,
      "balance_loss_mlp": 1.02197325,
      "epoch": 0.5657428004569229,
      "flos": 24349732997760.0,
      "grad_norm": 2.0418473158796253,
      "language_loss": 0.84935415,
      "learning_rate": 1.6727520668775476e-06,
      "loss": 0.87142408,
      "num_input_tokens_seen": 101504330,
      "step": 4705,
      "time_per_iteration": 3.6212432384490967
    },
    {
      "auxiliary_loss_clip": 0.0118148,
      "auxiliary_loss_mlp": 0.01027966,
      "balance_loss_clip": 1.05333734,
      "balance_loss_mlp": 1.0196507,
      "epoch": 0.5658630433475621,
      "flos": 21944041562880.0,
      "grad_norm": 2.3702811686855423,
      "language_loss": 0.75584364,
      "learning_rate": 1.6719836184427275e-06,
      "loss": 0.77793813,
      "num_input_tokens_seen": 101524635,
      "step": 4706,
      "time_per_iteration": 2.6061768531799316
    },
    {
      "auxiliary_loss_clip": 0.0117358,
      "auxiliary_loss_mlp": 0.01023822,
      "balance_loss_clip": 1.04913127,
      "balance_loss_mlp": 1.01582944,
      "epoch": 0.5659832862382012,
      "flos": 30409012218240.0,
      "grad_norm": 1.9868130313032346,
      "language_loss": 0.64561498,
      "learning_rate": 1.671215219767733e-06,
      "loss": 0.66758901,
      "num_input_tokens_seen": 101544095,
      "step": 4707,
      "time_per_iteration": 2.6289069652557373
    },
    {
      "auxiliary_loss_clip": 0.01176935,
      "auxiliary_loss_mlp": 0.01031287,
      "balance_loss_clip": 1.05123019,
      "balance_loss_mlp": 1.02351141,
      "epoch": 0.5661035291288402,
      "flos": 13188194570880.0,
      "grad_norm": 2.3611399021551875,
      "language_loss": 0.75964022,
      "learning_rate": 1.670446870969127e-06,
      "loss": 0.78172243,
      "num_input_tokens_seen": 101561760,
      "step": 4708,
      "time_per_iteration": 3.4810285568237305
    },
    {
      "auxiliary_loss_clip": 0.01176982,
      "auxiliary_loss_mlp": 0.01022806,
      "balance_loss_clip": 1.05073118,
      "balance_loss_mlp": 1.01506293,
      "epoch": 0.5662237720194794,
      "flos": 16143032108160.0,
      "grad_norm": 2.588098097418897,
      "language_loss": 0.80142081,
      "learning_rate": 1.6696785721634685e-06,
      "loss": 0.82341874,
      "num_input_tokens_seen": 101576245,
      "step": 4709,
      "time_per_iteration": 2.5698578357696533
    },
    {
      "auxiliary_loss_clip": 0.01182446,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.05487275,
      "balance_loss_mlp": 1.0173378,
      "epoch": 0.5663440149101184,
      "flos": 17676848718720.0,
      "grad_norm": 2.2732002279084824,
      "language_loss": 0.73800755,
      "learning_rate": 1.6689103234673086e-06,
      "loss": 0.76008677,
      "num_input_tokens_seen": 101594565,
      "step": 4710,
      "time_per_iteration": 2.5862460136413574
    },
    {
      "auxiliary_loss_clip": 0.01180572,
      "auxiliary_loss_mlp": 0.010334,
      "balance_loss_clip": 1.05138397,
      "balance_loss_mlp": 1.02504969,
      "epoch": 0.5664642578007575,
      "flos": 23368330627200.0,
      "grad_norm": 10.538372866692013,
      "language_loss": 0.77286685,
      "learning_rate": 1.668142124997189e-06,
      "loss": 0.79500651,
      "num_input_tokens_seen": 101614225,
      "step": 4711,
      "time_per_iteration": 2.7185564041137695
    },
    {
      "auxiliary_loss_clip": 0.0107857,
      "auxiliary_loss_mlp": 0.01001163,
      "balance_loss_clip": 1.01603019,
      "balance_loss_mlp": 1.00008988,
      "epoch": 0.5665845006913967,
      "flos": 65516470945920.0,
      "grad_norm": 1.1644887266971056,
      "language_loss": 0.59722668,
      "learning_rate": 1.6673739768696453e-06,
      "loss": 0.61802399,
      "num_input_tokens_seen": 101680795,
      "step": 4712,
      "time_per_iteration": 4.169605493545532
    },
    {
      "auxiliary_loss_clip": 0.01178097,
      "auxiliary_loss_mlp": 0.01024113,
      "balance_loss_clip": 1.05291748,
      "balance_loss_mlp": 1.01604819,
      "epoch": 0.5667047435820357,
      "flos": 26140885620480.0,
      "grad_norm": 12.482538803206513,
      "language_loss": 0.77699089,
      "learning_rate": 1.6666058792012052e-06,
      "loss": 0.79901302,
      "num_input_tokens_seen": 101701680,
      "step": 4713,
      "time_per_iteration": 2.654763698577881
    },
    {
      "auxiliary_loss_clip": 0.01079059,
      "auxiliary_loss_mlp": 0.01001319,
      "balance_loss_clip": 1.01619518,
      "balance_loss_mlp": 1.00018024,
      "epoch": 0.5668249864726748,
      "flos": 71866949725440.0,
      "grad_norm": 0.8816556081637069,
      "language_loss": 0.68741417,
      "learning_rate": 1.6658378321083878e-06,
      "loss": 0.70821798,
      "num_input_tokens_seen": 101766010,
      "step": 4714,
      "time_per_iteration": 4.153012275695801
    },
    {
      "auxiliary_loss_clip": 0.01176452,
      "auxiliary_loss_mlp": 0.01022936,
      "balance_loss_clip": 1.05069005,
      "balance_loss_mlp": 1.01577747,
      "epoch": 0.5669452293633139,
      "flos": 22195667312640.0,
      "grad_norm": 1.99632408588491,
      "language_loss": 0.82518655,
      "learning_rate": 1.6650698357077055e-06,
      "loss": 0.84718037,
      "num_input_tokens_seen": 101783055,
      "step": 4715,
      "time_per_iteration": 2.625800609588623
    },
    {
      "auxiliary_loss_clip": 0.01178105,
      "auxiliary_loss_mlp": 0.01036329,
      "balance_loss_clip": 1.04986238,
      "balance_loss_mlp": 1.02799654,
      "epoch": 0.567065472253953,
      "flos": 18223193560320.0,
      "grad_norm": 4.133754370881215,
      "language_loss": 0.81192404,
      "learning_rate": 1.6643018901156632e-06,
      "loss": 0.83406836,
      "num_input_tokens_seen": 101802150,
      "step": 4716,
      "time_per_iteration": 2.654419183731079
    },
    {
      "auxiliary_loss_clip": 0.01177336,
      "auxiliary_loss_mlp": 0.01027909,
      "balance_loss_clip": 1.05074453,
      "balance_loss_mlp": 1.01959395,
      "epoch": 0.567185715144592,
      "flos": 20371548983040.0,
      "grad_norm": 3.479300548521405,
      "language_loss": 0.80105543,
      "learning_rate": 1.6635339954487566e-06,
      "loss": 0.8231079,
      "num_input_tokens_seen": 101818025,
      "step": 4717,
      "time_per_iteration": 2.5802602767944336
    },
    {
      "auxiliary_loss_clip": 0.01182042,
      "auxiliary_loss_mlp": 0.0102631,
      "balance_loss_clip": 1.05379856,
      "balance_loss_mlp": 1.01779246,
      "epoch": 0.5673059580352312,
      "flos": 23221348174080.0,
      "grad_norm": 3.8135996789705766,
      "language_loss": 0.81942546,
      "learning_rate": 1.6627661518234765e-06,
      "loss": 0.84150898,
      "num_input_tokens_seen": 101837280,
      "step": 4718,
      "time_per_iteration": 2.6385324001312256
    },
    {
      "auxiliary_loss_clip": 0.01181385,
      "auxiliary_loss_mlp": 0.0103309,
      "balance_loss_clip": 1.0534997,
      "balance_loss_mlp": 1.02469802,
      "epoch": 0.5674262009258703,
      "flos": 21719599430400.0,
      "grad_norm": 3.617684131081261,
      "language_loss": 0.85519457,
      "learning_rate": 1.661998359356302e-06,
      "loss": 0.8773393,
      "num_input_tokens_seen": 101856310,
      "step": 4719,
      "time_per_iteration": 2.629178047180176
    },
    {
      "auxiliary_loss_clip": 0.01078852,
      "auxiliary_loss_mlp": 0.01000743,
      "balance_loss_clip": 1.01686239,
      "balance_loss_mlp": 0.9996168,
      "epoch": 0.5675464438165093,
      "flos": 67470369114240.0,
      "grad_norm": 0.8731258513276929,
      "language_loss": 0.55756772,
      "learning_rate": 1.6612306181637077e-06,
      "loss": 0.57836366,
      "num_input_tokens_seen": 101915635,
      "step": 4720,
      "time_per_iteration": 3.108353853225708
    },
    {
      "auxiliary_loss_clip": 0.01177134,
      "auxiliary_loss_mlp": 0.01034962,
      "balance_loss_clip": 1.05006826,
      "balance_loss_mlp": 1.02742219,
      "epoch": 0.5676666867071485,
      "flos": 18879173688960.0,
      "grad_norm": 2.0708214980147814,
      "language_loss": 0.66078025,
      "learning_rate": 1.6604629283621598e-06,
      "loss": 0.68290126,
      "num_input_tokens_seen": 101933565,
      "step": 4721,
      "time_per_iteration": 2.6464173793792725
    },
    {
      "auxiliary_loss_clip": 0.01182989,
      "auxiliary_loss_mlp": 0.01027851,
      "balance_loss_clip": 1.05419803,
      "balance_loss_mlp": 1.01895189,
      "epoch": 0.5677869295977875,
      "flos": 33546778744320.0,
      "grad_norm": 2.3611255460849567,
      "language_loss": 0.74218446,
      "learning_rate": 1.6596952900681152e-06,
      "loss": 0.76429284,
      "num_input_tokens_seen": 101954325,
      "step": 4722,
      "time_per_iteration": 2.763087511062622
    },
    {
      "auxiliary_loss_clip": 0.01180942,
      "auxiliary_loss_mlp": 0.01027713,
      "balance_loss_clip": 1.05495834,
      "balance_loss_mlp": 1.01881385,
      "epoch": 0.5679071724884266,
      "flos": 28037256157440.0,
      "grad_norm": 7.197377865076929,
      "language_loss": 0.82281339,
      "learning_rate": 1.658927703398025e-06,
      "loss": 0.84490001,
      "num_input_tokens_seen": 101974390,
      "step": 4723,
      "time_per_iteration": 2.72314715385437
    },
    {
      "auxiliary_loss_clip": 0.01177453,
      "auxiliary_loss_mlp": 0.01027605,
      "balance_loss_clip": 1.05102217,
      "balance_loss_mlp": 1.01990986,
      "epoch": 0.5680274153790658,
      "flos": 23550110380800.0,
      "grad_norm": 2.5775246487283447,
      "language_loss": 0.77828157,
      "learning_rate": 1.6581601684683309e-06,
      "loss": 0.80033213,
      "num_input_tokens_seen": 101994815,
      "step": 4724,
      "time_per_iteration": 2.6035263538360596
    },
    {
      "auxiliary_loss_clip": 0.01179065,
      "auxiliary_loss_mlp": 0.01025833,
      "balance_loss_clip": 1.0527854,
      "balance_loss_mlp": 1.01705289,
      "epoch": 0.5681476582697048,
      "flos": 22455158140800.0,
      "grad_norm": 2.8547028039818922,
      "language_loss": 0.69113731,
      "learning_rate": 1.6573926853954674e-06,
      "loss": 0.71318626,
      "num_input_tokens_seen": 102012400,
      "step": 4725,
      "time_per_iteration": 2.6847095489501953
    },
    {
      "auxiliary_loss_clip": 0.01175235,
      "auxiliary_loss_mlp": 0.01025874,
      "balance_loss_clip": 1.05244064,
      "balance_loss_mlp": 1.0184623,
      "epoch": 0.5682679011603439,
      "flos": 19536913584000.0,
      "grad_norm": 2.30154536332773,
      "language_loss": 0.82790661,
      "learning_rate": 1.6566252542958608e-06,
      "loss": 0.84991765,
      "num_input_tokens_seen": 102031900,
      "step": 4726,
      "time_per_iteration": 2.5841989517211914
    },
    {
      "auxiliary_loss_clip": 0.01180351,
      "auxiliary_loss_mlp": 0.01029626,
      "balance_loss_clip": 1.05283928,
      "balance_loss_mlp": 1.02102494,
      "epoch": 0.568388144050983,
      "flos": 28765488493440.0,
      "grad_norm": 1.9603298211500761,
      "language_loss": 0.78720391,
      "learning_rate": 1.6558578752859305e-06,
      "loss": 0.8093037,
      "num_input_tokens_seen": 102050860,
      "step": 4727,
      "time_per_iteration": 2.696094274520874
    },
    {
      "auxiliary_loss_clip": 0.01178821,
      "auxiliary_loss_mlp": 0.01023847,
      "balance_loss_clip": 1.05210233,
      "balance_loss_mlp": 1.01536548,
      "epoch": 0.5685083869416221,
      "flos": 21209452519680.0,
      "grad_norm": 2.137811038506697,
      "language_loss": 0.78594995,
      "learning_rate": 1.6550905484820865e-06,
      "loss": 0.80797666,
      "num_input_tokens_seen": 102069320,
      "step": 4728,
      "time_per_iteration": 2.6367669105529785
    },
    {
      "auxiliary_loss_clip": 0.01179923,
      "auxiliary_loss_mlp": 0.01022492,
      "balance_loss_clip": 1.05115438,
      "balance_loss_mlp": 1.01430786,
      "epoch": 0.5686286298322611,
      "flos": 24827021942400.0,
      "grad_norm": 2.0664181228222565,
      "language_loss": 0.789747,
      "learning_rate": 1.6543232740007328e-06,
      "loss": 0.81177115,
      "num_input_tokens_seen": 102086435,
      "step": 4729,
      "time_per_iteration": 2.690279960632324
    },
    {
      "auxiliary_loss_clip": 0.01182399,
      "auxiliary_loss_mlp": 0.01025639,
      "balance_loss_clip": 1.05452943,
      "balance_loss_mlp": 1.0174315,
      "epoch": 0.5687488727229003,
      "flos": 26615121909120.0,
      "grad_norm": 2.4584425767639573,
      "language_loss": 0.67333913,
      "learning_rate": 1.653556051958263e-06,
      "loss": 0.69541955,
      "num_input_tokens_seen": 102106115,
      "step": 4730,
      "time_per_iteration": 2.6463866233825684
    },
    {
      "auxiliary_loss_clip": 0.01177916,
      "auxiliary_loss_mlp": 0.01026299,
      "balance_loss_clip": 1.05286145,
      "balance_loss_mlp": 1.0182519,
      "epoch": 0.5688691156135394,
      "flos": 20808725414400.0,
      "grad_norm": 2.9432293387214696,
      "language_loss": 0.73626518,
      "learning_rate": 1.6527888824710642e-06,
      "loss": 0.75830734,
      "num_input_tokens_seen": 102125715,
      "step": 4731,
      "time_per_iteration": 2.6293821334838867
    },
    {
      "auxiliary_loss_clip": 0.01179841,
      "auxiliary_loss_mlp": 0.01032479,
      "balance_loss_clip": 1.05193913,
      "balance_loss_mlp": 1.02430677,
      "epoch": 0.5689893585041784,
      "flos": 25880963829120.0,
      "grad_norm": 4.877672031128343,
      "language_loss": 0.77029252,
      "learning_rate": 1.6520217656555166e-06,
      "loss": 0.79241562,
      "num_input_tokens_seen": 102145005,
      "step": 4732,
      "time_per_iteration": 3.604454517364502
    },
    {
      "auxiliary_loss_clip": 0.01174615,
      "auxiliary_loss_mlp": 0.01024802,
      "balance_loss_clip": 1.05099535,
      "balance_loss_mlp": 1.01711845,
      "epoch": 0.5691096013948175,
      "flos": 23477463123840.0,
      "grad_norm": 1.7067412039480732,
      "language_loss": 0.70554805,
      "learning_rate": 1.65125470162799e-06,
      "loss": 0.72754228,
      "num_input_tokens_seen": 102165360,
      "step": 4733,
      "time_per_iteration": 2.611245632171631
    },
    {
      "auxiliary_loss_clip": 0.0117854,
      "auxiliary_loss_mlp": 0.01028498,
      "balance_loss_clip": 1.05025649,
      "balance_loss_mlp": 1.02055883,
      "epoch": 0.5692298442854566,
      "flos": 18075600576000.0,
      "grad_norm": 2.28701970394194,
      "language_loss": 0.69616979,
      "learning_rate": 1.6504876905048485e-06,
      "loss": 0.71824014,
      "num_input_tokens_seen": 102182320,
      "step": 4734,
      "time_per_iteration": 3.530435085296631
    },
    {
      "auxiliary_loss_clip": 0.01176045,
      "auxiliary_loss_mlp": 0.01027115,
      "balance_loss_clip": 1.05124974,
      "balance_loss_mlp": 1.02015924,
      "epoch": 0.5693500871760957,
      "flos": 23039317025280.0,
      "grad_norm": 2.3033539262063334,
      "language_loss": 0.72199565,
      "learning_rate": 1.6497207324024464e-06,
      "loss": 0.74402726,
      "num_input_tokens_seen": 102201220,
      "step": 4735,
      "time_per_iteration": 2.6250832080841064
    },
    {
      "auxiliary_loss_clip": 0.01177314,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.05140352,
      "balance_loss_mlp": 1.01811922,
      "epoch": 0.5694703300667348,
      "flos": 18989670902400.0,
      "grad_norm": 2.4402346289922563,
      "language_loss": 0.82506466,
      "learning_rate": 1.6489538274371305e-06,
      "loss": 0.84710145,
      "num_input_tokens_seen": 102219825,
      "step": 4736,
      "time_per_iteration": 2.62996244430542
    },
    {
      "auxiliary_loss_clip": 0.01174741,
      "auxiliary_loss_mlp": 0.01027349,
      "balance_loss_clip": 1.0513854,
      "balance_loss_mlp": 1.02002978,
      "epoch": 0.5695905729573739,
      "flos": 21908705558400.0,
      "grad_norm": 2.4804793418127082,
      "language_loss": 0.82963264,
      "learning_rate": 1.6481869757252396e-06,
      "loss": 0.85165364,
      "num_input_tokens_seen": 102238160,
      "step": 4737,
      "time_per_iteration": 2.669724702835083
    },
    {
      "auxiliary_loss_clip": 0.01176032,
      "auxiliary_loss_mlp": 0.01028622,
      "balance_loss_clip": 1.0518117,
      "balance_loss_mlp": 1.02067661,
      "epoch": 0.569710815848013,
      "flos": 28476659232000.0,
      "grad_norm": 1.411262131144837,
      "language_loss": 0.72036564,
      "learning_rate": 1.647420177383105e-06,
      "loss": 0.74241215,
      "num_input_tokens_seen": 102261030,
      "step": 4738,
      "time_per_iteration": 3.6124961376190186
    },
    {
      "auxiliary_loss_clip": 0.01173618,
      "auxiliary_loss_mlp": 0.01023745,
      "balance_loss_clip": 1.05083585,
      "balance_loss_mlp": 1.01610374,
      "epoch": 0.569831058738652,
      "flos": 28366162018560.0,
      "grad_norm": 1.9389152359997377,
      "language_loss": 0.72604668,
      "learning_rate": 1.646653432527049e-06,
      "loss": 0.74802035,
      "num_input_tokens_seen": 102281670,
      "step": 4739,
      "time_per_iteration": 2.6835596561431885
    },
    {
      "auxiliary_loss_clip": 0.01176332,
      "auxiliary_loss_mlp": 0.01029069,
      "balance_loss_clip": 1.05150342,
      "balance_loss_mlp": 1.02143407,
      "epoch": 0.5699513016292912,
      "flos": 25849973370240.0,
      "grad_norm": 2.1961736146849744,
      "language_loss": 0.74473953,
      "learning_rate": 1.645886741273387e-06,
      "loss": 0.76679349,
      "num_input_tokens_seen": 102303485,
      "step": 4740,
      "time_per_iteration": 3.6229825019836426
    },
    {
      "auxiliary_loss_clip": 0.01179123,
      "auxiliary_loss_mlp": 0.01027036,
      "balance_loss_clip": 1.05277312,
      "balance_loss_mlp": 1.01856041,
      "epoch": 0.5700715445199303,
      "flos": 18037858360320.0,
      "grad_norm": 4.2110216441866,
      "language_loss": 0.73812813,
      "learning_rate": 1.645120103738424e-06,
      "loss": 0.76018977,
      "num_input_tokens_seen": 102320995,
      "step": 4741,
      "time_per_iteration": 2.5911192893981934
    },
    {
      "auxiliary_loss_clip": 0.01174149,
      "auxiliary_loss_mlp": 0.01023677,
      "balance_loss_clip": 1.0512991,
      "balance_loss_mlp": 1.01681972,
      "epoch": 0.5701917874105693,
      "flos": 11473352392320.0,
      "grad_norm": 2.789612922102767,
      "language_loss": 0.840653,
      "learning_rate": 1.6443535200384591e-06,
      "loss": 0.86263132,
      "num_input_tokens_seen": 102339170,
      "step": 4742,
      "time_per_iteration": 2.581331968307495
    },
    {
      "auxiliary_loss_clip": 0.01174921,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.05064774,
      "balance_loss_mlp": 1.0199796,
      "epoch": 0.5703120303012085,
      "flos": 21761759018880.0,
      "grad_norm": 1.8600929002843218,
      "language_loss": 0.7068665,
      "learning_rate": 1.6435869902897827e-06,
      "loss": 0.72889531,
      "num_input_tokens_seen": 102357750,
      "step": 4743,
      "time_per_iteration": 2.6815898418426514
    },
    {
      "auxiliary_loss_clip": 0.01075535,
      "auxiliary_loss_mlp": 0.01001906,
      "balance_loss_clip": 1.01431799,
      "balance_loss_mlp": 1.0007261,
      "epoch": 0.5704322731918475,
      "flos": 56746258513920.0,
      "grad_norm": 0.7984912580876271,
      "language_loss": 0.62038004,
      "learning_rate": 1.6428205146086764e-06,
      "loss": 0.64115453,
      "num_input_tokens_seen": 102419730,
      "step": 4744,
      "time_per_iteration": 3.242302179336548
    },
    {
      "auxiliary_loss_clip": 0.01175979,
      "auxiliary_loss_mlp": 0.01032896,
      "balance_loss_clip": 1.05059254,
      "balance_loss_mlp": 1.02498007,
      "epoch": 0.5705525160824866,
      "flos": 20741141975040.0,
      "grad_norm": 1.897467688219957,
      "language_loss": 0.70753241,
      "learning_rate": 1.6420540931114142e-06,
      "loss": 0.72962117,
      "num_input_tokens_seen": 102440320,
      "step": 4745,
      "time_per_iteration": 2.6853697299957275
    },
    {
      "auxiliary_loss_clip": 0.01177791,
      "auxiliary_loss_mlp": 0.01028678,
      "balance_loss_clip": 1.05103719,
      "balance_loss_mlp": 1.0204463,
      "epoch": 0.5706727589731257,
      "flos": 18771262254720.0,
      "grad_norm": 2.2476767233161157,
      "language_loss": 0.79182291,
      "learning_rate": 1.6412877259142616e-06,
      "loss": 0.81388766,
      "num_input_tokens_seen": 102460240,
      "step": 4746,
      "time_per_iteration": 2.6359121799468994
    },
    {
      "auxiliary_loss_clip": 0.0118039,
      "auxiliary_loss_mlp": 0.01029149,
      "balance_loss_clip": 1.05369878,
      "balance_loss_mlp": 1.0209657,
      "epoch": 0.5707930018637648,
      "flos": 27634733372160.0,
      "grad_norm": 2.1419638650480746,
      "language_loss": 0.74241155,
      "learning_rate": 1.6405214131334757e-06,
      "loss": 0.76450694,
      "num_input_tokens_seen": 102478765,
      "step": 4747,
      "time_per_iteration": 2.6727898120880127
    },
    {
      "auxiliary_loss_clip": 0.0117821,
      "auxiliary_loss_mlp": 0.01027997,
      "balance_loss_clip": 1.05355859,
      "balance_loss_mlp": 1.0200789,
      "epoch": 0.5709132447544039,
      "flos": 27597673514880.0,
      "grad_norm": 2.3336266032050212,
      "language_loss": 0.79468453,
      "learning_rate": 1.6397551548853052e-06,
      "loss": 0.81674659,
      "num_input_tokens_seen": 102496930,
      "step": 4748,
      "time_per_iteration": 2.6447229385375977
    },
    {
      "auxiliary_loss_clip": 0.01179312,
      "auxiliary_loss_mlp": 0.01022671,
      "balance_loss_clip": 1.05203772,
      "balance_loss_mlp": 1.01438022,
      "epoch": 0.571033487645043,
      "flos": 21686095019520.0,
      "grad_norm": 1.9772351742313552,
      "language_loss": 0.70806676,
      "learning_rate": 1.6389889512859917e-06,
      "loss": 0.73008662,
      "num_input_tokens_seen": 102516590,
      "step": 4749,
      "time_per_iteration": 2.598935604095459
    },
    {
      "auxiliary_loss_clip": 0.01074853,
      "auxiliary_loss_mlp": 0.01000624,
      "balance_loss_clip": 1.01362979,
      "balance_loss_mlp": 0.99946731,
      "epoch": 0.5711537305356821,
      "flos": 70181445980160.0,
      "grad_norm": 0.821438247805042,
      "language_loss": 0.60330844,
      "learning_rate": 1.638222802451767e-06,
      "loss": 0.62406325,
      "num_input_tokens_seen": 102578070,
      "step": 4750,
      "time_per_iteration": 3.230593681335449
    },
    {
      "auxiliary_loss_clip": 0.01170087,
      "auxiliary_loss_mlp": 0.01022489,
      "balance_loss_clip": 1.04796076,
      "balance_loss_mlp": 1.01502347,
      "epoch": 0.5712739734263211,
      "flos": 24717494396160.0,
      "grad_norm": 1.9084027448016148,
      "language_loss": 0.75355482,
      "learning_rate": 1.6374567084988561e-06,
      "loss": 0.77548057,
      "num_input_tokens_seen": 102599255,
      "step": 4751,
      "time_per_iteration": 2.6813178062438965
    },
    {
      "auxiliary_loss_clip": 0.01182573,
      "auxiliary_loss_mlp": 0.01030998,
      "balance_loss_clip": 1.05484378,
      "balance_loss_mlp": 1.02308857,
      "epoch": 0.5713942163169603,
      "flos": 26578169792640.0,
      "grad_norm": 1.6938135976931776,
      "language_loss": 0.76376545,
      "learning_rate": 1.6366906695434738e-06,
      "loss": 0.78590119,
      "num_input_tokens_seen": 102621775,
      "step": 4752,
      "time_per_iteration": 2.657411575317383
    },
    {
      "auxiliary_loss_clip": 0.01179059,
      "auxiliary_loss_mlp": 0.01025711,
      "balance_loss_clip": 1.05292702,
      "balance_loss_mlp": 1.01841569,
      "epoch": 0.5715144592075994,
      "flos": 21142443697920.0,
      "grad_norm": 2.5278534846124328,
      "language_loss": 0.86072028,
      "learning_rate": 1.6359246857018275e-06,
      "loss": 0.88276798,
      "num_input_tokens_seen": 102639305,
      "step": 4753,
      "time_per_iteration": 2.601224660873413
    },
    {
      "auxiliary_loss_clip": 0.01173665,
      "auxiliary_loss_mlp": 0.01023042,
      "balance_loss_clip": 1.05023098,
      "balance_loss_mlp": 1.0158751,
      "epoch": 0.5716347020982384,
      "flos": 23330265189120.0,
      "grad_norm": 1.9138730784755298,
      "language_loss": 0.78665286,
      "learning_rate": 1.6351587570901178e-06,
      "loss": 0.80861998,
      "num_input_tokens_seen": 102659430,
      "step": 4754,
      "time_per_iteration": 2.6433351039886475
    },
    {
      "auxiliary_loss_clip": 0.01174655,
      "auxiliary_loss_mlp": 0.01028492,
      "balance_loss_clip": 1.04970932,
      "balance_loss_mlp": 1.02052879,
      "epoch": 0.5717549449888776,
      "flos": 17009555806080.0,
      "grad_norm": 3.2456947456696312,
      "language_loss": 0.76246536,
      "learning_rate": 1.634392883824534e-06,
      "loss": 0.78449678,
      "num_input_tokens_seen": 102671430,
      "step": 4755,
      "time_per_iteration": 2.575711250305176
    },
    {
      "auxiliary_loss_clip": 0.01177382,
      "auxiliary_loss_mlp": 0.01023689,
      "balance_loss_clip": 1.05015254,
      "balance_loss_mlp": 1.01581526,
      "epoch": 0.5718751878795166,
      "flos": 35518130922240.0,
      "grad_norm": 2.6494035873122255,
      "language_loss": 0.6769371,
      "learning_rate": 1.6336270660212595e-06,
      "loss": 0.69894779,
      "num_input_tokens_seen": 102693025,
      "step": 4756,
      "time_per_iteration": 2.695815324783325
    },
    {
      "auxiliary_loss_clip": 0.01179462,
      "auxiliary_loss_mlp": 0.0102953,
      "balance_loss_clip": 1.05219722,
      "balance_loss_mlp": 1.0214448,
      "epoch": 0.5719954307701557,
      "flos": 38613989255040.0,
      "grad_norm": 3.239972232429629,
      "language_loss": 0.6601932,
      "learning_rate": 1.6328613037964676e-06,
      "loss": 0.68228316,
      "num_input_tokens_seen": 102716090,
      "step": 4757,
      "time_per_iteration": 2.795531749725342
    },
    {
      "auxiliary_loss_clip": 0.01172201,
      "auxiliary_loss_mlp": 0.01024296,
      "balance_loss_clip": 1.0480994,
      "balance_loss_mlp": 1.01599872,
      "epoch": 0.5721156736607949,
      "flos": 20631111638400.0,
      "grad_norm": 1.7889956022459748,
      "language_loss": 0.67737156,
      "learning_rate": 1.6320955972663241e-06,
      "loss": 0.69933653,
      "num_input_tokens_seen": 102735685,
      "step": 4758,
      "time_per_iteration": 3.6716010570526123
    },
    {
      "auxiliary_loss_clip": 0.01176227,
      "auxiliary_loss_mlp": 0.01030653,
      "balance_loss_clip": 1.04951406,
      "balance_loss_mlp": 1.02242112,
      "epoch": 0.5722359165514339,
      "flos": 37415076076800.0,
      "grad_norm": 4.638276872055367,
      "language_loss": 0.65371764,
      "learning_rate": 1.6313299465469857e-06,
      "loss": 0.6757865,
      "num_input_tokens_seen": 102758415,
      "step": 4759,
      "time_per_iteration": 2.724033832550049
    },
    {
      "auxiliary_loss_clip": 0.01180349,
      "auxiliary_loss_mlp": 0.0103077,
      "balance_loss_clip": 1.05399048,
      "balance_loss_mlp": 1.0227654,
      "epoch": 0.572356159442073,
      "flos": 21972877205760.0,
      "grad_norm": 3.2199713342336724,
      "language_loss": 0.7951982,
      "learning_rate": 1.6305643517546014e-06,
      "loss": 0.81730938,
      "num_input_tokens_seen": 102773795,
      "step": 4760,
      "time_per_iteration": 2.655329465866089
    },
    {
      "auxiliary_loss_clip": 0.01176088,
      "auxiliary_loss_mlp": 0.01027047,
      "balance_loss_clip": 1.05153263,
      "balance_loss_mlp": 1.01986814,
      "epoch": 0.5724764023327121,
      "flos": 19135540033920.0,
      "grad_norm": 2.078903172169758,
      "language_loss": 0.84756172,
      "learning_rate": 1.629798813005311e-06,
      "loss": 0.86959302,
      "num_input_tokens_seen": 102793515,
      "step": 4761,
      "time_per_iteration": 2.594393014907837
    },
    {
      "auxiliary_loss_clip": 0.01176671,
      "auxiliary_loss_mlp": 0.01023562,
      "balance_loss_clip": 1.05228627,
      "balance_loss_mlp": 1.01570582,
      "epoch": 0.5725966452233512,
      "flos": 22819759142400.0,
      "grad_norm": 1.8820575063436997,
      "language_loss": 0.71112633,
      "learning_rate": 1.6290333304152473e-06,
      "loss": 0.73312867,
      "num_input_tokens_seen": 102813390,
      "step": 4762,
      "time_per_iteration": 3.561507225036621
    },
    {
      "auxiliary_loss_clip": 0.0117747,
      "auxiliary_loss_mlp": 0.01025001,
      "balance_loss_clip": 1.0523901,
      "balance_loss_mlp": 1.01734138,
      "epoch": 0.5727168881139902,
      "flos": 41496610498560.0,
      "grad_norm": 8.318521512272135,
      "language_loss": 0.56764352,
      "learning_rate": 1.6282679041005314e-06,
      "loss": 0.58966827,
      "num_input_tokens_seen": 102838980,
      "step": 4763,
      "time_per_iteration": 2.7673258781433105
    },
    {
      "auxiliary_loss_clip": 0.01176109,
      "auxiliary_loss_mlp": 0.01027113,
      "balance_loss_clip": 1.05153537,
      "balance_loss_mlp": 1.01881611,
      "epoch": 0.5728371310046293,
      "flos": 14647675985280.0,
      "grad_norm": 5.034985965941005,
      "language_loss": 0.87381279,
      "learning_rate": 1.6275025341772789e-06,
      "loss": 0.89584506,
      "num_input_tokens_seen": 102855285,
      "step": 4764,
      "time_per_iteration": 2.603419303894043
    },
    {
      "auxiliary_loss_clip": 0.01176549,
      "auxiliary_loss_mlp": 0.01026683,
      "balance_loss_clip": 1.04993463,
      "balance_loss_mlp": 1.01835001,
      "epoch": 0.5729573738952685,
      "flos": 21506613736320.0,
      "grad_norm": 15.221534941291141,
      "language_loss": 0.82045245,
      "learning_rate": 1.626737220761596e-06,
      "loss": 0.84248471,
      "num_input_tokens_seen": 102872750,
      "step": 4765,
      "time_per_iteration": 3.5378034114837646
    },
    {
      "auxiliary_loss_clip": 0.01178197,
      "auxiliary_loss_mlp": 0.01026407,
      "balance_loss_clip": 1.05178666,
      "balance_loss_mlp": 1.01846194,
      "epoch": 0.5730776167859075,
      "flos": 23621680229760.0,
      "grad_norm": 2.283964145538755,
      "language_loss": 0.79224122,
      "learning_rate": 1.62597196396958e-06,
      "loss": 0.8142873,
      "num_input_tokens_seen": 102890920,
      "step": 4766,
      "time_per_iteration": 3.517773389816284
    },
    {
      "auxiliary_loss_clip": 0.01177761,
      "auxiliary_loss_mlp": 0.01023566,
      "balance_loss_clip": 1.05133462,
      "balance_loss_mlp": 1.01600838,
      "epoch": 0.5731978596765466,
      "flos": 25739224761600.0,
      "grad_norm": 5.6332453961200954,
      "language_loss": 0.8545168,
      "learning_rate": 1.6252067639173197e-06,
      "loss": 0.87653005,
      "num_input_tokens_seen": 102912830,
      "step": 4767,
      "time_per_iteration": 2.7047154903411865
    },
    {
      "auxiliary_loss_clip": 0.01173639,
      "auxiliary_loss_mlp": 0.01029081,
      "balance_loss_clip": 1.0491426,
      "balance_loss_mlp": 1.02098393,
      "epoch": 0.5733181025671857,
      "flos": 26359509749760.0,
      "grad_norm": 1.9010293838726087,
      "language_loss": 0.69732428,
      "learning_rate": 1.6244416207208956e-06,
      "loss": 0.71935147,
      "num_input_tokens_seen": 102933765,
      "step": 4768,
      "time_per_iteration": 2.6379213333129883
    },
    {
      "auxiliary_loss_clip": 0.01177081,
      "auxiliary_loss_mlp": 0.0103047,
      "balance_loss_clip": 1.0516454,
      "balance_loss_mlp": 1.02255499,
      "epoch": 0.5734383454578248,
      "flos": 29423874833280.0,
      "grad_norm": 2.0390501350815935,
      "language_loss": 0.73761904,
      "learning_rate": 1.6236765344963787e-06,
      "loss": 0.75969452,
      "num_input_tokens_seen": 102955025,
      "step": 4769,
      "time_per_iteration": 2.7116119861602783
    },
    {
      "auxiliary_loss_clip": 0.01177565,
      "auxiliary_loss_mlp": 0.01026359,
      "balance_loss_clip": 1.0508945,
      "balance_loss_mlp": 1.01870537,
      "epoch": 0.5735585883484638,
      "flos": 34969954487040.0,
      "grad_norm": 3.237428234926885,
      "language_loss": 0.69223869,
      "learning_rate": 1.6229115053598322e-06,
      "loss": 0.71427792,
      "num_input_tokens_seen": 102976780,
      "step": 4770,
      "time_per_iteration": 2.707967519760132
    },
    {
      "auxiliary_loss_clip": 0.01179657,
      "auxiliary_loss_mlp": 0.01022034,
      "balance_loss_clip": 1.053406,
      "balance_loss_mlp": 1.01436293,
      "epoch": 0.573678831239103,
      "flos": 18770759464320.0,
      "grad_norm": 4.462367210018057,
      "language_loss": 0.72120011,
      "learning_rate": 1.6221465334273108e-06,
      "loss": 0.74321705,
      "num_input_tokens_seen": 102995990,
      "step": 4771,
      "time_per_iteration": 2.6765148639678955
    },
    {
      "auxiliary_loss_clip": 0.01176178,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.04957819,
      "balance_loss_mlp": 1.01889396,
      "epoch": 0.5737990741297421,
      "flos": 25702883176320.0,
      "grad_norm": 2.7474555386989956,
      "language_loss": 0.62277746,
      "learning_rate": 1.6213816188148593e-06,
      "loss": 0.64480126,
      "num_input_tokens_seen": 103014695,
      "step": 4772,
      "time_per_iteration": 2.664396286010742
    },
    {
      "auxiliary_loss_clip": 0.01174432,
      "auxiliary_loss_mlp": 0.0102644,
      "balance_loss_clip": 1.0516355,
      "balance_loss_mlp": 1.0191381,
      "epoch": 0.5739193170203811,
      "flos": 27269234530560.0,
      "grad_norm": 2.208531357439374,
      "language_loss": 0.77317595,
      "learning_rate": 1.6206167616385162e-06,
      "loss": 0.79518467,
      "num_input_tokens_seen": 103035760,
      "step": 4773,
      "time_per_iteration": 2.6375374794006348
    },
    {
      "auxiliary_loss_clip": 0.01181435,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.05368114,
      "balance_loss_mlp": 1.01898718,
      "epoch": 0.5740395599110203,
      "flos": 12239721993600.0,
      "grad_norm": 2.4236797492866335,
      "language_loss": 0.73910904,
      "learning_rate": 1.6198519620143078e-06,
      "loss": 0.76120025,
      "num_input_tokens_seen": 103052915,
      "step": 4774,
      "time_per_iteration": 2.6944079399108887
    },
    {
      "auxiliary_loss_clip": 0.01177223,
      "auxiliary_loss_mlp": 0.01030881,
      "balance_loss_clip": 1.05316377,
      "balance_loss_mlp": 1.0232935,
      "epoch": 0.5741598028016593,
      "flos": 25921399564800.0,
      "grad_norm": 2.671977268482226,
      "language_loss": 0.78186184,
      "learning_rate": 1.6190872200582546e-06,
      "loss": 0.80394292,
      "num_input_tokens_seen": 103074655,
      "step": 4775,
      "time_per_iteration": 2.6569972038269043
    },
    {
      "auxiliary_loss_clip": 0.01176767,
      "auxiliary_loss_mlp": 0.01024264,
      "balance_loss_clip": 1.05272555,
      "balance_loss_mlp": 1.01669168,
      "epoch": 0.5742800456922984,
      "flos": 19244133826560.0,
      "grad_norm": 2.102814444521196,
      "language_loss": 0.77852631,
      "learning_rate": 1.6183225358863676e-06,
      "loss": 0.80053663,
      "num_input_tokens_seen": 103091550,
      "step": 4776,
      "time_per_iteration": 2.598371982574463
    },
    {
      "auxiliary_loss_clip": 0.01176427,
      "auxiliary_loss_mlp": 0.01028698,
      "balance_loss_clip": 1.04969156,
      "balance_loss_mlp": 1.02099097,
      "epoch": 0.5744002885829376,
      "flos": 30920487932160.0,
      "grad_norm": 3.1053619388315155,
      "language_loss": 0.7174657,
      "learning_rate": 1.617557909614648e-06,
      "loss": 0.73951697,
      "num_input_tokens_seen": 103110985,
      "step": 4777,
      "time_per_iteration": 2.6612613201141357
    },
    {
      "auxiliary_loss_clip": 0.01172929,
      "auxiliary_loss_mlp": 0.0102286,
      "balance_loss_clip": 1.04964733,
      "balance_loss_mlp": 1.01574361,
      "epoch": 0.5745205314735766,
      "flos": 23840017050240.0,
      "grad_norm": 2.054104422298209,
      "language_loss": 0.86077923,
      "learning_rate": 1.6167933413590899e-06,
      "loss": 0.8827371,
      "num_input_tokens_seen": 103129890,
      "step": 4778,
      "time_per_iteration": 2.7061219215393066
    },
    {
      "auxiliary_loss_clip": 0.01178134,
      "auxiliary_loss_mlp": 0.01025335,
      "balance_loss_clip": 1.05404854,
      "balance_loss_mlp": 1.01753283,
      "epoch": 0.5746407743642157,
      "flos": 12311902373760.0,
      "grad_norm": 2.609113845699121,
      "language_loss": 0.90276021,
      "learning_rate": 1.6160288312356773e-06,
      "loss": 0.92479491,
      "num_input_tokens_seen": 103147020,
      "step": 4779,
      "time_per_iteration": 2.592432975769043
    },
    {
      "auxiliary_loss_clip": 0.01173004,
      "auxiliary_loss_mlp": 0.01026267,
      "balance_loss_clip": 1.0487963,
      "balance_loss_mlp": 1.01912022,
      "epoch": 0.5747610172548548,
      "flos": 24133658734080.0,
      "grad_norm": 1.7210965495494939,
      "language_loss": 0.81694734,
      "learning_rate": 1.6152643793603857e-06,
      "loss": 0.83894002,
      "num_input_tokens_seen": 103167370,
      "step": 4780,
      "time_per_iteration": 2.65303635597229
    },
    {
      "auxiliary_loss_clip": 0.0117414,
      "auxiliary_loss_mlp": 0.01023947,
      "balance_loss_clip": 1.05092442,
      "balance_loss_mlp": 1.01662135,
      "epoch": 0.5748812601454939,
      "flos": 25408451393280.0,
      "grad_norm": 2.3737598321682793,
      "language_loss": 0.87970102,
      "learning_rate": 1.6144999858491815e-06,
      "loss": 0.90168184,
      "num_input_tokens_seen": 103186000,
      "step": 4781,
      "time_per_iteration": 2.6081807613372803
    },
    {
      "auxiliary_loss_clip": 0.01174849,
      "auxiliary_loss_mlp": 0.01025891,
      "balance_loss_clip": 1.04863596,
      "balance_loss_mlp": 1.01782107,
      "epoch": 0.575001503036133,
      "flos": 30624942827520.0,
      "grad_norm": 2.5342180981360296,
      "language_loss": 0.85777009,
      "learning_rate": 1.6137356508180232e-06,
      "loss": 0.87977755,
      "num_input_tokens_seen": 103207710,
      "step": 4782,
      "time_per_iteration": 2.6965720653533936
    },
    {
      "auxiliary_loss_clip": 0.01174272,
      "auxiliary_loss_mlp": 0.01026767,
      "balance_loss_clip": 1.04924738,
      "balance_loss_mlp": 1.01890206,
      "epoch": 0.5751217459267721,
      "flos": 21726566668800.0,
      "grad_norm": 2.098226188469332,
      "language_loss": 0.81224298,
      "learning_rate": 1.6129713743828593e-06,
      "loss": 0.83425331,
      "num_input_tokens_seen": 103226720,
      "step": 4783,
      "time_per_iteration": 2.5827603340148926
    },
    {
      "auxiliary_loss_clip": 0.01172968,
      "auxiliary_loss_mlp": 0.01026269,
      "balance_loss_clip": 1.0496769,
      "balance_loss_mlp": 1.0189321,
      "epoch": 0.5752419888174112,
      "flos": 21651620941440.0,
      "grad_norm": 1.5979252462983613,
      "language_loss": 0.7566123,
      "learning_rate": 1.6122071566596306e-06,
      "loss": 0.77860469,
      "num_input_tokens_seen": 103246995,
      "step": 4784,
      "time_per_iteration": 2.6330373287200928
    },
    {
      "auxiliary_loss_clip": 0.01178052,
      "auxiliary_loss_mlp": 0.01030023,
      "balance_loss_clip": 1.05183721,
      "balance_loss_mlp": 1.02250051,
      "epoch": 0.5753622317080502,
      "flos": 17775997234560.0,
      "grad_norm": 26.715488056528127,
      "language_loss": 0.83516502,
      "learning_rate": 1.6114429977642674e-06,
      "loss": 0.8572458,
      "num_input_tokens_seen": 103261500,
      "step": 4785,
      "time_per_iteration": 3.447356939315796
    },
    {
      "auxiliary_loss_clip": 0.01175423,
      "auxiliary_loss_mlp": 0.01028365,
      "balance_loss_clip": 1.05095172,
      "balance_loss_mlp": 1.02051473,
      "epoch": 0.5754824745986894,
      "flos": 19789616741760.0,
      "grad_norm": 2.0085829940699664,
      "language_loss": 0.73847914,
      "learning_rate": 1.6106788978126926e-06,
      "loss": 0.760517,
      "num_input_tokens_seen": 103280475,
      "step": 4786,
      "time_per_iteration": 2.59505033493042
    },
    {
      "auxiliary_loss_clip": 0.01173548,
      "auxiliary_loss_mlp": 0.01022404,
      "balance_loss_clip": 1.04833174,
      "balance_loss_mlp": 1.01441646,
      "epoch": 0.5756027174893285,
      "flos": 30985665160320.0,
      "grad_norm": 2.465729735219615,
      "language_loss": 0.79089302,
      "learning_rate": 1.6099148569208196e-06,
      "loss": 0.8128525,
      "num_input_tokens_seen": 103297695,
      "step": 4787,
      "time_per_iteration": 2.6786856651306152
    },
    {
      "auxiliary_loss_clip": 0.01178336,
      "auxiliary_loss_mlp": 0.01025637,
      "balance_loss_clip": 1.05027735,
      "balance_loss_mlp": 1.01739383,
      "epoch": 0.5757229603799675,
      "flos": 28546864364160.0,
      "grad_norm": 3.2856981403179866,
      "language_loss": 0.62867415,
      "learning_rate": 1.6091508752045523e-06,
      "loss": 0.65071386,
      "num_input_tokens_seen": 103318575,
      "step": 4788,
      "time_per_iteration": 2.594362735748291
    },
    {
      "auxiliary_loss_clip": 0.01172681,
      "auxiliary_loss_mlp": 0.01022236,
      "balance_loss_clip": 1.04930067,
      "balance_loss_mlp": 1.01454067,
      "epoch": 0.5758432032706067,
      "flos": 22999024944000.0,
      "grad_norm": 1.6448352624608542,
      "language_loss": 0.86697614,
      "learning_rate": 1.608386952779787e-06,
      "loss": 0.88892531,
      "num_input_tokens_seen": 103337945,
      "step": 4789,
      "time_per_iteration": 3.5730326175689697
    },
    {
      "auxiliary_loss_clip": 0.01173271,
      "auxiliary_loss_mlp": 0.01021766,
      "balance_loss_clip": 1.0503329,
      "balance_loss_mlp": 1.01426494,
      "epoch": 0.5759634461612457,
      "flos": 25739727552000.0,
      "grad_norm": 1.841276593537957,
      "language_loss": 0.74495447,
      "learning_rate": 1.6076230897624098e-06,
      "loss": 0.76690483,
      "num_input_tokens_seen": 103360150,
      "step": 4790,
      "time_per_iteration": 2.6210885047912598
    },
    {
      "auxiliary_loss_clip": 0.01174301,
      "auxiliary_loss_mlp": 0.01022184,
      "balance_loss_clip": 1.0489254,
      "balance_loss_mlp": 1.01476371,
      "epoch": 0.5760836890518848,
      "flos": 30591761639040.0,
      "grad_norm": 3.1225598591942005,
      "language_loss": 0.77673995,
      "learning_rate": 1.6068592862682974e-06,
      "loss": 0.7987048,
      "num_input_tokens_seen": 103378305,
      "step": 4791,
      "time_per_iteration": 2.663290500640869
    },
    {
      "auxiliary_loss_clip": 0.01175384,
      "auxiliary_loss_mlp": 0.01026685,
      "balance_loss_clip": 1.05041492,
      "balance_loss_mlp": 1.01872802,
      "epoch": 0.576203931942524,
      "flos": 36538963447680.0,
      "grad_norm": 2.1112146098351787,
      "language_loss": 0.74235725,
      "learning_rate": 1.6060955424133187e-06,
      "loss": 0.76437801,
      "num_input_tokens_seen": 103399230,
      "step": 4792,
      "time_per_iteration": 4.532433032989502
    },
    {
      "auxiliary_loss_clip": 0.01177092,
      "auxiliary_loss_mlp": 0.01025783,
      "balance_loss_clip": 1.05041695,
      "balance_loss_mlp": 1.01779592,
      "epoch": 0.576324174833163,
      "flos": 25516937445120.0,
      "grad_norm": 3.30519703009889,
      "language_loss": 0.89782101,
      "learning_rate": 1.6053318583133332e-06,
      "loss": 0.91984975,
      "num_input_tokens_seen": 103420100,
      "step": 4793,
      "time_per_iteration": 2.6591503620147705
    },
    {
      "auxiliary_loss_clip": 0.01172758,
      "auxiliary_loss_mlp": 0.01024887,
      "balance_loss_clip": 1.04882383,
      "balance_loss_mlp": 1.01743078,
      "epoch": 0.5764444177238021,
      "flos": 25119262995840.0,
      "grad_norm": 3.1727649533886333,
      "language_loss": 0.75146818,
      "learning_rate": 1.6045682340841907e-06,
      "loss": 0.77344453,
      "num_input_tokens_seen": 103439025,
      "step": 4794,
      "time_per_iteration": 2.6178793907165527
    },
    {
      "auxiliary_loss_clip": 0.01077333,
      "auxiliary_loss_mlp": 0.01003203,
      "balance_loss_clip": 1.01550364,
      "balance_loss_mlp": 1.0020113,
      "epoch": 0.5765646606144411,
      "flos": 62212687758720.0,
      "grad_norm": 0.7526396123171004,
      "language_loss": 0.5791629,
      "learning_rate": 1.6038046698417336e-06,
      "loss": 0.59996825,
      "num_input_tokens_seen": 103499920,
      "step": 4795,
      "time_per_iteration": 3.1713979244232178
    },
    {
      "auxiliary_loss_clip": 0.0117734,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.0526557,
      "balance_loss_mlp": 1.01829028,
      "epoch": 0.5766849035050803,
      "flos": 25118760205440.0,
      "grad_norm": 3.154562748711575,
      "language_loss": 0.68862712,
      "learning_rate": 1.6030411657017919e-06,
      "loss": 0.7106654,
      "num_input_tokens_seen": 103519575,
      "step": 4796,
      "time_per_iteration": 2.6768386363983154
    },
    {
      "auxiliary_loss_clip": 0.01175458,
      "auxiliary_loss_mlp": 0.01022725,
      "balance_loss_clip": 1.05020499,
      "balance_loss_mlp": 1.01566815,
      "epoch": 0.5768051463957193,
      "flos": 15991093578240.0,
      "grad_norm": 1.7993329358086259,
      "language_loss": 0.84383869,
      "learning_rate": 1.6022777217801903e-06,
      "loss": 0.86582053,
      "num_input_tokens_seen": 103536530,
      "step": 4797,
      "time_per_iteration": 2.6363279819488525
    },
    {
      "auxiliary_loss_clip": 0.01178886,
      "auxiliary_loss_mlp": 0.0102769,
      "balance_loss_clip": 1.05343771,
      "balance_loss_mlp": 1.01958394,
      "epoch": 0.5769253892863584,
      "flos": 22163635359360.0,
      "grad_norm": 2.1958181978638933,
      "language_loss": 0.74151647,
      "learning_rate": 1.601514338192742e-06,
      "loss": 0.76358223,
      "num_input_tokens_seen": 103556460,
      "step": 4798,
      "time_per_iteration": 2.584043502807617
    },
    {
      "auxiliary_loss_clip": 0.0117227,
      "auxiliary_loss_mlp": 0.01021715,
      "balance_loss_clip": 1.04966223,
      "balance_loss_mlp": 1.01460361,
      "epoch": 0.5770456321769976,
      "flos": 22856388036480.0,
      "grad_norm": 2.7871570287861243,
      "language_loss": 0.71481943,
      "learning_rate": 1.6007510150552514e-06,
      "loss": 0.73675931,
      "num_input_tokens_seen": 103574520,
      "step": 4799,
      "time_per_iteration": 2.679586887359619
    },
    {
      "auxiliary_loss_clip": 0.01177996,
      "auxiliary_loss_mlp": 0.01028914,
      "balance_loss_clip": 1.0507015,
      "balance_loss_mlp": 1.02098083,
      "epoch": 0.5771658750676366,
      "flos": 46353672489600.0,
      "grad_norm": 1.8547340690811989,
      "language_loss": 0.62180829,
      "learning_rate": 1.599987752483515e-06,
      "loss": 0.64387739,
      "num_input_tokens_seen": 103598965,
      "step": 4800,
      "time_per_iteration": 2.7791948318481445
    },
    {
      "auxiliary_loss_clip": 0.01177504,
      "auxiliary_loss_mlp": 0.01025003,
      "balance_loss_clip": 1.05272222,
      "balance_loss_mlp": 1.01698017,
      "epoch": 0.5772861179582757,
      "flos": 22159972172160.0,
      "grad_norm": 1.9527071160189817,
      "language_loss": 0.68202215,
      "learning_rate": 1.5992245505933184e-06,
      "loss": 0.70404726,
      "num_input_tokens_seen": 103618665,
      "step": 4801,
      "time_per_iteration": 2.7113378047943115
    },
    {
      "auxiliary_loss_clip": 0.01178062,
      "auxiliary_loss_mlp": 0.01030509,
      "balance_loss_clip": 1.05265331,
      "balance_loss_mlp": 1.0228498,
      "epoch": 0.5774063608489148,
      "flos": 31248926916480.0,
      "grad_norm": 3.124342348023743,
      "language_loss": 0.71270382,
      "learning_rate": 1.5984614095004388e-06,
      "loss": 0.73478955,
      "num_input_tokens_seen": 103639800,
      "step": 4802,
      "time_per_iteration": 2.651947259902954
    },
    {
      "auxiliary_loss_clip": 0.01177213,
      "auxiliary_loss_mlp": 0.01025434,
      "balance_loss_clip": 1.05119371,
      "balance_loss_mlp": 1.01708603,
      "epoch": 0.5775266037395539,
      "flos": 22527123039360.0,
      "grad_norm": 2.6425925134082977,
      "language_loss": 0.81027842,
      "learning_rate": 1.5976983293206438e-06,
      "loss": 0.83230484,
      "num_input_tokens_seen": 103655605,
      "step": 4803,
      "time_per_iteration": 2.649231433868408
    },
    {
      "auxiliary_loss_clip": 0.01175537,
      "auxiliary_loss_mlp": 0.01025791,
      "balance_loss_clip": 1.05039549,
      "balance_loss_mlp": 1.01843596,
      "epoch": 0.577646846630193,
      "flos": 21068790860160.0,
      "grad_norm": 2.339120344335614,
      "language_loss": 0.71264911,
      "learning_rate": 1.5969353101696928e-06,
      "loss": 0.73466235,
      "num_input_tokens_seen": 103674045,
      "step": 4804,
      "time_per_iteration": 2.573765516281128
    },
    {
      "auxiliary_loss_clip": 0.01173708,
      "auxiliary_loss_mlp": 0.01025944,
      "balance_loss_clip": 1.05057335,
      "balance_loss_mlp": 1.01891375,
      "epoch": 0.5777670895208321,
      "flos": 29714284293120.0,
      "grad_norm": 1.8863080066279874,
      "language_loss": 0.79784566,
      "learning_rate": 1.5961723521633341e-06,
      "loss": 0.81984222,
      "num_input_tokens_seen": 103695285,
      "step": 4805,
      "time_per_iteration": 2.7015891075134277
    },
    {
      "auxiliary_loss_clip": 0.01174184,
      "auxiliary_loss_mlp": 0.01030853,
      "balance_loss_clip": 1.04952347,
      "balance_loss_mlp": 1.02300322,
      "epoch": 0.5778873324114712,
      "flos": 19500428344320.0,
      "grad_norm": 2.532486720509808,
      "language_loss": 0.90979624,
      "learning_rate": 1.5954094554173097e-06,
      "loss": 0.93184668,
      "num_input_tokens_seen": 103713275,
      "step": 4806,
      "time_per_iteration": 2.5991265773773193
    },
    {
      "auxiliary_loss_clip": 0.01174736,
      "auxiliary_loss_mlp": 0.01021359,
      "balance_loss_clip": 1.05055594,
      "balance_loss_mlp": 1.01454329,
      "epoch": 0.5780075753021102,
      "flos": 14136846716160.0,
      "grad_norm": 2.1300281106560126,
      "language_loss": 0.79065973,
      "learning_rate": 1.5946466200473482e-06,
      "loss": 0.81262064,
      "num_input_tokens_seen": 103731185,
      "step": 4807,
      "time_per_iteration": 2.599271059036255
    },
    {
      "auxiliary_loss_clip": 0.01172169,
      "auxiliary_loss_mlp": 0.01024282,
      "balance_loss_clip": 1.04979134,
      "balance_loss_mlp": 1.01684976,
      "epoch": 0.5781278181927494,
      "flos": 15262178883840.0,
      "grad_norm": 2.6185131839588456,
      "language_loss": 0.83369386,
      "learning_rate": 1.5938838461691723e-06,
      "loss": 0.85565841,
      "num_input_tokens_seen": 103748095,
      "step": 4808,
      "time_per_iteration": 2.5787017345428467
    },
    {
      "auxiliary_loss_clip": 0.01179425,
      "auxiliary_loss_mlp": 0.01028397,
      "balance_loss_clip": 1.05289078,
      "balance_loss_mlp": 1.02033257,
      "epoch": 0.5782480610833884,
      "flos": 16726831856640.0,
      "grad_norm": 2.4715238903089536,
      "language_loss": 0.83090591,
      "learning_rate": 1.593121133898494e-06,
      "loss": 0.85298407,
      "num_input_tokens_seen": 103765300,
      "step": 4809,
      "time_per_iteration": 2.575007200241089
    },
    {
      "auxiliary_loss_clip": 0.01179166,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.05199087,
      "balance_loss_mlp": 1.02263856,
      "epoch": 0.5783683039740275,
      "flos": 25482140144640.0,
      "grad_norm": 2.443680273282509,
      "language_loss": 0.79233527,
      "learning_rate": 1.592358483351016e-06,
      "loss": 0.81443536,
      "num_input_tokens_seen": 103785475,
      "step": 4810,
      "time_per_iteration": 2.598682165145874
    },
    {
      "auxiliary_loss_clip": 0.01171763,
      "auxiliary_loss_mlp": 0.01023988,
      "balance_loss_clip": 1.04913592,
      "balance_loss_mlp": 1.01641822,
      "epoch": 0.5784885468646667,
      "flos": 18405835240320.0,
      "grad_norm": 5.740035410777151,
      "language_loss": 0.72413182,
      "learning_rate": 1.5915958946424326e-06,
      "loss": 0.74608934,
      "num_input_tokens_seen": 103804160,
      "step": 4811,
      "time_per_iteration": 3.5357506275177
    },
    {
      "auxiliary_loss_clip": 0.01175735,
      "auxiliary_loss_mlp": 0.01029694,
      "balance_loss_clip": 1.05064225,
      "balance_loss_mlp": 1.02204716,
      "epoch": 0.5786087897553057,
      "flos": 46100717936640.0,
      "grad_norm": 1.5865804652556992,
      "language_loss": 0.74392599,
      "learning_rate": 1.5908333678884271e-06,
      "loss": 0.7659803,
      "num_input_tokens_seen": 103830580,
      "step": 4812,
      "time_per_iteration": 2.8563570976257324
    },
    {
      "auxiliary_loss_clip": 0.01175593,
      "auxiliary_loss_mlp": 0.01026448,
      "balance_loss_clip": 1.05160689,
      "balance_loss_mlp": 1.01857448,
      "epoch": 0.5787290326459448,
      "flos": 12385950261120.0,
      "grad_norm": 2.2207108008329945,
      "language_loss": 0.73771656,
      "learning_rate": 1.5900709032046743e-06,
      "loss": 0.75973701,
      "num_input_tokens_seen": 103848655,
      "step": 4813,
      "time_per_iteration": 2.663141965866089
    },
    {
      "auxiliary_loss_clip": 0.01176933,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.05282831,
      "balance_loss_mlp": 1.01743734,
      "epoch": 0.5788492755365839,
      "flos": 23290332243840.0,
      "grad_norm": 1.9937929768652336,
      "language_loss": 0.78143811,
      "learning_rate": 1.5893085007068391e-06,
      "loss": 0.8034541,
      "num_input_tokens_seen": 103866215,
      "step": 4814,
      "time_per_iteration": 2.644390344619751
    },
    {
      "auxiliary_loss_clip": 0.01175769,
      "auxiliary_loss_mlp": 0.01028109,
      "balance_loss_clip": 1.05261159,
      "balance_loss_mlp": 1.02096581,
      "epoch": 0.578969518427223,
      "flos": 24061047390720.0,
      "grad_norm": 2.8937404496877903,
      "language_loss": 0.70889407,
      "learning_rate": 1.5885461605105786e-06,
      "loss": 0.73093289,
      "num_input_tokens_seen": 103887815,
      "step": 4815,
      "time_per_iteration": 2.633620023727417
    },
    {
      "auxiliary_loss_clip": 0.01176751,
      "auxiliary_loss_mlp": 0.01024704,
      "balance_loss_clip": 1.05088449,
      "balance_loss_mlp": 1.01668763,
      "epoch": 0.579089761317862,
      "flos": 21871825269120.0,
      "grad_norm": 2.032274662887844,
      "language_loss": 0.76622957,
      "learning_rate": 1.5877838827315375e-06,
      "loss": 0.78824413,
      "num_input_tokens_seen": 103906360,
      "step": 4816,
      "time_per_iteration": 3.469082832336426
    },
    {
      "auxiliary_loss_clip": 0.01175334,
      "auxiliary_loss_mlp": 0.01028072,
      "balance_loss_clip": 1.05185628,
      "balance_loss_mlp": 1.01966763,
      "epoch": 0.5792100042085012,
      "flos": 22929681738240.0,
      "grad_norm": 8.387007933869596,
      "language_loss": 0.70155859,
      "learning_rate": 1.587021667485355e-06,
      "loss": 0.72359264,
      "num_input_tokens_seen": 103925730,
      "step": 4817,
      "time_per_iteration": 2.658205986022949
    },
    {
      "auxiliary_loss_clip": 0.01177053,
      "auxiliary_loss_mlp": 0.01027348,
      "balance_loss_clip": 1.05102789,
      "balance_loss_mlp": 1.01911068,
      "epoch": 0.5793302470991403,
      "flos": 21470056669440.0,
      "grad_norm": 1.9461135424991978,
      "language_loss": 0.78406644,
      "learning_rate": 1.5862595148876559e-06,
      "loss": 0.80611044,
      "num_input_tokens_seen": 103945835,
      "step": 4818,
      "time_per_iteration": 3.4782795906066895
    },
    {
      "auxiliary_loss_clip": 0.01176909,
      "auxiliary_loss_mlp": 0.01031667,
      "balance_loss_clip": 1.05138993,
      "balance_loss_mlp": 1.02360797,
      "epoch": 0.5794504899897793,
      "flos": 12711013367040.0,
      "grad_norm": 2.4585222613878335,
      "language_loss": 0.76432484,
      "learning_rate": 1.58549742505406e-06,
      "loss": 0.78641063,
      "num_input_tokens_seen": 103960580,
      "step": 4819,
      "time_per_iteration": 3.5327413082122803
    },
    {
      "auxiliary_loss_clip": 0.01176307,
      "auxiliary_loss_mlp": 0.01027749,
      "balance_loss_clip": 1.05056727,
      "balance_loss_mlp": 1.01966655,
      "epoch": 0.5795707328804185,
      "flos": 14867054300160.0,
      "grad_norm": 2.5878478107429332,
      "language_loss": 0.75992858,
      "learning_rate": 1.5847353981001747e-06,
      "loss": 0.78196913,
      "num_input_tokens_seen": 103977760,
      "step": 4820,
      "time_per_iteration": 2.6140973567962646
    },
    {
      "auxiliary_loss_clip": 0.01176342,
      "auxiliary_loss_mlp": 0.01023528,
      "balance_loss_clip": 1.05096316,
      "balance_loss_mlp": 1.01618803,
      "epoch": 0.5796909757710575,
      "flos": 36430046432640.0,
      "grad_norm": 2.147735376677562,
      "language_loss": 0.69959897,
      "learning_rate": 1.5839734341415993e-06,
      "loss": 0.72159773,
      "num_input_tokens_seen": 103999960,
      "step": 4821,
      "time_per_iteration": 2.760908842086792
    },
    {
      "auxiliary_loss_clip": 0.0117458,
      "auxiliary_loss_mlp": 0.0102474,
      "balance_loss_clip": 1.0525744,
      "balance_loss_mlp": 1.01774263,
      "epoch": 0.5798112186616966,
      "flos": 23039891642880.0,
      "grad_norm": 2.4998948976073705,
      "language_loss": 0.76541686,
      "learning_rate": 1.5832115332939238e-06,
      "loss": 0.78741014,
      "num_input_tokens_seen": 104018400,
      "step": 4822,
      "time_per_iteration": 2.606684684753418
    },
    {
      "auxiliary_loss_clip": 0.01177456,
      "auxiliary_loss_mlp": 0.01025097,
      "balance_loss_clip": 1.0517745,
      "balance_loss_mlp": 1.01794505,
      "epoch": 0.5799314615523358,
      "flos": 16652604401280.0,
      "grad_norm": 1.8839129019852316,
      "language_loss": 0.75085038,
      "learning_rate": 1.5824496956727272e-06,
      "loss": 0.77287591,
      "num_input_tokens_seen": 104035605,
      "step": 4823,
      "time_per_iteration": 2.540309429168701
    },
    {
      "auxiliary_loss_clip": 0.01176068,
      "auxiliary_loss_mlp": 0.01020871,
      "balance_loss_clip": 1.05186367,
      "balance_loss_mlp": 1.01340795,
      "epoch": 0.5800517044429748,
      "flos": 20485673470080.0,
      "grad_norm": 1.90699588032479,
      "language_loss": 0.73387337,
      "learning_rate": 1.5816879213935797e-06,
      "loss": 0.7558428,
      "num_input_tokens_seen": 104054415,
      "step": 4824,
      "time_per_iteration": 2.7535974979400635
    },
    {
      "auxiliary_loss_clip": 0.01174981,
      "auxiliary_loss_mlp": 0.01024098,
      "balance_loss_clip": 1.05211508,
      "balance_loss_mlp": 1.01685572,
      "epoch": 0.5801719473336139,
      "flos": 31538258968320.0,
      "grad_norm": 2.299517638044643,
      "language_loss": 0.79581785,
      "learning_rate": 1.5809262105720416e-06,
      "loss": 0.81780863,
      "num_input_tokens_seen": 104075455,
      "step": 4825,
      "time_per_iteration": 2.6785829067230225
    },
    {
      "auxiliary_loss_clip": 0.01172455,
      "auxiliary_loss_mlp": 0.01022471,
      "balance_loss_clip": 1.04919136,
      "balance_loss_mlp": 1.01548219,
      "epoch": 0.580292190224253,
      "flos": 20375966355840.0,
      "grad_norm": 2.4834533737542226,
      "language_loss": 0.79450881,
      "learning_rate": 1.5801645633236644e-06,
      "loss": 0.81645805,
      "num_input_tokens_seen": 104096440,
      "step": 4826,
      "time_per_iteration": 2.6599204540252686
    },
    {
      "auxiliary_loss_clip": 0.01171862,
      "auxiliary_loss_mlp": 0.01024625,
      "balance_loss_clip": 1.05026174,
      "balance_loss_mlp": 1.01698971,
      "epoch": 0.5804124331148921,
      "flos": 26615373304320.0,
      "grad_norm": 1.8736535445705447,
      "language_loss": 0.77373219,
      "learning_rate": 1.579402979763989e-06,
      "loss": 0.79569709,
      "num_input_tokens_seen": 104116775,
      "step": 4827,
      "time_per_iteration": 2.611832857131958
    },
    {
      "auxiliary_loss_clip": 0.01178334,
      "auxiliary_loss_mlp": 0.01025987,
      "balance_loss_clip": 1.05239475,
      "balance_loss_mlp": 1.01746941,
      "epoch": 0.5805326760055312,
      "flos": 13478496289920.0,
      "grad_norm": 2.299024687700711,
      "language_loss": 0.80997801,
      "learning_rate": 1.578641460008548e-06,
      "loss": 0.83202118,
      "num_input_tokens_seen": 104134510,
      "step": 4828,
      "time_per_iteration": 2.6064436435699463
    },
    {
      "auxiliary_loss_clip": 0.01173095,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.05012369,
      "balance_loss_mlp": 1.02146757,
      "epoch": 0.5806529188961702,
      "flos": 12091374823680.0,
      "grad_norm": 2.0881580167341496,
      "language_loss": 0.6821022,
      "learning_rate": 1.5778800041728613e-06,
      "loss": 0.70411682,
      "num_input_tokens_seen": 104150800,
      "step": 4829,
      "time_per_iteration": 2.5506410598754883
    },
    {
      "auxiliary_loss_clip": 0.0117524,
      "auxiliary_loss_mlp": 0.01025887,
      "balance_loss_clip": 1.0518291,
      "balance_loss_mlp": 1.01860297,
      "epoch": 0.5807731617868094,
      "flos": 26214107495040.0,
      "grad_norm": 1.4760771726035715,
      "language_loss": 0.66078162,
      "learning_rate": 1.577118612372443e-06,
      "loss": 0.6827929,
      "num_input_tokens_seen": 104172640,
      "step": 4830,
      "time_per_iteration": 2.660579204559326
    },
    {
      "auxiliary_loss_clip": 0.01178543,
      "auxiliary_loss_mlp": 0.01027644,
      "balance_loss_clip": 1.05149949,
      "balance_loss_mlp": 1.01981759,
      "epoch": 0.5808934046774484,
      "flos": 37962139190400.0,
      "grad_norm": 2.111009103068764,
      "language_loss": 0.70527911,
      "learning_rate": 1.5763572847227943e-06,
      "loss": 0.72734094,
      "num_input_tokens_seen": 104193525,
      "step": 4831,
      "time_per_iteration": 2.7386839389801025
    },
    {
      "auxiliary_loss_clip": 0.01172738,
      "auxiliary_loss_mlp": 0.01020704,
      "balance_loss_clip": 1.04864037,
      "balance_loss_mlp": 1.01353335,
      "epoch": 0.5810136475680875,
      "flos": 20485853038080.0,
      "grad_norm": 4.188954458857625,
      "language_loss": 0.81585491,
      "learning_rate": 1.5755960213394091e-06,
      "loss": 0.8377893,
      "num_input_tokens_seen": 104210625,
      "step": 4832,
      "time_per_iteration": 2.6278445720672607
    },
    {
      "auxiliary_loss_clip": 0.01175467,
      "auxiliary_loss_mlp": 0.01024986,
      "balance_loss_clip": 1.0512588,
      "balance_loss_mlp": 1.01708245,
      "epoch": 0.5811338904587267,
      "flos": 17530153574400.0,
      "grad_norm": 1.7646630712263511,
      "language_loss": 0.78457582,
      "learning_rate": 1.5748348223377703e-06,
      "loss": 0.80658031,
      "num_input_tokens_seen": 104228180,
      "step": 4833,
      "time_per_iteration": 2.575746774673462
    },
    {
      "auxiliary_loss_clip": 0.01174356,
      "auxiliary_loss_mlp": 0.01024924,
      "balance_loss_clip": 1.05022371,
      "balance_loss_mlp": 1.017241,
      "epoch": 0.5812541333493657,
      "flos": 19458017360640.0,
      "grad_norm": 2.1411501242943505,
      "language_loss": 0.77828002,
      "learning_rate": 1.5740736878333507e-06,
      "loss": 0.80027282,
      "num_input_tokens_seen": 104246020,
      "step": 4834,
      "time_per_iteration": 2.626434326171875
    },
    {
      "auxiliary_loss_clip": 0.01177993,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.05140662,
      "balance_loss_mlp": 1.01866782,
      "epoch": 0.5813743762400048,
      "flos": 20594949621120.0,
      "grad_norm": 2.8954217842760763,
      "language_loss": 0.77471852,
      "learning_rate": 1.5733126179416143e-06,
      "loss": 0.79676104,
      "num_input_tokens_seen": 104260505,
      "step": 4835,
      "time_per_iteration": 2.5483574867248535
    },
    {
      "auxiliary_loss_clip": 0.01174244,
      "auxiliary_loss_mlp": 0.01022931,
      "balance_loss_clip": 1.05045795,
      "balance_loss_mlp": 1.01491404,
      "epoch": 0.5814946191306439,
      "flos": 33178227246720.0,
      "grad_norm": 2.545359193882761,
      "language_loss": 0.72892159,
      "learning_rate": 1.5725516127780137e-06,
      "loss": 0.75089335,
      "num_input_tokens_seen": 104282640,
      "step": 4836,
      "time_per_iteration": 2.6547722816467285
    },
    {
      "auxiliary_loss_clip": 0.01179024,
      "auxiliary_loss_mlp": 0.0102098,
      "balance_loss_clip": 1.04996026,
      "balance_loss_mlp": 1.01277256,
      "epoch": 0.581614862021283,
      "flos": 16143283503360.0,
      "grad_norm": 3.3162836185137294,
      "language_loss": 0.8865906,
      "learning_rate": 1.5717906724579943e-06,
      "loss": 0.90859067,
      "num_input_tokens_seen": 104299700,
      "step": 4837,
      "time_per_iteration": 2.5757768154144287
    },
    {
      "auxiliary_loss_clip": 0.01178493,
      "auxiliary_loss_mlp": 0.01033382,
      "balance_loss_clip": 1.05301332,
      "balance_loss_mlp": 1.02547276,
      "epoch": 0.581735104911922,
      "flos": 33802642298880.0,
      "grad_norm": 2.799301270243501,
      "language_loss": 0.67838454,
      "learning_rate": 1.571029797096989e-06,
      "loss": 0.70050329,
      "num_input_tokens_seen": 104320805,
      "step": 4838,
      "time_per_iteration": 3.6135926246643066
    },
    {
      "auxiliary_loss_clip": 0.01172096,
      "auxiliary_loss_mlp": 0.01023976,
      "balance_loss_clip": 1.04854548,
      "balance_loss_mlp": 1.0164299,
      "epoch": 0.5818553478025612,
      "flos": 23331163029120.0,
      "grad_norm": 2.264914985271504,
      "language_loss": 0.78758675,
      "learning_rate": 1.570268986810423e-06,
      "loss": 0.80954742,
      "num_input_tokens_seen": 104340700,
      "step": 4839,
      "time_per_iteration": 2.6012275218963623
    },
    {
      "auxiliary_loss_clip": 0.01175014,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 1.05103707,
      "balance_loss_mlp": 1.01771569,
      "epoch": 0.5819755906932003,
      "flos": 20996143603200.0,
      "grad_norm": 2.008314342317228,
      "language_loss": 0.75138772,
      "learning_rate": 1.5695082417137096e-06,
      "loss": 0.77338898,
      "num_input_tokens_seen": 104358575,
      "step": 4840,
      "time_per_iteration": 2.5814926624298096
    },
    {
      "auxiliary_loss_clip": 0.01173324,
      "auxiliary_loss_mlp": 0.01023997,
      "balance_loss_clip": 1.05098915,
      "balance_loss_mlp": 1.0169245,
      "epoch": 0.5820958335838393,
      "flos": 21431668008960.0,
      "grad_norm": 1.8805426238720917,
      "language_loss": 0.74986225,
      "learning_rate": 1.5687475619222539e-06,
      "loss": 0.77183545,
      "num_input_tokens_seen": 104378530,
      "step": 4841,
      "time_per_iteration": 2.6076934337615967
    },
    {
      "auxiliary_loss_clip": 0.01177859,
      "auxiliary_loss_mlp": 0.01025345,
      "balance_loss_clip": 1.05157471,
      "balance_loss_mlp": 1.01664019,
      "epoch": 0.5822160764744785,
      "flos": 17967473660160.0,
      "grad_norm": 4.0598778521480625,
      "language_loss": 0.73576128,
      "learning_rate": 1.5679869475514496e-06,
      "loss": 0.75779331,
      "num_input_tokens_seen": 104395465,
      "step": 4842,
      "time_per_iteration": 3.4396049976348877
    },
    {
      "auxiliary_loss_clip": 0.01176699,
      "auxiliary_loss_mlp": 0.01025943,
      "balance_loss_clip": 1.05077994,
      "balance_loss_mlp": 1.01757491,
      "epoch": 0.5823363193651175,
      "flos": 23033858158080.0,
      "grad_norm": 2.670559405160225,
      "language_loss": 0.81435466,
      "learning_rate": 1.567226398716682e-06,
      "loss": 0.83638102,
      "num_input_tokens_seen": 104415380,
      "step": 4843,
      "time_per_iteration": 2.6271142959594727
    },
    {
      "auxiliary_loss_clip": 0.01178943,
      "auxiliary_loss_mlp": 0.01030212,
      "balance_loss_clip": 1.05198288,
      "balance_loss_mlp": 1.02188575,
      "epoch": 0.5824565622557566,
      "flos": 32891840110080.0,
      "grad_norm": 1.8601681577106708,
      "language_loss": 0.61872923,
      "learning_rate": 1.566465915533326e-06,
      "loss": 0.6408208,
      "num_input_tokens_seen": 104437410,
      "step": 4844,
      "time_per_iteration": 3.585444450378418
    },
    {
      "auxiliary_loss_clip": 0.011752,
      "auxiliary_loss_mlp": 0.01026263,
      "balance_loss_clip": 1.05104041,
      "balance_loss_mlp": 1.01825547,
      "epoch": 0.5825768051463958,
      "flos": 22229674513920.0,
      "grad_norm": 1.872450496625505,
      "language_loss": 0.88176382,
      "learning_rate": 1.5657054981167458e-06,
      "loss": 0.90377843,
      "num_input_tokens_seen": 104456305,
      "step": 4845,
      "time_per_iteration": 3.525031805038452
    },
    {
      "auxiliary_loss_clip": 0.01174868,
      "auxiliary_loss_mlp": 0.01028598,
      "balance_loss_clip": 1.05268812,
      "balance_loss_mlp": 1.02119529,
      "epoch": 0.5826970480370348,
      "flos": 28001561016960.0,
      "grad_norm": 1.8323747738608862,
      "language_loss": 0.67520714,
      "learning_rate": 1.5649451465822965e-06,
      "loss": 0.69724178,
      "num_input_tokens_seen": 104477695,
      "step": 4846,
      "time_per_iteration": 2.6935133934020996
    },
    {
      "auxiliary_loss_clip": 0.01173339,
      "auxiliary_loss_mlp": 0.01024216,
      "balance_loss_clip": 1.05093634,
      "balance_loss_mlp": 1.01681578,
      "epoch": 0.5828172909276739,
      "flos": 17858053854720.0,
      "grad_norm": 1.7290631418370943,
      "language_loss": 0.83790582,
      "learning_rate": 1.5641848610453218e-06,
      "loss": 0.85988134,
      "num_input_tokens_seen": 104496355,
      "step": 4847,
      "time_per_iteration": 2.5220179557800293
    },
    {
      "auxiliary_loss_clip": 0.01175619,
      "auxiliary_loss_mlp": 0.01024411,
      "balance_loss_clip": 1.05147243,
      "balance_loss_mlp": 1.01648343,
      "epoch": 0.582937533818313,
      "flos": 19865244827520.0,
      "grad_norm": 3.305069654238593,
      "language_loss": 0.85877991,
      "learning_rate": 1.563424641621158e-06,
      "loss": 0.88078022,
      "num_input_tokens_seen": 104515535,
      "step": 4848,
      "time_per_iteration": 2.651080846786499
    },
    {
      "auxiliary_loss_clip": 0.01178026,
      "auxiliary_loss_mlp": 0.01026627,
      "balance_loss_clip": 1.05214405,
      "balance_loss_mlp": 1.01831794,
      "epoch": 0.5830577767089521,
      "flos": 26870734068480.0,
      "grad_norm": 1.9152295690390717,
      "language_loss": 0.69852775,
      "learning_rate": 1.5626644884251282e-06,
      "loss": 0.72057426,
      "num_input_tokens_seen": 104535055,
      "step": 4849,
      "time_per_iteration": 2.603252649307251
    },
    {
      "auxiliary_loss_clip": 0.01174632,
      "auxiliary_loss_mlp": 0.0102152,
      "balance_loss_clip": 1.05045223,
      "balance_loss_mlp": 1.0143348,
      "epoch": 0.5831780195995911,
      "flos": 25298205575040.0,
      "grad_norm": 1.878817753775426,
      "language_loss": 0.87776387,
      "learning_rate": 1.5619044015725488e-06,
      "loss": 0.89972544,
      "num_input_tokens_seen": 104554745,
      "step": 4850,
      "time_per_iteration": 2.644857168197632
    },
    {
      "auxiliary_loss_clip": 0.01185779,
      "auxiliary_loss_mlp": 0.01027808,
      "balance_loss_clip": 1.0568831,
      "balance_loss_mlp": 1.01922441,
      "epoch": 0.5832982624902303,
      "flos": 14756988049920.0,
      "grad_norm": 3.8463049156028157,
      "language_loss": 0.86972612,
      "learning_rate": 1.5611443811787224e-06,
      "loss": 0.89186203,
      "num_input_tokens_seen": 104568870,
      "step": 4851,
      "time_per_iteration": 2.553152561187744
    },
    {
      "auxiliary_loss_clip": 0.01171992,
      "auxiliary_loss_mlp": 0.01023,
      "balance_loss_clip": 1.05035424,
      "balance_loss_mlp": 1.01639569,
      "epoch": 0.5834185053808694,
      "flos": 20444555376000.0,
      "grad_norm": 2.0735835697261282,
      "language_loss": 0.6909427,
      "learning_rate": 1.560384427358945e-06,
      "loss": 0.71289265,
      "num_input_tokens_seen": 104588415,
      "step": 4852,
      "time_per_iteration": 2.6110875606536865
    },
    {
      "auxiliary_loss_clip": 0.01176984,
      "auxiliary_loss_mlp": 0.01023252,
      "balance_loss_clip": 1.04996157,
      "balance_loss_mlp": 1.01560223,
      "epoch": 0.5835387482715084,
      "flos": 27200394115200.0,
      "grad_norm": 1.6300942800788296,
      "language_loss": 0.73074961,
      "learning_rate": 1.5596245402284998e-06,
      "loss": 0.75275195,
      "num_input_tokens_seen": 104611940,
      "step": 4853,
      "time_per_iteration": 2.671952247619629
    },
    {
      "auxiliary_loss_clip": 0.01176769,
      "auxiliary_loss_mlp": 0.01027477,
      "balance_loss_clip": 1.05216169,
      "balance_loss_mlp": 1.01969302,
      "epoch": 0.5836589911621476,
      "flos": 16654615562880.0,
      "grad_norm": 3.647311990427533,
      "language_loss": 0.81877476,
      "learning_rate": 1.5588647199026619e-06,
      "loss": 0.84081727,
      "num_input_tokens_seen": 104629675,
      "step": 4854,
      "time_per_iteration": 2.5439257621765137
    },
    {
      "auxiliary_loss_clip": 0.0118149,
      "auxiliary_loss_mlp": 0.0102692,
      "balance_loss_clip": 1.05355906,
      "balance_loss_mlp": 1.01862264,
      "epoch": 0.5837792340527866,
      "flos": 20446817932800.0,
      "grad_norm": 2.4482104765871053,
      "language_loss": 0.87324113,
      "learning_rate": 1.5581049664966956e-06,
      "loss": 0.89532524,
      "num_input_tokens_seen": 104647435,
      "step": 4855,
      "time_per_iteration": 2.6182940006256104
    },
    {
      "auxiliary_loss_clip": 0.01077324,
      "auxiliary_loss_mlp": 0.01001737,
      "balance_loss_clip": 1.01615071,
      "balance_loss_mlp": 1.00050282,
      "epoch": 0.5838994769434257,
      "flos": 65995480765440.0,
      "grad_norm": 0.9912933526616694,
      "language_loss": 0.65084434,
      "learning_rate": 1.5573452801258545e-06,
      "loss": 0.67163503,
      "num_input_tokens_seen": 104694605,
      "step": 4856,
      "time_per_iteration": 2.968928813934326
    },
    {
      "auxiliary_loss_clip": 0.01179591,
      "auxiliary_loss_mlp": 0.01038238,
      "balance_loss_clip": 1.05391884,
      "balance_loss_mlp": 1.03029907,
      "epoch": 0.5840197198340649,
      "flos": 21470523546240.0,
      "grad_norm": 3.6275945411833597,
      "language_loss": 0.63525307,
      "learning_rate": 1.5565856609053824e-06,
      "loss": 0.65743142,
      "num_input_tokens_seen": 104713400,
      "step": 4857,
      "time_per_iteration": 2.656747579574585
    },
    {
      "auxiliary_loss_clip": 0.01178394,
      "auxiliary_loss_mlp": 0.01022615,
      "balance_loss_clip": 1.05344427,
      "balance_loss_mlp": 1.01467514,
      "epoch": 0.5841399627247039,
      "flos": 19135144984320.0,
      "grad_norm": 2.2628596779660857,
      "language_loss": 0.80172199,
      "learning_rate": 1.5558261089505127e-06,
      "loss": 0.82373214,
      "num_input_tokens_seen": 104732130,
      "step": 4858,
      "time_per_iteration": 2.6578075885772705
    },
    {
      "auxiliary_loss_clip": 0.01180405,
      "auxiliary_loss_mlp": 0.01024305,
      "balance_loss_clip": 1.05459499,
      "balance_loss_mlp": 1.01652372,
      "epoch": 0.584260205615343,
      "flos": 26425692558720.0,
      "grad_norm": 2.015780695873506,
      "language_loss": 0.80188936,
      "learning_rate": 1.5550666243764697e-06,
      "loss": 0.82393646,
      "num_input_tokens_seen": 104750290,
      "step": 4859,
      "time_per_iteration": 2.684985637664795
    },
    {
      "auxiliary_loss_clip": 0.01178262,
      "auxiliary_loss_mlp": 0.01026185,
      "balance_loss_clip": 1.05212641,
      "balance_loss_mlp": 1.0186038,
      "epoch": 0.584380448505982,
      "flos": 13881809174400.0,
      "grad_norm": 2.5795986607161363,
      "language_loss": 0.77362716,
      "learning_rate": 1.554307207298465e-06,
      "loss": 0.79567158,
      "num_input_tokens_seen": 104768550,
      "step": 4860,
      "time_per_iteration": 2.6506502628326416
    },
    {
      "auxiliary_loss_clip": 0.01178144,
      "auxiliary_loss_mlp": 0.01028003,
      "balance_loss_clip": 1.05172527,
      "balance_loss_mlp": 1.01962852,
      "epoch": 0.5845006913966212,
      "flos": 21543709507200.0,
      "grad_norm": 1.9391962394725728,
      "language_loss": 0.78550291,
      "learning_rate": 1.553547857831704e-06,
      "loss": 0.80756438,
      "num_input_tokens_seen": 104785060,
      "step": 4861,
      "time_per_iteration": 2.6641786098480225
    },
    {
      "auxiliary_loss_clip": 0.01077893,
      "auxiliary_loss_mlp": 0.01001756,
      "balance_loss_clip": 1.01694524,
      "balance_loss_mlp": 1.00047421,
      "epoch": 0.5846209342872603,
      "flos": 58375452712320.0,
      "grad_norm": 0.8916383129137174,
      "language_loss": 0.64135623,
      "learning_rate": 1.5527885760913771e-06,
      "loss": 0.66215265,
      "num_input_tokens_seen": 104834950,
      "step": 4862,
      "time_per_iteration": 2.996105909347534
    },
    {
      "auxiliary_loss_clip": 0.01176096,
      "auxiliary_loss_mlp": 0.01023697,
      "balance_loss_clip": 1.05405521,
      "balance_loss_mlp": 1.01625788,
      "epoch": 0.5847411771778993,
      "flos": 18588045957120.0,
      "grad_norm": 1.9173689746955593,
      "language_loss": 0.76657265,
      "learning_rate": 1.552029362192668e-06,
      "loss": 0.78857058,
      "num_input_tokens_seen": 104854210,
      "step": 4863,
      "time_per_iteration": 2.648219108581543
    },
    {
      "auxiliary_loss_clip": 0.01177409,
      "auxiliary_loss_mlp": 0.01025986,
      "balance_loss_clip": 1.05216455,
      "balance_loss_mlp": 1.01811218,
      "epoch": 0.5848614200685385,
      "flos": 24240780069120.0,
      "grad_norm": 1.8306495454723604,
      "language_loss": 0.72339624,
      "learning_rate": 1.5512702162507478e-06,
      "loss": 0.74543023,
      "num_input_tokens_seen": 104874525,
      "step": 4864,
      "time_per_iteration": 3.550361156463623
    },
    {
      "auxiliary_loss_clip": 0.01078069,
      "auxiliary_loss_mlp": 0.01001468,
      "balance_loss_clip": 1.01730943,
      "balance_loss_mlp": 1.00026405,
      "epoch": 0.5849816629591775,
      "flos": 71660245933440.0,
      "grad_norm": 1.1043492415697513,
      "language_loss": 0.55707622,
      "learning_rate": 1.5505111383807792e-06,
      "loss": 0.57787156,
      "num_input_tokens_seen": 104937195,
      "step": 4865,
      "time_per_iteration": 3.221541166305542
    },
    {
      "auxiliary_loss_clip": 0.01175901,
      "auxiliary_loss_mlp": 0.01024924,
      "balance_loss_clip": 1.04993153,
      "balance_loss_mlp": 1.01719308,
      "epoch": 0.5851019058498166,
      "flos": 23802095266560.0,
      "grad_norm": 1.8551326287064278,
      "language_loss": 0.80929911,
      "learning_rate": 1.5497521286979138e-06,
      "loss": 0.83130741,
      "num_input_tokens_seen": 104957435,
      "step": 4866,
      "time_per_iteration": 2.6205546855926514
    },
    {
      "auxiliary_loss_clip": 0.01180129,
      "auxiliary_loss_mlp": 0.0102799,
      "balance_loss_clip": 1.05183744,
      "balance_loss_mlp": 1.01965141,
      "epoch": 0.5852221487404557,
      "flos": 24388516707840.0,
      "grad_norm": 2.7025114146923372,
      "language_loss": 0.74381143,
      "learning_rate": 1.5489931873172927e-06,
      "loss": 0.76589262,
      "num_input_tokens_seen": 104978755,
      "step": 4867,
      "time_per_iteration": 2.6225976943969727
    },
    {
      "auxiliary_loss_clip": 0.01177422,
      "auxiliary_loss_mlp": 0.01026946,
      "balance_loss_clip": 1.05363011,
      "balance_loss_mlp": 1.01956737,
      "epoch": 0.5853423916310948,
      "flos": 27271425260160.0,
      "grad_norm": 1.9144599302453882,
      "language_loss": 0.79052842,
      "learning_rate": 1.5482343143540467e-06,
      "loss": 0.81257212,
      "num_input_tokens_seen": 105000020,
      "step": 4868,
      "time_per_iteration": 3.636089563369751
    },
    {
      "auxiliary_loss_clip": 0.01178176,
      "auxiliary_loss_mlp": 0.01020273,
      "balance_loss_clip": 1.0516963,
      "balance_loss_mlp": 1.01285851,
      "epoch": 0.5854626345217339,
      "flos": 11983786611840.0,
      "grad_norm": 2.1169098329441876,
      "language_loss": 0.82922566,
      "learning_rate": 1.547475509923295e-06,
      "loss": 0.85121012,
      "num_input_tokens_seen": 105017060,
      "step": 4869,
      "time_per_iteration": 2.6562561988830566
    },
    {
      "auxiliary_loss_clip": 0.0107684,
      "auxiliary_loss_mlp": 0.01001664,
      "balance_loss_clip": 1.01632822,
      "balance_loss_mlp": 1.0003531,
      "epoch": 0.585582877412373,
      "flos": 64342335173760.0,
      "grad_norm": 0.7415933271725251,
      "language_loss": 0.55997133,
      "learning_rate": 1.5467167741401495e-06,
      "loss": 0.58075631,
      "num_input_tokens_seen": 105078540,
      "step": 4870,
      "time_per_iteration": 3.1967504024505615
    },
    {
      "auxiliary_loss_clip": 0.01178114,
      "auxiliary_loss_mlp": 0.01024073,
      "balance_loss_clip": 1.05389619,
      "balance_loss_mlp": 1.0165689,
      "epoch": 0.5857031203030121,
      "flos": 17011926103680.0,
      "grad_norm": 2.4543843806072467,
      "language_loss": 0.71445185,
      "learning_rate": 1.5459581071197083e-06,
      "loss": 0.73647368,
      "num_input_tokens_seen": 105094200,
      "step": 4871,
      "time_per_iteration": 3.4080166816711426
    },
    {
      "auxiliary_loss_clip": 0.01179889,
      "auxiliary_loss_mlp": 0.01026839,
      "balance_loss_clip": 1.0532608,
      "balance_loss_mlp": 1.0191257,
      "epoch": 0.5858233631936511,
      "flos": 20885682303360.0,
      "grad_norm": 2.9467787526371843,
      "language_loss": 0.83298939,
      "learning_rate": 1.5451995089770624e-06,
      "loss": 0.85505664,
      "num_input_tokens_seen": 105113985,
      "step": 4872,
      "time_per_iteration": 3.5458316802978516
    },
    {
      "auxiliary_loss_clip": 0.01174939,
      "auxiliary_loss_mlp": 0.01027899,
      "balance_loss_clip": 1.0507859,
      "balance_loss_mlp": 1.02061546,
      "epoch": 0.5859436060842903,
      "flos": 23191902000000.0,
      "grad_norm": 1.3800365706913065,
      "language_loss": 0.72057301,
      "learning_rate": 1.5444409798272885e-06,
      "loss": 0.74260139,
      "num_input_tokens_seen": 105138075,
      "step": 4873,
      "time_per_iteration": 2.670271158218384
    },
    {
      "auxiliary_loss_clip": 0.01179907,
      "auxiliary_loss_mlp": 0.01023598,
      "balance_loss_clip": 1.05330265,
      "balance_loss_mlp": 1.01583147,
      "epoch": 0.5860638489749294,
      "flos": 22492648961280.0,
      "grad_norm": 2.900161497476379,
      "language_loss": 0.80520135,
      "learning_rate": 1.543682519785456e-06,
      "loss": 0.82723641,
      "num_input_tokens_seen": 105156555,
      "step": 4874,
      "time_per_iteration": 2.6269545555114746
    },
    {
      "auxiliary_loss_clip": 0.0117647,
      "auxiliary_loss_mlp": 0.01021319,
      "balance_loss_clip": 1.05186033,
      "balance_loss_mlp": 1.01421142,
      "epoch": 0.5861840918655684,
      "flos": 17566243764480.0,
      "grad_norm": 2.3959722491943634,
      "language_loss": 0.8018322,
      "learning_rate": 1.5429241289666219e-06,
      "loss": 0.8238101,
      "num_input_tokens_seen": 105174055,
      "step": 4875,
      "time_per_iteration": 2.576286554336548
    },
    {
      "auxiliary_loss_clip": 0.01174717,
      "auxiliary_loss_mlp": 0.01026075,
      "balance_loss_clip": 1.05195498,
      "balance_loss_mlp": 1.01861811,
      "epoch": 0.5863043347562076,
      "flos": 25556152118400.0,
      "grad_norm": 2.2312006576131984,
      "language_loss": 0.70097649,
      "learning_rate": 1.5421658074858342e-06,
      "loss": 0.72298437,
      "num_input_tokens_seen": 105192160,
      "step": 4876,
      "time_per_iteration": 2.6581575870513916
    },
    {
      "auxiliary_loss_clip": 0.01183361,
      "auxiliary_loss_mlp": 0.01034073,
      "balance_loss_clip": 1.0549649,
      "balance_loss_mlp": 1.02629805,
      "epoch": 0.5864245776468466,
      "flos": 20667525050880.0,
      "grad_norm": 2.2481324938332805,
      "language_loss": 0.66463339,
      "learning_rate": 1.5414075554581298e-06,
      "loss": 0.68680775,
      "num_input_tokens_seen": 105210205,
      "step": 4877,
      "time_per_iteration": 2.5864450931549072
    },
    {
      "auxiliary_loss_clip": 0.01177992,
      "auxiliary_loss_mlp": 0.01028634,
      "balance_loss_clip": 1.05109358,
      "balance_loss_mlp": 1.02049804,
      "epoch": 0.5865448205374857,
      "flos": 28913907490560.0,
      "grad_norm": 2.0704945747344508,
      "language_loss": 0.78857219,
      "learning_rate": 1.5406493729985348e-06,
      "loss": 0.81063855,
      "num_input_tokens_seen": 105229400,
      "step": 4878,
      "time_per_iteration": 2.8111093044281006
    },
    {
      "auxiliary_loss_clip": 0.01180519,
      "auxiliary_loss_mlp": 0.01024201,
      "balance_loss_clip": 1.05378437,
      "balance_loss_mlp": 1.01571965,
      "epoch": 0.5866650634281249,
      "flos": 25842575168640.0,
      "grad_norm": 10.617970432626604,
      "language_loss": 0.7202493,
      "learning_rate": 1.5398912602220644e-06,
      "loss": 0.74229646,
      "num_input_tokens_seen": 105248675,
      "step": 4879,
      "time_per_iteration": 2.80692195892334
    },
    {
      "auxiliary_loss_clip": 0.01177805,
      "auxiliary_loss_mlp": 0.01024657,
      "balance_loss_clip": 1.05196249,
      "balance_loss_mlp": 1.01653326,
      "epoch": 0.5867853063187639,
      "flos": 17052325925760.0,
      "grad_norm": 2.9038318433907198,
      "language_loss": 0.78906494,
      "learning_rate": 1.539133217243724e-06,
      "loss": 0.81108958,
      "num_input_tokens_seen": 105265695,
      "step": 4880,
      "time_per_iteration": 2.6548728942871094
    },
    {
      "auxiliary_loss_clip": 0.01179712,
      "auxiliary_loss_mlp": 0.01027345,
      "balance_loss_clip": 1.05185366,
      "balance_loss_mlp": 1.01812446,
      "epoch": 0.586905549209403,
      "flos": 24645026707200.0,
      "grad_norm": 2.4250811924773426,
      "language_loss": 0.76217955,
      "learning_rate": 1.5383752441785081e-06,
      "loss": 0.78425014,
      "num_input_tokens_seen": 105284920,
      "step": 4881,
      "time_per_iteration": 2.658369302749634
    },
    {
      "auxiliary_loss_clip": 0.01180204,
      "auxiliary_loss_mlp": 0.01028128,
      "balance_loss_clip": 1.05436134,
      "balance_loss_mlp": 1.02044749,
      "epoch": 0.5870257921000421,
      "flos": 14720538723840.0,
      "grad_norm": 2.6889298903419006,
      "language_loss": 0.85688221,
      "learning_rate": 1.5376173411414003e-06,
      "loss": 0.87896556,
      "num_input_tokens_seen": 105302960,
      "step": 4882,
      "time_per_iteration": 2.6328847408294678
    },
    {
      "auxiliary_loss_clip": 0.01178405,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.05288601,
      "balance_loss_mlp": 1.02082682,
      "epoch": 0.5871460349906812,
      "flos": 23914998691200.0,
      "grad_norm": 2.012876863479698,
      "language_loss": 0.78986049,
      "learning_rate": 1.5368595082473753e-06,
      "loss": 0.81193602,
      "num_input_tokens_seen": 105321260,
      "step": 4883,
      "time_per_iteration": 2.626833915710449
    },
    {
      "auxiliary_loss_clip": 0.01175365,
      "auxiliary_loss_mlp": 0.01023223,
      "balance_loss_clip": 1.05171263,
      "balance_loss_mlp": 1.01563835,
      "epoch": 0.5872662778813202,
      "flos": 22164174063360.0,
      "grad_norm": 9.319962935331164,
      "language_loss": 0.78357106,
      "learning_rate": 1.5361017456113935e-06,
      "loss": 0.80555695,
      "num_input_tokens_seen": 105341610,
      "step": 4884,
      "time_per_iteration": 2.6035828590393066
    },
    {
      "auxiliary_loss_clip": 0.01178005,
      "auxiliary_loss_mlp": 0.01023753,
      "balance_loss_clip": 1.05191231,
      "balance_loss_mlp": 1.01547432,
      "epoch": 0.5873865207719594,
      "flos": 18441925430400.0,
      "grad_norm": 2.1384949250847964,
      "language_loss": 0.86274505,
      "learning_rate": 1.5353440533484085e-06,
      "loss": 0.88476264,
      "num_input_tokens_seen": 105360465,
      "step": 4885,
      "time_per_iteration": 2.6077325344085693
    },
    {
      "auxiliary_loss_clip": 0.01178056,
      "auxiliary_loss_mlp": 0.01027318,
      "balance_loss_clip": 1.05192685,
      "balance_loss_mlp": 1.01862788,
      "epoch": 0.5875067636625985,
      "flos": 54015321427200.0,
      "grad_norm": 3.5334019198670608,
      "language_loss": 0.66274977,
      "learning_rate": 1.534586431573361e-06,
      "loss": 0.68480343,
      "num_input_tokens_seen": 105385405,
      "step": 4886,
      "time_per_iteration": 2.848257303237915
    },
    {
      "auxiliary_loss_clip": 0.01182026,
      "auxiliary_loss_mlp": 0.01029542,
      "balance_loss_clip": 1.05419183,
      "balance_loss_mlp": 1.02044022,
      "epoch": 0.5876270065532375,
      "flos": 27995707100160.0,
      "grad_norm": 2.1925976818534934,
      "language_loss": 0.79177892,
      "learning_rate": 1.5338288804011817e-06,
      "loss": 0.81389457,
      "num_input_tokens_seen": 105404905,
      "step": 4887,
      "time_per_iteration": 2.6770756244659424
    },
    {
      "auxiliary_loss_clip": 0.01177173,
      "auxiliary_loss_mlp": 0.01027758,
      "balance_loss_clip": 1.050897,
      "balance_loss_mlp": 1.01985478,
      "epoch": 0.5877472494438767,
      "flos": 21361462876800.0,
      "grad_norm": 3.7437581307435415,
      "language_loss": 0.70987856,
      "learning_rate": 1.533071399946791e-06,
      "loss": 0.73192787,
      "num_input_tokens_seen": 105423650,
      "step": 4888,
      "time_per_iteration": 2.557332754135132
    },
    {
      "auxiliary_loss_clip": 0.01175932,
      "auxiliary_loss_mlp": 0.01024586,
      "balance_loss_clip": 1.04943836,
      "balance_loss_mlp": 1.0170635,
      "epoch": 0.5878674923345157,
      "flos": 22383013674240.0,
      "grad_norm": 3.917717152949058,
      "language_loss": 0.57340908,
      "learning_rate": 1.5323139903250977e-06,
      "loss": 0.59541428,
      "num_input_tokens_seen": 105444255,
      "step": 4889,
      "time_per_iteration": 2.6794066429138184
    },
    {
      "auxiliary_loss_clip": 0.01179453,
      "auxiliary_loss_mlp": 0.01030469,
      "balance_loss_clip": 1.05535626,
      "balance_loss_mlp": 1.0220412,
      "epoch": 0.5879877352251548,
      "flos": 21868664872320.0,
      "grad_norm": 1.5404615551140532,
      "language_loss": 0.77089918,
      "learning_rate": 1.5315566516510002e-06,
      "loss": 0.79299837,
      "num_input_tokens_seen": 105462425,
      "step": 4890,
      "time_per_iteration": 2.6029715538024902
    },
    {
      "auxiliary_loss_clip": 0.01178421,
      "auxiliary_loss_mlp": 0.01029355,
      "balance_loss_clip": 1.05329871,
      "balance_loss_mlp": 1.02149296,
      "epoch": 0.5881079781157939,
      "flos": 17493811989120.0,
      "grad_norm": 1.9982398771137804,
      "language_loss": 0.67424154,
      "learning_rate": 1.5307993840393857e-06,
      "loss": 0.69631928,
      "num_input_tokens_seen": 105480505,
      "step": 4891,
      "time_per_iteration": 3.636220693588257
    },
    {
      "auxiliary_loss_clip": 0.01177206,
      "auxiliary_loss_mlp": 0.01027354,
      "balance_loss_clip": 1.05140722,
      "balance_loss_mlp": 1.01963544,
      "epoch": 0.588228221006433,
      "flos": 22601853285120.0,
      "grad_norm": 2.342328090549335,
      "language_loss": 0.80929363,
      "learning_rate": 1.530042187605132e-06,
      "loss": 0.83133924,
      "num_input_tokens_seen": 105499760,
      "step": 4892,
      "time_per_iteration": 2.5602269172668457
    },
    {
      "auxiliary_loss_clip": 0.01175093,
      "auxiliary_loss_mlp": 0.01026918,
      "balance_loss_clip": 1.05173182,
      "balance_loss_mlp": 1.01963389,
      "epoch": 0.5883484638970721,
      "flos": 26176939896960.0,
      "grad_norm": 1.5609643243536842,
      "language_loss": 0.84218657,
      "learning_rate": 1.5292850624631044e-06,
      "loss": 0.86420667,
      "num_input_tokens_seen": 105521955,
      "step": 4893,
      "time_per_iteration": 2.644244909286499
    },
    {
      "auxiliary_loss_clip": 0.01178538,
      "auxiliary_loss_mlp": 0.01027154,
      "balance_loss_clip": 1.05291903,
      "balance_loss_mlp": 1.01936698,
      "epoch": 0.5884687067877111,
      "flos": 30443737691520.0,
      "grad_norm": 2.5729631085122153,
      "language_loss": 0.8010053,
      "learning_rate": 1.5285280087281593e-06,
      "loss": 0.82306224,
      "num_input_tokens_seen": 105542685,
      "step": 4894,
      "time_per_iteration": 3.52278995513916
    },
    {
      "auxiliary_loss_clip": 0.01076982,
      "auxiliary_loss_mlp": 0.0100092,
      "balance_loss_clip": 1.01697302,
      "balance_loss_mlp": 0.99963808,
      "epoch": 0.5885889496783503,
      "flos": 70507550580480.0,
      "grad_norm": 0.6418874337907482,
      "language_loss": 0.56597281,
      "learning_rate": 1.5277710265151398e-06,
      "loss": 0.58675182,
      "num_input_tokens_seen": 105612165,
      "step": 4895,
      "time_per_iteration": 3.344661235809326
    },
    {
      "auxiliary_loss_clip": 0.01183114,
      "auxiliary_loss_mlp": 0.01029183,
      "balance_loss_clip": 1.05552769,
      "balance_loss_mlp": 1.02029586,
      "epoch": 0.5887091925689893,
      "flos": 19098767485440.0,
      "grad_norm": 3.022993330339846,
      "language_loss": 0.77455831,
      "learning_rate": 1.5270141159388803e-06,
      "loss": 0.79668128,
      "num_input_tokens_seen": 105629185,
      "step": 4896,
      "time_per_iteration": 2.6019208431243896
    },
    {
      "auxiliary_loss_clip": 0.01174701,
      "auxiliary_loss_mlp": 0.01025537,
      "balance_loss_clip": 1.04868364,
      "balance_loss_mlp": 1.01747298,
      "epoch": 0.5888294354596284,
      "flos": 23294282739840.0,
      "grad_norm": 1.9773690763505702,
      "language_loss": 0.80638152,
      "learning_rate": 1.526257277114203e-06,
      "loss": 0.82838392,
      "num_input_tokens_seen": 105650260,
      "step": 4897,
      "time_per_iteration": 3.514923095703125
    },
    {
      "auxiliary_loss_clip": 0.01176254,
      "auxiliary_loss_mlp": 0.01028357,
      "balance_loss_clip": 1.05177689,
      "balance_loss_mlp": 1.02051258,
      "epoch": 0.5889496783502676,
      "flos": 21981532383360.0,
      "grad_norm": 2.1468101009711935,
      "language_loss": 0.79477429,
      "learning_rate": 1.5255005101559201e-06,
      "loss": 0.81682038,
      "num_input_tokens_seen": 105667870,
      "step": 4898,
      "time_per_iteration": 3.5349810123443604
    },
    {
      "auxiliary_loss_clip": 0.01175519,
      "auxiliary_loss_mlp": 0.01020171,
      "balance_loss_clip": 1.05190706,
      "balance_loss_mlp": 1.01277447,
      "epoch": 0.5890699212409066,
      "flos": 21685233093120.0,
      "grad_norm": 3.115991905500501,
      "language_loss": 0.76613402,
      "learning_rate": 1.524743815178833e-06,
      "loss": 0.78809088,
      "num_input_tokens_seen": 105685830,
      "step": 4899,
      "time_per_iteration": 2.603900909423828
    },
    {
      "auxiliary_loss_clip": 0.01173798,
      "auxiliary_loss_mlp": 0.01029426,
      "balance_loss_clip": 1.05013871,
      "balance_loss_mlp": 1.02208281,
      "epoch": 0.5891901641315457,
      "flos": 19464553635840.0,
      "grad_norm": 2.008798230405004,
      "language_loss": 0.80555201,
      "learning_rate": 1.5239871922977315e-06,
      "loss": 0.82758427,
      "num_input_tokens_seen": 105705745,
      "step": 4900,
      "time_per_iteration": 2.5682601928710938
    },
    {
      "auxiliary_loss_clip": 0.01178322,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.05462623,
      "balance_loss_mlp": 1.0203774,
      "epoch": 0.5893104070221848,
      "flos": 19609884063360.0,
      "grad_norm": 2.2356128031614246,
      "language_loss": 0.89772999,
      "learning_rate": 1.523230641627394e-06,
      "loss": 0.91979831,
      "num_input_tokens_seen": 105724730,
      "step": 4901,
      "time_per_iteration": 2.616145610809326
    },
    {
      "auxiliary_loss_clip": 0.01176371,
      "auxiliary_loss_mlp": 0.01021809,
      "balance_loss_clip": 1.05301869,
      "balance_loss_mlp": 1.01460862,
      "epoch": 0.5894306499128239,
      "flos": 29060063930880.0,
      "grad_norm": 1.9427720030669124,
      "language_loss": 0.73152316,
      "learning_rate": 1.5224741632825888e-06,
      "loss": 0.75350499,
      "num_input_tokens_seen": 105744920,
      "step": 4902,
      "time_per_iteration": 2.7376060485839844
    },
    {
      "auxiliary_loss_clip": 0.01181065,
      "auxiliary_loss_mlp": 0.01030056,
      "balance_loss_clip": 1.05423248,
      "balance_loss_mlp": 1.0215801,
      "epoch": 0.589550892803463,
      "flos": 42298890721920.0,
      "grad_norm": 1.8757667004952725,
      "language_loss": 0.69580454,
      "learning_rate": 1.521717757378074e-06,
      "loss": 0.71791577,
      "num_input_tokens_seen": 105765465,
      "step": 4903,
      "time_per_iteration": 2.7691445350646973
    },
    {
      "auxiliary_loss_clip": 0.0117919,
      "auxiliary_loss_mlp": 0.01024605,
      "balance_loss_clip": 1.0527699,
      "balance_loss_mlp": 1.01625729,
      "epoch": 0.5896711356941021,
      "flos": 14137062197760.0,
      "grad_norm": 1.7812896752671916,
      "language_loss": 0.68947053,
      "learning_rate": 1.5209614240285943e-06,
      "loss": 0.71150839,
      "num_input_tokens_seen": 105783120,
      "step": 4904,
      "time_per_iteration": 2.582972526550293
    },
    {
      "auxiliary_loss_clip": 0.01177652,
      "auxiliary_loss_mlp": 0.01026159,
      "balance_loss_clip": 1.05197453,
      "balance_loss_mlp": 1.01814842,
      "epoch": 0.5897913785847412,
      "flos": 17201355454080.0,
      "grad_norm": 6.641616976172932,
      "language_loss": 0.84718776,
      "learning_rate": 1.520205163348887e-06,
      "loss": 0.86922586,
      "num_input_tokens_seen": 105801055,
      "step": 4905,
      "time_per_iteration": 2.587540864944458
    },
    {
      "auxiliary_loss_clip": 0.01076994,
      "auxiliary_loss_mlp": 0.01001543,
      "balance_loss_clip": 1.01697421,
      "balance_loss_mlp": 1.00029087,
      "epoch": 0.5899116214753802,
      "flos": 48794164202880.0,
      "grad_norm": 0.7256469059880558,
      "language_loss": 0.56956935,
      "learning_rate": 1.519448975453674e-06,
      "loss": 0.59035474,
      "num_input_tokens_seen": 105856155,
      "step": 4906,
      "time_per_iteration": 3.03707218170166
    },
    {
      "auxiliary_loss_clip": 0.0118123,
      "auxiliary_loss_mlp": 0.01028202,
      "balance_loss_clip": 1.05559492,
      "balance_loss_mlp": 1.0205605,
      "epoch": 0.5900318643660194,
      "flos": 21103659987840.0,
      "grad_norm": 2.0246346582152146,
      "language_loss": 0.76037061,
      "learning_rate": 1.5186928604576696e-06,
      "loss": 0.78246498,
      "num_input_tokens_seen": 105873350,
      "step": 4907,
      "time_per_iteration": 2.620103597640991
    },
    {
      "auxiliary_loss_clip": 0.01176219,
      "auxiliary_loss_mlp": 0.01028703,
      "balance_loss_clip": 1.05293226,
      "balance_loss_mlp": 1.02154136,
      "epoch": 0.5901521072566585,
      "flos": 21178390233600.0,
      "grad_norm": 2.2931181562991703,
      "language_loss": 0.76902103,
      "learning_rate": 1.5179368184755752e-06,
      "loss": 0.79107022,
      "num_input_tokens_seen": 105891435,
      "step": 4908,
      "time_per_iteration": 2.6113369464874268
    },
    {
      "auxiliary_loss_clip": 0.01176951,
      "auxiliary_loss_mlp": 0.01025688,
      "balance_loss_clip": 1.05319154,
      "balance_loss_mlp": 1.01806438,
      "epoch": 0.5902723501472975,
      "flos": 20225967160320.0,
      "grad_norm": 1.9432442136100974,
      "language_loss": 0.82522678,
      "learning_rate": 1.5171808496220821e-06,
      "loss": 0.8472532,
      "num_input_tokens_seen": 105910190,
      "step": 4909,
      "time_per_iteration": 2.563103199005127
    },
    {
      "auxiliary_loss_clip": 0.01176311,
      "auxiliary_loss_mlp": 0.01024628,
      "balance_loss_clip": 1.05122304,
      "balance_loss_mlp": 1.01681972,
      "epoch": 0.5903925930379367,
      "flos": 22964407211520.0,
      "grad_norm": 2.4684029055860526,
      "language_loss": 0.81575888,
      "learning_rate": 1.5164249540118708e-06,
      "loss": 0.83776826,
      "num_input_tokens_seen": 105929315,
      "step": 4910,
      "time_per_iteration": 2.642763137817383
    },
    {
      "auxiliary_loss_clip": 0.01178229,
      "auxiliary_loss_mlp": 0.01029032,
      "balance_loss_clip": 1.05286729,
      "balance_loss_mlp": 1.02111614,
      "epoch": 0.5905128359285757,
      "flos": 23367720096000.0,
      "grad_norm": 1.6823537907370927,
      "language_loss": 0.8311255,
      "learning_rate": 1.5156691317596093e-06,
      "loss": 0.85319811,
      "num_input_tokens_seen": 105950740,
      "step": 4911,
      "time_per_iteration": 2.5942182540893555
    },
    {
      "auxiliary_loss_clip": 0.0117892,
      "auxiliary_loss_mlp": 0.01027156,
      "balance_loss_clip": 1.05410194,
      "balance_loss_mlp": 1.01906133,
      "epoch": 0.5906330788192148,
      "flos": 28032335994240.0,
      "grad_norm": 2.0206431460433723,
      "language_loss": 0.66912329,
      "learning_rate": 1.5149133829799556e-06,
      "loss": 0.69118404,
      "num_input_tokens_seen": 105968735,
      "step": 4912,
      "time_per_iteration": 2.6028125286102295
    },
    {
      "auxiliary_loss_clip": 0.01176113,
      "auxiliary_loss_mlp": 0.01028771,
      "balance_loss_clip": 1.05172276,
      "balance_loss_mlp": 1.0205394,
      "epoch": 0.590753321709854,
      "flos": 18477943793280.0,
      "grad_norm": 2.096971696611953,
      "language_loss": 0.80348974,
      "learning_rate": 1.5141577077875556e-06,
      "loss": 0.82553852,
      "num_input_tokens_seen": 105986060,
      "step": 4913,
      "time_per_iteration": 2.5179529190063477
    },
    {
      "auxiliary_loss_clip": 0.01179128,
      "auxiliary_loss_mlp": 0.01026513,
      "balance_loss_clip": 1.05421698,
      "balance_loss_mlp": 1.01879096,
      "epoch": 0.590873564600493,
      "flos": 16873706568960.0,
      "grad_norm": 13.328018113661749,
      "language_loss": 0.72253329,
      "learning_rate": 1.5134021062970451e-06,
      "loss": 0.74458975,
      "num_input_tokens_seen": 106004440,
      "step": 4914,
      "time_per_iteration": 2.6454265117645264
    },
    {
      "auxiliary_loss_clip": 0.01176856,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.05406988,
      "balance_loss_mlp": 1.01705432,
      "epoch": 0.5909938074911321,
      "flos": 13516166678400.0,
      "grad_norm": 2.1488111016710105,
      "language_loss": 0.81093132,
      "learning_rate": 1.5126465786230483e-06,
      "loss": 0.83294582,
      "num_input_tokens_seen": 106021215,
      "step": 4915,
      "time_per_iteration": 2.5910773277282715
    },
    {
      "auxiliary_loss_clip": 0.01174832,
      "auxiliary_loss_mlp": 0.01029408,
      "balance_loss_clip": 1.05094814,
      "balance_loss_mlp": 1.02204967,
      "epoch": 0.5911140503817712,
      "flos": 26024067613440.0,
      "grad_norm": 2.3019870424159654,
      "language_loss": 0.81953728,
      "learning_rate": 1.5118911248801787e-06,
      "loss": 0.84157968,
      "num_input_tokens_seen": 106039225,
      "step": 4916,
      "time_per_iteration": 2.6401748657226562
    },
    {
      "auxiliary_loss_clip": 0.01172803,
      "auxiliary_loss_mlp": 0.01028985,
      "balance_loss_clip": 1.04925609,
      "balance_loss_mlp": 1.02161455,
      "epoch": 0.5912342932724103,
      "flos": 23258731253760.0,
      "grad_norm": 2.234258250041989,
      "language_loss": 0.79628193,
      "learning_rate": 1.5111357451830364e-06,
      "loss": 0.81829977,
      "num_input_tokens_seen": 106057920,
      "step": 4917,
      "time_per_iteration": 3.6149866580963135
    },
    {
      "auxiliary_loss_clip": 0.0117333,
      "auxiliary_loss_mlp": 0.01025291,
      "balance_loss_clip": 1.05085182,
      "balance_loss_mlp": 1.01792693,
      "epoch": 0.5913545361630493,
      "flos": 19573039687680.0,
      "grad_norm": 2.24789272855543,
      "language_loss": 0.71078902,
      "learning_rate": 1.5103804396462131e-06,
      "loss": 0.73277521,
      "num_input_tokens_seen": 106077855,
      "step": 4918,
      "time_per_iteration": 2.596020221710205
    },
    {
      "auxiliary_loss_clip": 0.01180033,
      "auxiliary_loss_mlp": 0.0102666,
      "balance_loss_clip": 1.05229962,
      "balance_loss_mlp": 1.017856,
      "epoch": 0.5914747790536885,
      "flos": 26213532877440.0,
      "grad_norm": 4.521518610784576,
      "language_loss": 0.80051076,
      "learning_rate": 1.5096252083842877e-06,
      "loss": 0.82257771,
      "num_input_tokens_seen": 106097065,
      "step": 4919,
      "time_per_iteration": 2.685591220855713
    },
    {
      "auxiliary_loss_clip": 0.01177456,
      "auxiliary_loss_mlp": 0.01028012,
      "balance_loss_clip": 1.05109668,
      "balance_loss_mlp": 1.01954246,
      "epoch": 0.5915950219443276,
      "flos": 27417545786880.0,
      "grad_norm": 2.1394154359593176,
      "language_loss": 0.85513175,
      "learning_rate": 1.5088700515118285e-06,
      "loss": 0.87718642,
      "num_input_tokens_seen": 106116385,
      "step": 4920,
      "time_per_iteration": 3.5172035694122314
    },
    {
      "auxiliary_loss_clip": 0.01177273,
      "auxiliary_loss_mlp": 0.01029742,
      "balance_loss_clip": 1.05230558,
      "balance_loss_mlp": 1.02193069,
      "epoch": 0.5917152648349666,
      "flos": 21907879545600.0,
      "grad_norm": 2.115577570271301,
      "language_loss": 0.66574192,
      "learning_rate": 1.508114969143392e-06,
      "loss": 0.68781209,
      "num_input_tokens_seen": 106136370,
      "step": 4921,
      "time_per_iteration": 2.6293272972106934
    },
    {
      "auxiliary_loss_clip": 0.01174279,
      "auxiliary_loss_mlp": 0.0102307,
      "balance_loss_clip": 1.0517993,
      "balance_loss_mlp": 1.01520777,
      "epoch": 0.5918355077256057,
      "flos": 28109185142400.0,
      "grad_norm": 1.6960494618705213,
      "language_loss": 0.77503055,
      "learning_rate": 1.5073599613935238e-06,
      "loss": 0.79700398,
      "num_input_tokens_seen": 106158490,
      "step": 4922,
      "time_per_iteration": 2.6307132244110107
    },
    {
      "auxiliary_loss_clip": 0.011797,
      "auxiliary_loss_mlp": 0.01028027,
      "balance_loss_clip": 1.0538609,
      "balance_loss_mlp": 1.02031088,
      "epoch": 0.5919557506162448,
      "flos": 28183807647360.0,
      "grad_norm": 1.7381560032086856,
      "language_loss": 0.57636756,
      "learning_rate": 1.5066050283767574e-06,
      "loss": 0.59844488,
      "num_input_tokens_seen": 106179170,
      "step": 4923,
      "time_per_iteration": 2.6204771995544434
    },
    {
      "auxiliary_loss_clip": 0.01177291,
      "auxiliary_loss_mlp": 0.01025355,
      "balance_loss_clip": 1.05463314,
      "balance_loss_mlp": 1.01720679,
      "epoch": 0.5920759935068839,
      "flos": 12094355652480.0,
      "grad_norm": 2.653997982070505,
      "language_loss": 0.82967651,
      "learning_rate": 1.505850170207616e-06,
      "loss": 0.85170299,
      "num_input_tokens_seen": 106196035,
      "step": 4924,
      "time_per_iteration": 3.457054853439331
    },
    {
      "auxiliary_loss_clip": 0.01177645,
      "auxiliary_loss_mlp": 0.01024216,
      "balance_loss_clip": 1.05310333,
      "balance_loss_mlp": 1.01652741,
      "epoch": 0.592196236397523,
      "flos": 29424772673280.0,
      "grad_norm": 2.179783313492548,
      "language_loss": 0.78344274,
      "learning_rate": 1.505095387000611e-06,
      "loss": 0.80546135,
      "num_input_tokens_seen": 106218335,
      "step": 4925,
      "time_per_iteration": 2.736973762512207
    },
    {
      "auxiliary_loss_clip": 0.0117733,
      "auxiliary_loss_mlp": 0.01028182,
      "balance_loss_clip": 1.05267978,
      "balance_loss_mlp": 1.02056479,
      "epoch": 0.5923164792881621,
      "flos": 24384709866240.0,
      "grad_norm": 2.1916576698393535,
      "language_loss": 0.74047607,
      "learning_rate": 1.504340678870242e-06,
      "loss": 0.76253122,
      "num_input_tokens_seen": 106236550,
      "step": 4926,
      "time_per_iteration": 3.5417611598968506
    },
    {
      "auxiliary_loss_clip": 0.01176671,
      "auxiliary_loss_mlp": 0.01027916,
      "balance_loss_clip": 1.05057859,
      "balance_loss_mlp": 1.02015245,
      "epoch": 0.5924367221788012,
      "flos": 24024238928640.0,
      "grad_norm": 2.2341867951586023,
      "language_loss": 0.90056723,
      "learning_rate": 1.5035860459309989e-06,
      "loss": 0.92261314,
      "num_input_tokens_seen": 106254265,
      "step": 4927,
      "time_per_iteration": 2.579444646835327
    },
    {
      "auxiliary_loss_clip": 0.01177702,
      "auxiliary_loss_mlp": 0.01021591,
      "balance_loss_clip": 1.0515573,
      "balance_loss_mlp": 1.01345539,
      "epoch": 0.5925569650694402,
      "flos": 26870590414080.0,
      "grad_norm": 1.8771297778498626,
      "language_loss": 0.64236659,
      "learning_rate": 1.5028314882973568e-06,
      "loss": 0.66435951,
      "num_input_tokens_seen": 106274670,
      "step": 4928,
      "time_per_iteration": 2.632901191711426
    },
    {
      "auxiliary_loss_clip": 0.01176741,
      "auxiliary_loss_mlp": 0.01030061,
      "balance_loss_clip": 1.05083549,
      "balance_loss_mlp": 1.0220623,
      "epoch": 0.5926772079600794,
      "flos": 22302788647680.0,
      "grad_norm": 2.7302177077422165,
      "language_loss": 0.84446865,
      "learning_rate": 1.502077006083783e-06,
      "loss": 0.86653668,
      "num_input_tokens_seen": 106293330,
      "step": 4929,
      "time_per_iteration": 2.5695056915283203
    },
    {
      "auxiliary_loss_clip": 0.01176985,
      "auxiliary_loss_mlp": 0.01027763,
      "balance_loss_clip": 1.05245471,
      "balance_loss_mlp": 1.01960349,
      "epoch": 0.5927974508507184,
      "flos": 19865244827520.0,
      "grad_norm": 2.7381913990276026,
      "language_loss": 0.76649421,
      "learning_rate": 1.5013225994047315e-06,
      "loss": 0.78854167,
      "num_input_tokens_seen": 106310960,
      "step": 4930,
      "time_per_iteration": 2.5857644081115723
    },
    {
      "auxiliary_loss_clip": 0.01175533,
      "auxiliary_loss_mlp": 0.01028675,
      "balance_loss_clip": 1.05169272,
      "balance_loss_mlp": 1.02062285,
      "epoch": 0.5929176937413575,
      "flos": 15776743167360.0,
      "grad_norm": 1.700043106612933,
      "language_loss": 0.80629921,
      "learning_rate": 1.5005682683746452e-06,
      "loss": 0.82834136,
      "num_input_tokens_seen": 106329475,
      "step": 4931,
      "time_per_iteration": 2.5629310607910156
    },
    {
      "auxiliary_loss_clip": 0.01178431,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.05391717,
      "balance_loss_mlp": 1.01865864,
      "epoch": 0.5930379366319967,
      "flos": 17601472028160.0,
      "grad_norm": 2.5763325611388748,
      "language_loss": 0.72837985,
      "learning_rate": 1.4998140131079553e-06,
      "loss": 0.75043249,
      "num_input_tokens_seen": 106345565,
      "step": 4932,
      "time_per_iteration": 2.6270198822021484
    },
    {
      "auxiliary_loss_clip": 0.01174202,
      "auxiliary_loss_mlp": 0.01025903,
      "balance_loss_clip": 1.05232251,
      "balance_loss_mlp": 1.01843131,
      "epoch": 0.5931581795226357,
      "flos": 17704283731200.0,
      "grad_norm": 1.956279762146657,
      "language_loss": 0.7370528,
      "learning_rate": 1.4990598337190821e-06,
      "loss": 0.75905383,
      "num_input_tokens_seen": 106361920,
      "step": 4933,
      "time_per_iteration": 2.535400629043579
    },
    {
      "auxiliary_loss_clip": 0.01177834,
      "auxiliary_loss_mlp": 0.01020858,
      "balance_loss_clip": 1.05246091,
      "balance_loss_mlp": 1.01312733,
      "epoch": 0.5932784224132748,
      "flos": 24280102483200.0,
      "grad_norm": 2.007429094279009,
      "language_loss": 0.68133807,
      "learning_rate": 1.4983057303224338e-06,
      "loss": 0.70332503,
      "num_input_tokens_seen": 106381735,
      "step": 4934,
      "time_per_iteration": 2.6445505619049072
    },
    {
      "auxiliary_loss_clip": 0.01175023,
      "auxiliary_loss_mlp": 0.01028445,
      "balance_loss_clip": 1.0517205,
      "balance_loss_mlp": 1.02068496,
      "epoch": 0.5933986653039139,
      "flos": 22926700909440.0,
      "grad_norm": 1.980867951961189,
      "language_loss": 0.87764239,
      "learning_rate": 1.4975517030324072e-06,
      "loss": 0.89967704,
      "num_input_tokens_seen": 106399745,
      "step": 4935,
      "time_per_iteration": 2.6002440452575684
    },
    {
      "auxiliary_loss_clip": 0.01074908,
      "auxiliary_loss_mlp": 0.01001809,
      "balance_loss_clip": 1.01508808,
      "balance_loss_mlp": 1.00058675,
      "epoch": 0.593518908194553,
      "flos": 71121730256640.0,
      "grad_norm": 0.7806170481061708,
      "language_loss": 0.61675078,
      "learning_rate": 1.4967977519633882e-06,
      "loss": 0.63751799,
      "num_input_tokens_seen": 106457205,
      "step": 4936,
      "time_per_iteration": 3.269371509552002
    },
    {
      "auxiliary_loss_clip": 0.01181314,
      "auxiliary_loss_mlp": 0.01030375,
      "balance_loss_clip": 1.0534842,
      "balance_loss_mlp": 1.02158284,
      "epoch": 0.593639151085192,
      "flos": 20448649526400.0,
      "grad_norm": 2.3056867582469582,
      "language_loss": 0.78393966,
      "learning_rate": 1.4960438772297494e-06,
      "loss": 0.8060565,
      "num_input_tokens_seen": 106474250,
      "step": 4937,
      "time_per_iteration": 2.6539785861968994
    },
    {
      "auxiliary_loss_clip": 0.01180002,
      "auxiliary_loss_mlp": 0.01030302,
      "balance_loss_clip": 1.0528245,
      "balance_loss_mlp": 1.02220798,
      "epoch": 0.5937593939758312,
      "flos": 30883428074880.0,
      "grad_norm": 2.65245056893996,
      "language_loss": 0.73478186,
      "learning_rate": 1.495290078945855e-06,
      "loss": 0.75688493,
      "num_input_tokens_seen": 106494015,
      "step": 4938,
      "time_per_iteration": 2.6875431537628174
    },
    {
      "auxiliary_loss_clip": 0.01176515,
      "auxiliary_loss_mlp": 0.01032094,
      "balance_loss_clip": 1.05191696,
      "balance_loss_mlp": 1.02469671,
      "epoch": 0.5938796368664703,
      "flos": 36898069668480.0,
      "grad_norm": 1.9177655542300986,
      "language_loss": 0.74720693,
      "learning_rate": 1.4945363572260529e-06,
      "loss": 0.76929301,
      "num_input_tokens_seen": 106515010,
      "step": 4939,
      "time_per_iteration": 2.7432470321655273
    },
    {
      "auxiliary_loss_clip": 0.01178738,
      "auxiliary_loss_mlp": 0.01027181,
      "balance_loss_clip": 1.05311072,
      "balance_loss_mlp": 1.0190208,
      "epoch": 0.5939998797571093,
      "flos": 23842926051840.0,
      "grad_norm": 2.1000524958776587,
      "language_loss": 0.67609334,
      "learning_rate": 1.4937827121846845e-06,
      "loss": 0.69815254,
      "num_input_tokens_seen": 106535265,
      "step": 4940,
      "time_per_iteration": 2.6255059242248535
    },
    {
      "auxiliary_loss_clip": 0.01175991,
      "auxiliary_loss_mlp": 0.01033049,
      "balance_loss_clip": 1.05261707,
      "balance_loss_mlp": 1.0252229,
      "epoch": 0.5941201226477485,
      "flos": 25191407462400.0,
      "grad_norm": 2.2747560804387708,
      "language_loss": 0.73227352,
      "learning_rate": 1.4930291439360755e-06,
      "loss": 0.75436389,
      "num_input_tokens_seen": 106557830,
      "step": 4941,
      "time_per_iteration": 2.646926164627075
    },
    {
      "auxiliary_loss_clip": 0.01180497,
      "auxiliary_loss_mlp": 0.01026298,
      "balance_loss_clip": 1.05354977,
      "balance_loss_mlp": 1.01792097,
      "epoch": 0.5942403655383875,
      "flos": 22418996123520.0,
      "grad_norm": 2.607039127808717,
      "language_loss": 0.79026413,
      "learning_rate": 1.4922756525945427e-06,
      "loss": 0.81233215,
      "num_input_tokens_seen": 106577140,
      "step": 4942,
      "time_per_iteration": 2.5524585247039795
    },
    {
      "auxiliary_loss_clip": 0.01074801,
      "auxiliary_loss_mlp": 0.01003147,
      "balance_loss_clip": 1.01506424,
      "balance_loss_mlp": 1.00193071,
      "epoch": 0.5943606084290266,
      "flos": 67629310796160.0,
      "grad_norm": 0.911010948799586,
      "language_loss": 0.59552634,
      "learning_rate": 1.4915222382743894e-06,
      "loss": 0.61630583,
      "num_input_tokens_seen": 106635975,
      "step": 4943,
      "time_per_iteration": 3.197404384613037
    },
    {
      "auxiliary_loss_clip": 0.01179829,
      "auxiliary_loss_mlp": 0.01029427,
      "balance_loss_clip": 1.05401886,
      "balance_loss_mlp": 1.02110624,
      "epoch": 0.5944808513196658,
      "flos": 18223157646720.0,
      "grad_norm": 2.31745199424626,
      "language_loss": 0.71809435,
      "learning_rate": 1.4907689010899085e-06,
      "loss": 0.74018693,
      "num_input_tokens_seen": 106653555,
      "step": 4944,
      "time_per_iteration": 3.5157182216644287
    },
    {
      "auxiliary_loss_clip": 0.01176622,
      "auxiliary_loss_mlp": 0.01029595,
      "balance_loss_clip": 1.05019677,
      "balance_loss_mlp": 1.02177453,
      "epoch": 0.5946010942103048,
      "flos": 24790824011520.0,
      "grad_norm": 2.1737486044745427,
      "language_loss": 0.62384713,
      "learning_rate": 1.4900156411553804e-06,
      "loss": 0.64590925,
      "num_input_tokens_seen": 106673385,
      "step": 4945,
      "time_per_iteration": 2.6186532974243164
    },
    {
      "auxiliary_loss_clip": 0.01178358,
      "auxiliary_loss_mlp": 0.01032394,
      "balance_loss_clip": 1.05400467,
      "balance_loss_mlp": 1.02410328,
      "epoch": 0.5947213371009439,
      "flos": 15231619388160.0,
      "grad_norm": 2.5726063769076752,
      "language_loss": 0.86015624,
      "learning_rate": 1.4892624585850739e-06,
      "loss": 0.88226378,
      "num_input_tokens_seen": 106691740,
      "step": 4946,
      "time_per_iteration": 2.6431884765625
    },
    {
      "auxiliary_loss_clip": 0.01179099,
      "auxiliary_loss_mlp": 0.01027619,
      "balance_loss_clip": 1.051862,
      "balance_loss_mlp": 1.01961362,
      "epoch": 0.594841579991583,
      "flos": 25848069949440.0,
      "grad_norm": 2.195867665620698,
      "language_loss": 0.79731572,
      "learning_rate": 1.4885093534932465e-06,
      "loss": 0.81938291,
      "num_input_tokens_seen": 106709705,
      "step": 4947,
      "time_per_iteration": 3.679177761077881
    },
    {
      "auxiliary_loss_clip": 0.01180942,
      "auxiliary_loss_mlp": 0.01025736,
      "balance_loss_clip": 1.05489147,
      "balance_loss_mlp": 1.01791024,
      "epoch": 0.5949618228822221,
      "flos": 23981109672960.0,
      "grad_norm": 2.2234296685506316,
      "language_loss": 0.71314609,
      "learning_rate": 1.4877563259941433e-06,
      "loss": 0.73521292,
      "num_input_tokens_seen": 106727560,
      "step": 4948,
      "time_per_iteration": 2.6641604900360107
    },
    {
      "auxiliary_loss_clip": 0.01181511,
      "auxiliary_loss_mlp": 0.01027978,
      "balance_loss_clip": 1.05410099,
      "balance_loss_mlp": 1.01947832,
      "epoch": 0.5950820657728612,
      "flos": 40547491476480.0,
      "grad_norm": 2.9402080335782004,
      "language_loss": 0.68050277,
      "learning_rate": 1.4870033762019988e-06,
      "loss": 0.70259768,
      "num_input_tokens_seen": 106747725,
      "step": 4949,
      "time_per_iteration": 2.7812392711639404
    },
    {
      "auxiliary_loss_clip": 0.01179306,
      "auxiliary_loss_mlp": 0.01030903,
      "balance_loss_clip": 1.05331492,
      "balance_loss_mlp": 1.02257609,
      "epoch": 0.5952023086635003,
      "flos": 23184467884800.0,
      "grad_norm": 2.795376916716358,
      "language_loss": 0.7347002,
      "learning_rate": 1.4862505042310334e-06,
      "loss": 0.75680232,
      "num_input_tokens_seen": 106767010,
      "step": 4950,
      "time_per_iteration": 2.5635831356048584
    },
    {
      "auxiliary_loss_clip": 0.01176883,
      "auxiliary_loss_mlp": 0.01025758,
      "balance_loss_clip": 1.05301654,
      "balance_loss_mlp": 1.01801515,
      "epoch": 0.5953225515541394,
      "flos": 33653289548160.0,
      "grad_norm": 1.7062068890115347,
      "language_loss": 0.69709051,
      "learning_rate": 1.4854977101954587e-06,
      "loss": 0.71911687,
      "num_input_tokens_seen": 106789230,
      "step": 4951,
      "time_per_iteration": 3.612428903579712
    },
    {
      "auxiliary_loss_clip": 0.01179028,
      "auxiliary_loss_mlp": 0.01027744,
      "balance_loss_clip": 1.05222845,
      "balance_loss_mlp": 1.02000451,
      "epoch": 0.5954427944447784,
      "flos": 24459619680000.0,
      "grad_norm": 1.7684095523015524,
      "language_loss": 0.86424458,
      "learning_rate": 1.4847449942094716e-06,
      "loss": 0.88631231,
      "num_input_tokens_seen": 106808110,
      "step": 4952,
      "time_per_iteration": 2.662973403930664
    },
    {
      "auxiliary_loss_clip": 0.01175669,
      "auxiliary_loss_mlp": 0.01029082,
      "balance_loss_clip": 1.05101347,
      "balance_loss_mlp": 1.0216372,
      "epoch": 0.5955630373354175,
      "flos": 18551848026240.0,
      "grad_norm": 2.0846768937718867,
      "language_loss": 0.86129785,
      "learning_rate": 1.4839923563872598e-06,
      "loss": 0.88334537,
      "num_input_tokens_seen": 106826650,
      "step": 4953,
      "time_per_iteration": 3.4898765087127686
    },
    {
      "auxiliary_loss_clip": 0.01176908,
      "auxiliary_loss_mlp": 0.01024105,
      "balance_loss_clip": 1.05114448,
      "balance_loss_mlp": 1.01576042,
      "epoch": 0.5956832802260567,
      "flos": 19791699730560.0,
      "grad_norm": 2.0908074057249535,
      "language_loss": 0.7622534,
      "learning_rate": 1.483239796842997e-06,
      "loss": 0.78426349,
      "num_input_tokens_seen": 106844680,
      "step": 4954,
      "time_per_iteration": 2.6204543113708496
    },
    {
      "auxiliary_loss_clip": 0.01175106,
      "auxiliary_loss_mlp": 0.0103093,
      "balance_loss_clip": 1.05364513,
      "balance_loss_mlp": 1.02323842,
      "epoch": 0.5958035231166957,
      "flos": 19750868945280.0,
      "grad_norm": 2.2292692283769013,
      "language_loss": 0.83747494,
      "learning_rate": 1.4824873156908462e-06,
      "loss": 0.85953534,
      "num_input_tokens_seen": 106862605,
      "step": 4955,
      "time_per_iteration": 2.6036298274993896
    },
    {
      "auxiliary_loss_clip": 0.0118077,
      "auxiliary_loss_mlp": 0.01029063,
      "balance_loss_clip": 1.05468607,
      "balance_loss_mlp": 1.02047372,
      "epoch": 0.5959237660073348,
      "flos": 21652806090240.0,
      "grad_norm": 1.5981703165868377,
      "language_loss": 0.7554965,
      "learning_rate": 1.4817349130449584e-06,
      "loss": 0.77759486,
      "num_input_tokens_seen": 106882325,
      "step": 4956,
      "time_per_iteration": 2.613579511642456
    },
    {
      "auxiliary_loss_clip": 0.01172962,
      "auxiliary_loss_mlp": 0.01031201,
      "balance_loss_clip": 1.04929662,
      "balance_loss_mlp": 1.02354503,
      "epoch": 0.5960440088979739,
      "flos": 21171207513600.0,
      "grad_norm": 1.8755833274882605,
      "language_loss": 0.83070159,
      "learning_rate": 1.4809825890194717e-06,
      "loss": 0.85274315,
      "num_input_tokens_seen": 106900995,
      "step": 4957,
      "time_per_iteration": 2.5698635578155518
    },
    {
      "auxiliary_loss_clip": 0.01171545,
      "auxiliary_loss_mlp": 0.01027914,
      "balance_loss_clip": 1.0506084,
      "balance_loss_mlp": 1.01994526,
      "epoch": 0.596164251788613,
      "flos": 14757526753920.0,
      "grad_norm": 1.8740285963986814,
      "language_loss": 0.77212226,
      "learning_rate": 1.4802303437285139e-06,
      "loss": 0.79411685,
      "num_input_tokens_seen": 106918265,
      "step": 4958,
      "time_per_iteration": 2.5907058715820312
    },
    {
      "auxiliary_loss_clip": 0.01177097,
      "auxiliary_loss_mlp": 0.01022983,
      "balance_loss_clip": 1.05114579,
      "balance_loss_mlp": 1.01543665,
      "epoch": 0.596284494679252,
      "flos": 20485924865280.0,
      "grad_norm": 4.662012156590281,
      "language_loss": 0.80508673,
      "learning_rate": 1.4794781772861994e-06,
      "loss": 0.82708752,
      "num_input_tokens_seen": 106934760,
      "step": 4959,
      "time_per_iteration": 2.5775065422058105
    },
    {
      "auxiliary_loss_clip": 0.01176296,
      "auxiliary_loss_mlp": 0.01025468,
      "balance_loss_clip": 1.0531292,
      "balance_loss_mlp": 1.01810431,
      "epoch": 0.5964047375698912,
      "flos": 31212262108800.0,
      "grad_norm": 2.3221966482574694,
      "language_loss": 0.66552007,
      "learning_rate": 1.4787260898066324e-06,
      "loss": 0.68753767,
      "num_input_tokens_seen": 106954760,
      "step": 4960,
      "time_per_iteration": 2.6823253631591797
    },
    {
      "auxiliary_loss_clip": 0.01176959,
      "auxiliary_loss_mlp": 0.01021247,
      "balance_loss_clip": 1.05287969,
      "balance_loss_mlp": 1.01358151,
      "epoch": 0.5965249804605303,
      "flos": 27483620855040.0,
      "grad_norm": 2.3291861162632244,
      "language_loss": 0.85738534,
      "learning_rate": 1.4779740814039023e-06,
      "loss": 0.87936735,
      "num_input_tokens_seen": 106974845,
      "step": 4961,
      "time_per_iteration": 2.651567220687866
    },
    {
      "auxiliary_loss_clip": 0.01175164,
      "auxiliary_loss_mlp": 0.01021654,
      "balance_loss_clip": 1.0512166,
      "balance_loss_mlp": 1.01370239,
      "epoch": 0.5966452233511693,
      "flos": 30773936442240.0,
      "grad_norm": 3.1806113148116157,
      "language_loss": 0.68779242,
      "learning_rate": 1.4772221521920894e-06,
      "loss": 0.70976067,
      "num_input_tokens_seen": 106994870,
      "step": 4962,
      "time_per_iteration": 2.7074661254882812
    },
    {
      "auxiliary_loss_clip": 0.01178942,
      "auxiliary_loss_mlp": 0.01023464,
      "balance_loss_clip": 1.05514431,
      "balance_loss_mlp": 1.01581657,
      "epoch": 0.5967654662418085,
      "flos": 25481170477440.0,
      "grad_norm": 2.0106746156671487,
      "language_loss": 0.74363875,
      "learning_rate": 1.4764703022852598e-06,
      "loss": 0.76566285,
      "num_input_tokens_seen": 107015390,
      "step": 4963,
      "time_per_iteration": 2.652543544769287
    },
    {
      "auxiliary_loss_clip": 0.01173199,
      "auxiliary_loss_mlp": 0.01022245,
      "balance_loss_clip": 1.05099416,
      "balance_loss_mlp": 1.01473475,
      "epoch": 0.5968857091324475,
      "flos": 19099126621440.0,
      "grad_norm": 1.877791221525465,
      "language_loss": 0.77209187,
      "learning_rate": 1.4757185317974696e-06,
      "loss": 0.79404634,
      "num_input_tokens_seen": 107033775,
      "step": 4964,
      "time_per_iteration": 2.618866205215454
    },
    {
      "auxiliary_loss_clip": 0.01177969,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 1.0508455,
      "balance_loss_mlp": 1.01671219,
      "epoch": 0.5970059520230866,
      "flos": 23692711374720.0,
      "grad_norm": 2.498539031955606,
      "language_loss": 0.71265364,
      "learning_rate": 1.474966840842761e-06,
      "loss": 0.73468518,
      "num_input_tokens_seen": 107053355,
      "step": 4965,
      "time_per_iteration": 2.5902469158172607
    },
    {
      "auxiliary_loss_clip": 0.01175223,
      "auxiliary_loss_mlp": 0.01023798,
      "balance_loss_clip": 1.05128145,
      "balance_loss_mlp": 1.01622188,
      "epoch": 0.5971261949137258,
      "flos": 23185545292800.0,
      "grad_norm": 3.349157021523995,
      "language_loss": 0.87303782,
      "learning_rate": 1.4742152295351655e-06,
      "loss": 0.895028,
      "num_input_tokens_seen": 107072510,
      "step": 4966,
      "time_per_iteration": 2.630281448364258
    },
    {
      "auxiliary_loss_clip": 0.01178462,
      "auxiliary_loss_mlp": 0.010256,
      "balance_loss_clip": 1.05092669,
      "balance_loss_mlp": 1.01755285,
      "epoch": 0.5972464378043648,
      "flos": 20557710195840.0,
      "grad_norm": 3.966521449790314,
      "language_loss": 0.63815492,
      "learning_rate": 1.4734636979887016e-06,
      "loss": 0.66019547,
      "num_input_tokens_seen": 107089970,
      "step": 4967,
      "time_per_iteration": 2.5589759349823
    },
    {
      "auxiliary_loss_clip": 0.01176225,
      "auxiliary_loss_mlp": 0.01029555,
      "balance_loss_clip": 1.05027318,
      "balance_loss_mlp": 1.02137697,
      "epoch": 0.5973666806950039,
      "flos": 29387030457600.0,
      "grad_norm": 2.1398762843021877,
      "language_loss": 0.89895654,
      "learning_rate": 1.4727122463173755e-06,
      "loss": 0.92101431,
      "num_input_tokens_seen": 107108500,
      "step": 4968,
      "time_per_iteration": 2.6664459705352783
    },
    {
      "auxiliary_loss_clip": 0.01176535,
      "auxiliary_loss_mlp": 0.01028167,
      "balance_loss_clip": 1.05314136,
      "balance_loss_mlp": 1.01972079,
      "epoch": 0.597486923585643,
      "flos": 22273522041600.0,
      "grad_norm": 1.8600522416383558,
      "language_loss": 0.64717734,
      "learning_rate": 1.471960874635183e-06,
      "loss": 0.66922438,
      "num_input_tokens_seen": 107128060,
      "step": 4969,
      "time_per_iteration": 2.605513572692871
    },
    {
      "auxiliary_loss_clip": 0.01177171,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.05070138,
      "balance_loss_mlp": 1.01883757,
      "epoch": 0.5976071664762821,
      "flos": 13772461196160.0,
      "grad_norm": 2.5779496894505454,
      "language_loss": 0.7015329,
      "learning_rate": 1.4712095830561055e-06,
      "loss": 0.72357357,
      "num_input_tokens_seen": 107146550,
      "step": 4970,
      "time_per_iteration": 3.577730417251587
    },
    {
      "auxiliary_loss_clip": 0.01176895,
      "auxiliary_loss_mlp": 0.01027177,
      "balance_loss_clip": 1.05219054,
      "balance_loss_mlp": 1.01966119,
      "epoch": 0.5977274093669211,
      "flos": 19098623831040.0,
      "grad_norm": 2.116828523969225,
      "language_loss": 0.81176233,
      "learning_rate": 1.4704583716941147e-06,
      "loss": 0.833803,
      "num_input_tokens_seen": 107165415,
      "step": 4971,
      "time_per_iteration": 2.6083993911743164
    },
    {
      "auxiliary_loss_clip": 0.01178453,
      "auxiliary_loss_mlp": 0.01025773,
      "balance_loss_clip": 1.05455232,
      "balance_loss_mlp": 1.01806641,
      "epoch": 0.5978476522575603,
      "flos": 20376002269440.0,
      "grad_norm": 2.9169773324788504,
      "language_loss": 0.7245872,
      "learning_rate": 1.4697072406631672e-06,
      "loss": 0.74662948,
      "num_input_tokens_seen": 107185320,
      "step": 4972,
      "time_per_iteration": 2.5493686199188232
    },
    {
      "auxiliary_loss_clip": 0.01181485,
      "auxiliary_loss_mlp": 0.01030515,
      "balance_loss_clip": 1.05336452,
      "balance_loss_mlp": 1.02187228,
      "epoch": 0.5979678951481994,
      "flos": 29023147728000.0,
      "grad_norm": 1.935463792994196,
      "language_loss": 0.72923803,
      "learning_rate": 1.4689561900772097e-06,
      "loss": 0.75135803,
      "num_input_tokens_seen": 107205380,
      "step": 4973,
      "time_per_iteration": 2.6878483295440674
    },
    {
      "auxiliary_loss_clip": 0.01177216,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.05129683,
      "balance_loss_mlp": 1.01871824,
      "epoch": 0.5980881380388384,
      "flos": 17967689141760.0,
      "grad_norm": 2.4760354627579746,
      "language_loss": 0.72561467,
      "learning_rate": 1.4682052200501758e-06,
      "loss": 0.74764884,
      "num_input_tokens_seen": 107222585,
      "step": 4974,
      "time_per_iteration": 3.487170457839966
    },
    {
      "auxiliary_loss_clip": 0.01175993,
      "auxiliary_loss_mlp": 0.01023261,
      "balance_loss_clip": 1.0518105,
      "balance_loss_mlp": 1.01574206,
      "epoch": 0.5982083809294776,
      "flos": 22962827013120.0,
      "grad_norm": 1.8590606792823372,
      "language_loss": 0.80304456,
      "learning_rate": 1.4674543306959876e-06,
      "loss": 0.82503712,
      "num_input_tokens_seen": 107242055,
      "step": 4975,
      "time_per_iteration": 2.695566415786743
    },
    {
      "auxiliary_loss_clip": 0.01180987,
      "auxiliary_loss_mlp": 0.0102963,
      "balance_loss_clip": 1.05184758,
      "balance_loss_mlp": 1.02146423,
      "epoch": 0.5983286238201166,
      "flos": 20991941712000.0,
      "grad_norm": 3.2744869305376763,
      "language_loss": 0.84828866,
      "learning_rate": 1.4667035221285535e-06,
      "loss": 0.87039489,
      "num_input_tokens_seen": 107259695,
      "step": 4976,
      "time_per_iteration": 2.6260857582092285
    },
    {
      "auxiliary_loss_clip": 0.01179357,
      "auxiliary_loss_mlp": 0.01028116,
      "balance_loss_clip": 1.05355561,
      "balance_loss_mlp": 1.01995659,
      "epoch": 0.5984488667107557,
      "flos": 28183448511360.0,
      "grad_norm": 1.889174549179206,
      "language_loss": 0.74152732,
      "learning_rate": 1.4659527944617715e-06,
      "loss": 0.76360208,
      "num_input_tokens_seen": 107279640,
      "step": 4977,
      "time_per_iteration": 3.532301902770996
    },
    {
      "auxiliary_loss_clip": 0.01177578,
      "auxiliary_loss_mlp": 0.01022774,
      "balance_loss_clip": 1.05173671,
      "balance_loss_mlp": 1.0148257,
      "epoch": 0.5985691096013949,
      "flos": 16471794314880.0,
      "grad_norm": 2.1628413277727314,
      "language_loss": 0.76424599,
      "learning_rate": 1.465202147809526e-06,
      "loss": 0.78624952,
      "num_input_tokens_seen": 107298135,
      "step": 4978,
      "time_per_iteration": 2.6316020488739014
    },
    {
      "auxiliary_loss_clip": 0.01178817,
      "auxiliary_loss_mlp": 0.01029319,
      "balance_loss_clip": 1.05316699,
      "balance_loss_mlp": 1.02196932,
      "epoch": 0.5986893524920339,
      "flos": 26719046933760.0,
      "grad_norm": 2.0941651082748076,
      "language_loss": 0.76049995,
      "learning_rate": 1.4644515822856888e-06,
      "loss": 0.78258127,
      "num_input_tokens_seen": 107316570,
      "step": 4979,
      "time_per_iteration": 3.4597392082214355
    },
    {
      "auxiliary_loss_clip": 0.01074219,
      "auxiliary_loss_mlp": 0.01000971,
      "balance_loss_clip": 1.01445055,
      "balance_loss_mlp": 0.99968988,
      "epoch": 0.598809595382673,
      "flos": 61608061100160.0,
      "grad_norm": 0.7565268786148344,
      "language_loss": 0.56462139,
      "learning_rate": 1.4637010980041215e-06,
      "loss": 0.58537328,
      "num_input_tokens_seen": 107378680,
      "step": 4980,
      "time_per_iteration": 3.228929042816162
    },
    {
      "auxiliary_loss_clip": 0.01180646,
      "auxiliary_loss_mlp": 0.01028884,
      "balance_loss_clip": 1.05385804,
      "balance_loss_mlp": 1.02085793,
      "epoch": 0.5989298382733121,
      "flos": 11801719549440.0,
      "grad_norm": 2.332486006597508,
      "language_loss": 0.89204079,
      "learning_rate": 1.4629506950786707e-06,
      "loss": 0.91413611,
      "num_input_tokens_seen": 107394860,
      "step": 4981,
      "time_per_iteration": 2.640834093093872
    },
    {
      "auxiliary_loss_clip": 0.0107383,
      "auxiliary_loss_mlp": 0.01000688,
      "balance_loss_clip": 1.01411772,
      "balance_loss_mlp": 0.99941844,
      "epoch": 0.5990500811639512,
      "flos": 60025800021120.0,
      "grad_norm": 0.8143248027131771,
      "language_loss": 0.5609163,
      "learning_rate": 1.4622003736231733e-06,
      "loss": 0.58166158,
      "num_input_tokens_seen": 107453850,
      "step": 4982,
      "time_per_iteration": 3.2637650966644287
    },
    {
      "auxiliary_loss_clip": 0.0117787,
      "auxiliary_loss_mlp": 0.01028478,
      "balance_loss_clip": 1.05129552,
      "balance_loss_mlp": 1.01969218,
      "epoch": 0.5991703240545903,
      "flos": 18222726683520.0,
      "grad_norm": 2.432656863341259,
      "language_loss": 0.80571693,
      "learning_rate": 1.461450133751451e-06,
      "loss": 0.82778037,
      "num_input_tokens_seen": 107471920,
      "step": 4983,
      "time_per_iteration": 2.6625607013702393
    },
    {
      "auxiliary_loss_clip": 0.01177246,
      "auxiliary_loss_mlp": 0.01024247,
      "balance_loss_clip": 1.05293131,
      "balance_loss_mlp": 1.01605177,
      "epoch": 0.5992905669452293,
      "flos": 27709894581120.0,
      "grad_norm": 1.7970816795143507,
      "language_loss": 0.76144159,
      "learning_rate": 1.4606999755773153e-06,
      "loss": 0.78345644,
      "num_input_tokens_seen": 107493125,
      "step": 4984,
      "time_per_iteration": 2.6492419242858887
    },
    {
      "auxiliary_loss_clip": 0.01175788,
      "auxiliary_loss_mlp": 0.01027533,
      "balance_loss_clip": 1.05218065,
      "balance_loss_mlp": 1.02001703,
      "epoch": 0.5994108098358685,
      "flos": 20449008662400.0,
      "grad_norm": 1.5503416628089575,
      "language_loss": 0.8202486,
      "learning_rate": 1.4599498992145643e-06,
      "loss": 0.84228182,
      "num_input_tokens_seen": 107513150,
      "step": 4985,
      "time_per_iteration": 2.6860339641571045
    },
    {
      "auxiliary_loss_clip": 0.01178562,
      "auxiliary_loss_mlp": 0.0102492,
      "balance_loss_clip": 1.05297673,
      "balance_loss_mlp": 1.01711178,
      "epoch": 0.5995310527265075,
      "flos": 22269966595200.0,
      "grad_norm": 2.1716548698217433,
      "language_loss": 0.71594661,
      "learning_rate": 1.4591999047769846e-06,
      "loss": 0.73798138,
      "num_input_tokens_seen": 107532005,
      "step": 4986,
      "time_per_iteration": 2.5867440700531006
    },
    {
      "auxiliary_loss_clip": 0.01179256,
      "auxiliary_loss_mlp": 0.01027544,
      "balance_loss_clip": 1.05388665,
      "balance_loss_mlp": 1.01952744,
      "epoch": 0.5996512956171466,
      "flos": 18916951818240.0,
      "grad_norm": 1.850240788090755,
      "language_loss": 0.7579391,
      "learning_rate": 1.4584499923783486e-06,
      "loss": 0.78000706,
      "num_input_tokens_seen": 107550585,
      "step": 4987,
      "time_per_iteration": 2.653752326965332
    },
    {
      "auxiliary_loss_clip": 0.01173428,
      "auxiliary_loss_mlp": 0.01023668,
      "balance_loss_clip": 1.05025268,
      "balance_loss_mlp": 1.01642609,
      "epoch": 0.5997715385077858,
      "flos": 15370916330880.0,
      "grad_norm": 1.8182465266087933,
      "language_loss": 0.75836718,
      "learning_rate": 1.457700162132419e-06,
      "loss": 0.78033817,
      "num_input_tokens_seen": 107567575,
      "step": 4988,
      "time_per_iteration": 2.565922260284424
    },
    {
      "auxiliary_loss_clip": 0.01174263,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.04968226,
      "balance_loss_mlp": 1.01643372,
      "epoch": 0.5998917813984248,
      "flos": 25264844818560.0,
      "grad_norm": 2.7487526201063113,
      "language_loss": 0.72427636,
      "learning_rate": 1.4569504141529433e-06,
      "loss": 0.74626458,
      "num_input_tokens_seen": 107585410,
      "step": 4989,
      "time_per_iteration": 2.6231324672698975
    },
    {
      "auxiliary_loss_clip": 0.01178753,
      "auxiliary_loss_mlp": 0.01026464,
      "balance_loss_clip": 1.05143368,
      "balance_loss_mlp": 1.01834583,
      "epoch": 0.6000120242890639,
      "flos": 22054502862720.0,
      "grad_norm": 2.7397711170983245,
      "language_loss": 0.72105026,
      "learning_rate": 1.456200748553658e-06,
      "loss": 0.74310243,
      "num_input_tokens_seen": 107603405,
      "step": 4990,
      "time_per_iteration": 2.58186674118042
    },
    {
      "auxiliary_loss_clip": 0.01180497,
      "auxiliary_loss_mlp": 0.01025585,
      "balance_loss_clip": 1.05442655,
      "balance_loss_mlp": 1.01755047,
      "epoch": 0.600132267179703,
      "flos": 29863421562240.0,
      "grad_norm": 1.6061451215662446,
      "language_loss": 0.78799021,
      "learning_rate": 1.455451165448287e-06,
      "loss": 0.81005108,
      "num_input_tokens_seen": 107626060,
      "step": 4991,
      "time_per_iteration": 2.7448229789733887
    },
    {
      "auxiliary_loss_clip": 0.01174462,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.04982901,
      "balance_loss_mlp": 1.01696312,
      "epoch": 0.6002525100703421,
      "flos": 25045358762880.0,
      "grad_norm": 3.4308513503377913,
      "language_loss": 0.73303175,
      "learning_rate": 1.4547016649505407e-06,
      "loss": 0.7550236,
      "num_input_tokens_seen": 107644070,
      "step": 4992,
      "time_per_iteration": 2.6397249698638916
    },
    {
      "auxiliary_loss_clip": 0.01176105,
      "auxiliary_loss_mlp": 0.01026213,
      "balance_loss_clip": 1.05181384,
      "balance_loss_mlp": 1.01826715,
      "epoch": 0.6003727529609811,
      "flos": 20849592113280.0,
      "grad_norm": 4.290695476502926,
      "language_loss": 0.84925395,
      "learning_rate": 1.4539522471741193e-06,
      "loss": 0.87127721,
      "num_input_tokens_seen": 107661495,
      "step": 4993,
      "time_per_iteration": 2.6705291271209717
    },
    {
      "auxiliary_loss_clip": 0.01179407,
      "auxiliary_loss_mlp": 0.01027469,
      "balance_loss_clip": 1.05134392,
      "balance_loss_mlp": 1.01862919,
      "epoch": 0.6004929958516203,
      "flos": 15594604277760.0,
      "grad_norm": 2.823567325723112,
      "language_loss": 0.71188569,
      "learning_rate": 1.4532029122327067e-06,
      "loss": 0.73395443,
      "num_input_tokens_seen": 107678280,
      "step": 4994,
      "time_per_iteration": 2.6803319454193115
    },
    {
      "auxiliary_loss_clip": 0.01176024,
      "auxiliary_loss_mlp": 0.01026962,
      "balance_loss_clip": 1.04999804,
      "balance_loss_mlp": 1.01898086,
      "epoch": 0.6006132387422594,
      "flos": 21763267390080.0,
      "grad_norm": 2.1586185114352534,
      "language_loss": 0.75235939,
      "learning_rate": 1.4524536602399783e-06,
      "loss": 0.77438927,
      "num_input_tokens_seen": 107697370,
      "step": 4995,
      "time_per_iteration": 2.824716091156006
    },
    {
      "auxiliary_loss_clip": 0.01175762,
      "auxiliary_loss_mlp": 0.01020874,
      "balance_loss_clip": 1.05451584,
      "balance_loss_mlp": 1.01293421,
      "epoch": 0.6007334816328984,
      "flos": 22858542852480.0,
      "grad_norm": 1.9459160739049133,
      "language_loss": 0.77362061,
      "learning_rate": 1.4517044913095938e-06,
      "loss": 0.795587,
      "num_input_tokens_seen": 107717790,
      "step": 4996,
      "time_per_iteration": 3.680260419845581
    },
    {
      "auxiliary_loss_clip": 0.01177413,
      "auxiliary_loss_mlp": 0.01023413,
      "balance_loss_clip": 1.05181241,
      "balance_loss_mlp": 1.01505685,
      "epoch": 0.6008537245235376,
      "flos": 28324577047680.0,
      "grad_norm": 2.9361070889490306,
      "language_loss": 0.81660998,
      "learning_rate": 1.4509554055552022e-06,
      "loss": 0.83861822,
      "num_input_tokens_seen": 107738020,
      "step": 4997,
      "time_per_iteration": 2.7182326316833496
    },
    {
      "auxiliary_loss_clip": 0.01180495,
      "auxiliary_loss_mlp": 0.01031698,
      "balance_loss_clip": 1.05585802,
      "balance_loss_mlp": 1.0236752,
      "epoch": 0.6009739674141766,
      "flos": 20886113266560.0,
      "grad_norm": 3.043380864069961,
      "language_loss": 0.84128112,
      "learning_rate": 1.450206403090439e-06,
      "loss": 0.86340308,
      "num_input_tokens_seen": 107756215,
      "step": 4998,
      "time_per_iteration": 2.551417350769043
    },
    {
      "auxiliary_loss_clip": 0.01174136,
      "auxiliary_loss_mlp": 0.01026452,
      "balance_loss_clip": 1.05214572,
      "balance_loss_mlp": 1.01918066,
      "epoch": 0.6010942103048157,
      "flos": 20481004702080.0,
      "grad_norm": 2.926413052322533,
      "language_loss": 0.86823493,
      "learning_rate": 1.4494574840289274e-06,
      "loss": 0.89024079,
      "num_input_tokens_seen": 107773330,
      "step": 4999,
      "time_per_iteration": 2.555901050567627
    },
    {
      "auxiliary_loss_clip": 0.01180117,
      "auxiliary_loss_mlp": 0.01024237,
      "balance_loss_clip": 1.05211234,
      "balance_loss_mlp": 1.01538575,
      "epoch": 0.6012144531954549,
      "flos": 23805973935360.0,
      "grad_norm": 2.9732153136138395,
      "language_loss": 0.73947048,
      "learning_rate": 1.4487086484842782e-06,
      "loss": 0.76151395,
      "num_input_tokens_seen": 107791975,
      "step": 5000,
      "time_per_iteration": 2.6463875770568848
    },
    {
      "auxiliary_loss_clip": 0.01176172,
      "auxiliary_loss_mlp": 0.01030135,
      "balance_loss_clip": 1.05222511,
      "balance_loss_mlp": 1.02208281,
      "epoch": 0.6013346960860939,
      "flos": 18988378012800.0,
      "grad_norm": 2.2781867523912127,
      "language_loss": 0.60480082,
      "learning_rate": 1.4479598965700878e-06,
      "loss": 0.6268639,
      "num_input_tokens_seen": 107809240,
      "step": 5001,
      "time_per_iteration": 3.5461983680725098
    },
    {
      "auxiliary_loss_clip": 0.01177471,
      "auxiliary_loss_mlp": 0.01027496,
      "balance_loss_clip": 1.05316222,
      "balance_loss_mlp": 1.01980066,
      "epoch": 0.601454938976733,
      "flos": 24025316336640.0,
      "grad_norm": 2.536396186544169,
      "language_loss": 0.69232702,
      "learning_rate": 1.4472112283999427e-06,
      "loss": 0.71437663,
      "num_input_tokens_seen": 107827895,
      "step": 5002,
      "time_per_iteration": 2.7286605834960938
    },
    {
      "auxiliary_loss_clip": 0.01172353,
      "auxiliary_loss_mlp": 0.01029295,
      "balance_loss_clip": 1.04948092,
      "balance_loss_mlp": 1.02185631,
      "epoch": 0.6015751818673721,
      "flos": 26427129102720.0,
      "grad_norm": 1.953214343150901,
      "language_loss": 0.69436049,
      "learning_rate": 1.4464626440874143e-06,
      "loss": 0.7163769,
      "num_input_tokens_seen": 107847010,
      "step": 5003,
      "time_per_iteration": 3.540628433227539
    },
    {
      "auxiliary_loss_clip": 0.01180323,
      "auxiliary_loss_mlp": 0.01026046,
      "balance_loss_clip": 1.0538981,
      "balance_loss_mlp": 1.01743901,
      "epoch": 0.6016954247580112,
      "flos": 13115260005120.0,
      "grad_norm": 2.7508617911297493,
      "language_loss": 0.7475698,
      "learning_rate": 1.4457141437460636e-06,
      "loss": 0.76963341,
      "num_input_tokens_seen": 107864235,
      "step": 5004,
      "time_per_iteration": 2.580280303955078
    },
    {
      "auxiliary_loss_clip": 0.01180006,
      "auxiliary_loss_mlp": 0.01029393,
      "balance_loss_clip": 1.05300677,
      "balance_loss_mlp": 1.02053571,
      "epoch": 0.6018156676486502,
      "flos": 23768447201280.0,
      "grad_norm": 1.924088371026613,
      "language_loss": 0.73354506,
      "learning_rate": 1.444965727489436e-06,
      "loss": 0.75563914,
      "num_input_tokens_seen": 107883680,
      "step": 5005,
      "time_per_iteration": 3.5896291732788086
    },
    {
      "auxiliary_loss_clip": 0.01175048,
      "auxiliary_loss_mlp": 0.01030004,
      "balance_loss_clip": 1.05041826,
      "balance_loss_mlp": 1.0219276,
      "epoch": 0.6019359105392894,
      "flos": 26469360518400.0,
      "grad_norm": 2.6603179512599575,
      "language_loss": 0.63330072,
      "learning_rate": 1.444217395431066e-06,
      "loss": 0.65535128,
      "num_input_tokens_seen": 107906220,
      "step": 5006,
      "time_per_iteration": 2.6626455783843994
    },
    {
      "auxiliary_loss_clip": 0.01074257,
      "auxiliary_loss_mlp": 0.01003245,
      "balance_loss_clip": 1.01455069,
      "balance_loss_mlp": 1.00204086,
      "epoch": 0.6020561534299285,
      "flos": 69190849728000.0,
      "grad_norm": 0.8083623559882612,
      "language_loss": 0.55863011,
      "learning_rate": 1.4434691476844755e-06,
      "loss": 0.57940519,
      "num_input_tokens_seen": 107967195,
      "step": 5007,
      "time_per_iteration": 3.228623390197754
    },
    {
      "auxiliary_loss_clip": 0.01173443,
      "auxiliary_loss_mlp": 0.01027329,
      "balance_loss_clip": 1.05101299,
      "balance_loss_mlp": 1.01992261,
      "epoch": 0.6021763963205675,
      "flos": 21835304115840.0,
      "grad_norm": 5.075605811225046,
      "language_loss": 0.67036295,
      "learning_rate": 1.4427209843631729e-06,
      "loss": 0.69237065,
      "num_input_tokens_seen": 107984245,
      "step": 5008,
      "time_per_iteration": 2.691676378250122
    },
    {
      "auxiliary_loss_clip": 0.01173153,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.05097353,
      "balance_loss_mlp": 1.01674139,
      "epoch": 0.6022966392112067,
      "flos": 26578636669440.0,
      "grad_norm": 2.469840934817062,
      "language_loss": 0.81684363,
      "learning_rate": 1.4419729055806534e-06,
      "loss": 0.83881938,
      "num_input_tokens_seen": 108003680,
      "step": 5009,
      "time_per_iteration": 2.7979633808135986
    },
    {
      "auxiliary_loss_clip": 0.01174551,
      "auxiliary_loss_mlp": 0.01032049,
      "balance_loss_clip": 1.05373669,
      "balance_loss_mlp": 1.02455389,
      "epoch": 0.6024168821018457,
      "flos": 20703722981760.0,
      "grad_norm": 2.09002498835765,
      "language_loss": 0.82107878,
      "learning_rate": 1.441224911450401e-06,
      "loss": 0.84314477,
      "num_input_tokens_seen": 108019635,
      "step": 5010,
      "time_per_iteration": 2.648228406906128
    },
    {
      "auxiliary_loss_clip": 0.01182275,
      "auxiliary_loss_mlp": 0.01025956,
      "balance_loss_clip": 1.05568838,
      "balance_loss_mlp": 1.01840734,
      "epoch": 0.6025371249924848,
      "flos": 24680973242880.0,
      "grad_norm": 2.073174362506303,
      "language_loss": 0.8253628,
      "learning_rate": 1.4404770020858851e-06,
      "loss": 0.84744513,
      "num_input_tokens_seen": 108039120,
      "step": 5011,
      "time_per_iteration": 2.6607038974761963
    },
    {
      "auxiliary_loss_clip": 0.01174244,
      "auxiliary_loss_mlp": 0.01024915,
      "balance_loss_clip": 1.05096579,
      "balance_loss_mlp": 1.01763153,
      "epoch": 0.602657367883124,
      "flos": 25955801815680.0,
      "grad_norm": 1.6809056528663668,
      "language_loss": 0.86214429,
      "learning_rate": 1.439729177600563e-06,
      "loss": 0.8841359,
      "num_input_tokens_seen": 108059615,
      "step": 5012,
      "time_per_iteration": 2.672442674636841
    },
    {
      "auxiliary_loss_clip": 0.01177231,
      "auxiliary_loss_mlp": 0.01024515,
      "balance_loss_clip": 1.05224621,
      "balance_loss_mlp": 1.01702833,
      "epoch": 0.602777610773763,
      "flos": 16690633925760.0,
      "grad_norm": 3.470852466493991,
      "language_loss": 0.73190975,
      "learning_rate": 1.4389814381078793e-06,
      "loss": 0.75392717,
      "num_input_tokens_seen": 108078855,
      "step": 5013,
      "time_per_iteration": 2.6500842571258545
    },
    {
      "auxiliary_loss_clip": 0.01175628,
      "auxiliary_loss_mlp": 0.01034003,
      "balance_loss_clip": 1.05214679,
      "balance_loss_mlp": 1.02604544,
      "epoch": 0.6028978536644021,
      "flos": 13334243270400.0,
      "grad_norm": 2.2596187397621983,
      "language_loss": 0.79784334,
      "learning_rate": 1.438233783721265e-06,
      "loss": 0.81993967,
      "num_input_tokens_seen": 108095020,
      "step": 5014,
      "time_per_iteration": 2.6423583030700684
    },
    {
      "auxiliary_loss_clip": 0.01176944,
      "auxiliary_loss_mlp": 0.01029438,
      "balance_loss_clip": 1.05436122,
      "balance_loss_mlp": 1.02232409,
      "epoch": 0.6030180965550412,
      "flos": 19644825018240.0,
      "grad_norm": 2.8725347543643913,
      "language_loss": 0.77959883,
      "learning_rate": 1.43748621455414e-06,
      "loss": 0.80166268,
      "num_input_tokens_seen": 108111455,
      "step": 5015,
      "time_per_iteration": 2.5899147987365723
    },
    {
      "auxiliary_loss_clip": 0.01176456,
      "auxiliary_loss_mlp": 0.01029283,
      "balance_loss_clip": 1.05087757,
      "balance_loss_mlp": 1.02160037,
      "epoch": 0.6031383394456803,
      "flos": 14458390289280.0,
      "grad_norm": 3.069254326476987,
      "language_loss": 0.81159079,
      "learning_rate": 1.4367387307199082e-06,
      "loss": 0.83364826,
      "num_input_tokens_seen": 108128305,
      "step": 5016,
      "time_per_iteration": 2.612771511077881
    },
    {
      "auxiliary_loss_clip": 0.01176517,
      "auxiliary_loss_mlp": 0.01024328,
      "balance_loss_clip": 1.05263305,
      "balance_loss_mlp": 1.01707375,
      "epoch": 0.6032585823363193,
      "flos": 13917791623680.0,
      "grad_norm": 3.4829736238922617,
      "language_loss": 0.82343113,
      "learning_rate": 1.4359913323319632e-06,
      "loss": 0.84543961,
      "num_input_tokens_seen": 108145475,
      "step": 5017,
      "time_per_iteration": 2.6078083515167236
    },
    {
      "auxiliary_loss_clip": 0.01175159,
      "auxiliary_loss_mlp": 0.0103038,
      "balance_loss_clip": 1.05185747,
      "balance_loss_mlp": 1.022035,
      "epoch": 0.6033788252269584,
      "flos": 24353252530560.0,
      "grad_norm": 2.343014222755637,
      "language_loss": 0.7796911,
      "learning_rate": 1.4352440195036847e-06,
      "loss": 0.80174649,
      "num_input_tokens_seen": 108165650,
      "step": 5018,
      "time_per_iteration": 2.816455125808716
    },
    {
      "auxiliary_loss_clip": 0.01175663,
      "auxiliary_loss_mlp": 0.01027873,
      "balance_loss_clip": 1.05120659,
      "balance_loss_mlp": 1.02011824,
      "epoch": 0.6034990681175976,
      "flos": 25521247077120.0,
      "grad_norm": 1.6418282644544262,
      "language_loss": 0.80187172,
      "learning_rate": 1.4344967923484395e-06,
      "loss": 0.82390714,
      "num_input_tokens_seen": 108187620,
      "step": 5019,
      "time_per_iteration": 2.6123695373535156
    },
    {
      "auxiliary_loss_clip": 0.01174564,
      "auxiliary_loss_mlp": 0.01023825,
      "balance_loss_clip": 1.05062699,
      "balance_loss_mlp": 1.01613045,
      "epoch": 0.6036193110082366,
      "flos": 25958387594880.0,
      "grad_norm": 2.9730292136617513,
      "language_loss": 0.71846139,
      "learning_rate": 1.433749650979581e-06,
      "loss": 0.74044532,
      "num_input_tokens_seen": 108207605,
      "step": 5020,
      "time_per_iteration": 2.708888530731201
    },
    {
      "auxiliary_loss_clip": 0.01173417,
      "auxiliary_loss_mlp": 0.0102133,
      "balance_loss_clip": 1.04922366,
      "balance_loss_mlp": 1.01380169,
      "epoch": 0.6037395538988757,
      "flos": 25593427457280.0,
      "grad_norm": 2.2791340084661456,
      "language_loss": 0.68009466,
      "learning_rate": 1.433002595510451e-06,
      "loss": 0.70204216,
      "num_input_tokens_seen": 108226385,
      "step": 5021,
      "time_per_iteration": 2.7087457180023193
    },
    {
      "auxiliary_loss_clip": 0.01175657,
      "auxiliary_loss_mlp": 0.01024793,
      "balance_loss_clip": 1.05040383,
      "balance_loss_mlp": 1.01683021,
      "epoch": 0.6038597967895148,
      "flos": 17816253402240.0,
      "grad_norm": 2.0994337268423737,
      "language_loss": 0.72100335,
      "learning_rate": 1.4322556260543757e-06,
      "loss": 0.74300784,
      "num_input_tokens_seen": 108242960,
      "step": 5022,
      "time_per_iteration": 2.6019978523254395
    },
    {
      "auxiliary_loss_clip": 0.01073707,
      "auxiliary_loss_mlp": 0.01004261,
      "balance_loss_clip": 1.01398635,
      "balance_loss_mlp": 1.00307488,
      "epoch": 0.6039800396801539,
      "flos": 65169213235200.0,
      "grad_norm": 0.9230287953319276,
      "language_loss": 0.62650764,
      "learning_rate": 1.4315087427246703e-06,
      "loss": 0.64728737,
      "num_input_tokens_seen": 108296785,
      "step": 5023,
      "time_per_iteration": 3.9688808917999268
    },
    {
      "auxiliary_loss_clip": 0.0107384,
      "auxiliary_loss_mlp": 0.01005555,
      "balance_loss_clip": 1.01418686,
      "balance_loss_mlp": 1.00438082,
      "epoch": 0.604100282570793,
      "flos": 67386409073280.0,
      "grad_norm": 0.8711550070728244,
      "language_loss": 0.58468437,
      "learning_rate": 1.4307619456346372e-06,
      "loss": 0.60547829,
      "num_input_tokens_seen": 108341090,
      "step": 5024,
      "time_per_iteration": 2.8960399627685547
    },
    {
      "auxiliary_loss_clip": 0.01172711,
      "auxiliary_loss_mlp": 0.01023281,
      "balance_loss_clip": 1.04851162,
      "balance_loss_mlp": 1.01502848,
      "epoch": 0.6042205254614321,
      "flos": 35297495631360.0,
      "grad_norm": 6.325610960864422,
      "language_loss": 0.740767,
      "learning_rate": 1.430015234897564e-06,
      "loss": 0.7627269,
      "num_input_tokens_seen": 108364370,
      "step": 5025,
      "time_per_iteration": 2.779737710952759
    },
    {
      "auxiliary_loss_clip": 0.01174561,
      "auxiliary_loss_mlp": 0.01024642,
      "balance_loss_clip": 1.04994416,
      "balance_loss_mlp": 1.01714349,
      "epoch": 0.6043407683520712,
      "flos": 45658262206080.0,
      "grad_norm": 2.049874201011214,
      "language_loss": 0.66379219,
      "learning_rate": 1.4292686106267274e-06,
      "loss": 0.68578422,
      "num_input_tokens_seen": 108387220,
      "step": 5026,
      "time_per_iteration": 2.826047420501709
    },
    {
      "auxiliary_loss_clip": 0.01178013,
      "auxiliary_loss_mlp": 0.01029965,
      "balance_loss_clip": 1.05316091,
      "balance_loss_mlp": 1.02200818,
      "epoch": 0.6044610112427102,
      "flos": 16180020138240.0,
      "grad_norm": 1.7357901745870972,
      "language_loss": 0.77017939,
      "learning_rate": 1.4285220729353876e-06,
      "loss": 0.79225922,
      "num_input_tokens_seen": 108405760,
      "step": 5027,
      "time_per_iteration": 2.635364532470703
    },
    {
      "auxiliary_loss_clip": 0.01173661,
      "auxiliary_loss_mlp": 0.01025702,
      "balance_loss_clip": 1.0503155,
      "balance_loss_mlp": 1.01766098,
      "epoch": 0.6045812541333494,
      "flos": 13804062186240.0,
      "grad_norm": 1.8554166703991852,
      "language_loss": 0.77807575,
      "learning_rate": 1.4277756219367957e-06,
      "loss": 0.80006945,
      "num_input_tokens_seen": 108422785,
      "step": 5028,
      "time_per_iteration": 3.5974645614624023
    },
    {
      "auxiliary_loss_clip": 0.01179656,
      "auxiliary_loss_mlp": 0.01027257,
      "balance_loss_clip": 1.05339825,
      "balance_loss_mlp": 1.01959157,
      "epoch": 0.6047014970239885,
      "flos": 19975059682560.0,
      "grad_norm": 2.2687313538910594,
      "language_loss": 0.79494679,
      "learning_rate": 1.4270292577441864e-06,
      "loss": 0.81701601,
      "num_input_tokens_seen": 108442290,
      "step": 5029,
      "time_per_iteration": 2.575387954711914
    },
    {
      "auxiliary_loss_clip": 0.01176448,
      "auxiliary_loss_mlp": 0.01024836,
      "balance_loss_clip": 1.05105543,
      "balance_loss_mlp": 1.01665235,
      "epoch": 0.6048217399146275,
      "flos": 25337097025920.0,
      "grad_norm": 1.8529556534015323,
      "language_loss": 0.71928555,
      "learning_rate": 1.4262829804707836e-06,
      "loss": 0.74129838,
      "num_input_tokens_seen": 108464280,
      "step": 5030,
      "time_per_iteration": 3.5842039585113525
    },
    {
      "auxiliary_loss_clip": 0.01179094,
      "auxiliary_loss_mlp": 0.01026952,
      "balance_loss_clip": 1.05258608,
      "balance_loss_mlp": 1.01895308,
      "epoch": 0.6049419828052667,
      "flos": 26030819370240.0,
      "grad_norm": 1.434057250224776,
      "language_loss": 0.69921875,
      "learning_rate": 1.4255367902297958e-06,
      "loss": 0.72127914,
      "num_input_tokens_seen": 108485610,
      "step": 5031,
      "time_per_iteration": 3.53304123878479
    },
    {
      "auxiliary_loss_clip": 0.01173302,
      "auxiliary_loss_mlp": 0.01026191,
      "balance_loss_clip": 1.05097282,
      "balance_loss_mlp": 1.01862669,
      "epoch": 0.6050622256959057,
      "flos": 14648106948480.0,
      "grad_norm": 2.4378132710202167,
      "language_loss": 0.78806353,
      "learning_rate": 1.4247906871344215e-06,
      "loss": 0.81005847,
      "num_input_tokens_seen": 108501005,
      "step": 5032,
      "time_per_iteration": 2.587848663330078
    },
    {
      "auxiliary_loss_clip": 0.01170473,
      "auxiliary_loss_mlp": 0.01020739,
      "balance_loss_clip": 1.0484097,
      "balance_loss_mlp": 1.01329184,
      "epoch": 0.6051824685865448,
      "flos": 23331450337920.0,
      "grad_norm": 2.0267201667861383,
      "language_loss": 0.75233388,
      "learning_rate": 1.4240446712978415e-06,
      "loss": 0.77424604,
      "num_input_tokens_seen": 108519990,
      "step": 5033,
      "time_per_iteration": 2.69777774810791
    },
    {
      "auxiliary_loss_clip": 0.01176391,
      "auxiliary_loss_mlp": 0.01026632,
      "balance_loss_clip": 1.05227709,
      "balance_loss_mlp": 1.01859713,
      "epoch": 0.605302711477184,
      "flos": 27563307177600.0,
      "grad_norm": 1.952442069069523,
      "language_loss": 0.74274576,
      "learning_rate": 1.423298742833227e-06,
      "loss": 0.76477599,
      "num_input_tokens_seen": 108538650,
      "step": 5034,
      "time_per_iteration": 2.6429176330566406
    },
    {
      "auxiliary_loss_clip": 0.01175263,
      "auxiliary_loss_mlp": 0.01025799,
      "balance_loss_clip": 1.05019128,
      "balance_loss_mlp": 1.01791334,
      "epoch": 0.605422954367823,
      "flos": 15154698412800.0,
      "grad_norm": 4.973925466967652,
      "language_loss": 0.71574068,
      "learning_rate": 1.4225529018537352e-06,
      "loss": 0.73775125,
      "num_input_tokens_seen": 108554155,
      "step": 5035,
      "time_per_iteration": 2.6105988025665283
    },
    {
      "auxiliary_loss_clip": 0.01175364,
      "auxiliary_loss_mlp": 0.01027334,
      "balance_loss_clip": 1.05031824,
      "balance_loss_mlp": 1.0196805,
      "epoch": 0.6055431972584621,
      "flos": 27673912131840.0,
      "grad_norm": 1.742283602490978,
      "language_loss": 0.77920234,
      "learning_rate": 1.4218071484725082e-06,
      "loss": 0.80122936,
      "num_input_tokens_seen": 108576275,
      "step": 5036,
      "time_per_iteration": 2.653186798095703
    },
    {
      "auxiliary_loss_clip": 0.01179785,
      "auxiliary_loss_mlp": 0.01030008,
      "balance_loss_clip": 1.05386472,
      "balance_loss_mlp": 1.02167535,
      "epoch": 0.6056634401491012,
      "flos": 19387489006080.0,
      "grad_norm": 19.959431739369546,
      "language_loss": 0.765661,
      "learning_rate": 1.4210614828026786e-06,
      "loss": 0.78775901,
      "num_input_tokens_seen": 108594125,
      "step": 5037,
      "time_per_iteration": 2.6256637573242188
    },
    {
      "auxiliary_loss_clip": 0.01173606,
      "auxiliary_loss_mlp": 0.01022556,
      "balance_loss_clip": 1.04991674,
      "balance_loss_mlp": 1.01491451,
      "epoch": 0.6057836830397403,
      "flos": 24789459294720.0,
      "grad_norm": 1.5943336698732948,
      "language_loss": 0.74768114,
      "learning_rate": 1.4203159049573605e-06,
      "loss": 0.76964277,
      "num_input_tokens_seen": 108615360,
      "step": 5038,
      "time_per_iteration": 2.614166021347046
    },
    {
      "auxiliary_loss_clip": 0.01174228,
      "auxiliary_loss_mlp": 0.01030978,
      "balance_loss_clip": 1.05026567,
      "balance_loss_mlp": 1.0227468,
      "epoch": 0.6059039259303793,
      "flos": 20558248899840.0,
      "grad_norm": 2.1456627549320957,
      "language_loss": 0.87069231,
      "learning_rate": 1.4195704150496593e-06,
      "loss": 0.89274436,
      "num_input_tokens_seen": 108633075,
      "step": 5039,
      "time_per_iteration": 2.6539788246154785
    },
    {
      "auxiliary_loss_clip": 0.01176145,
      "auxiliary_loss_mlp": 0.01022399,
      "balance_loss_clip": 1.05215561,
      "balance_loss_mlp": 1.01504087,
      "epoch": 0.6060241688210185,
      "flos": 21069724613760.0,
      "grad_norm": 1.737171682821906,
      "language_loss": 0.73974341,
      "learning_rate": 1.4188250131926639e-06,
      "loss": 0.76172888,
      "num_input_tokens_seen": 108651875,
      "step": 5040,
      "time_per_iteration": 2.617994785308838
    },
    {
      "auxiliary_loss_clip": 0.01178779,
      "auxiliary_loss_mlp": 0.01027832,
      "balance_loss_clip": 1.05185986,
      "balance_loss_mlp": 1.01978517,
      "epoch": 0.6061444117116576,
      "flos": 16361081619840.0,
      "grad_norm": 2.4476884376089116,
      "language_loss": 0.80691707,
      "learning_rate": 1.4180796994994525e-06,
      "loss": 0.82898319,
      "num_input_tokens_seen": 108669290,
      "step": 5041,
      "time_per_iteration": 2.5909314155578613
    },
    {
      "auxiliary_loss_clip": 0.01175339,
      "auxiliary_loss_mlp": 0.01022653,
      "balance_loss_clip": 1.05082524,
      "balance_loss_mlp": 1.01528835,
      "epoch": 0.6062646546022966,
      "flos": 21507296094720.0,
      "grad_norm": 1.7872691144310773,
      "language_loss": 0.72100574,
      "learning_rate": 1.4173344740830877e-06,
      "loss": 0.74298567,
      "num_input_tokens_seen": 108688420,
      "step": 5042,
      "time_per_iteration": 2.690927267074585
    },
    {
      "auxiliary_loss_clip": 0.01178318,
      "auxiliary_loss_mlp": 0.01021702,
      "balance_loss_clip": 1.0546689,
      "balance_loss_mlp": 1.01408124,
      "epoch": 0.6063848974929358,
      "flos": 38983151283840.0,
      "grad_norm": 1.6075294477300714,
      "language_loss": 0.70612705,
      "learning_rate": 1.4165893370566206e-06,
      "loss": 0.72812724,
      "num_input_tokens_seen": 108712175,
      "step": 5043,
      "time_per_iteration": 2.7712841033935547
    },
    {
      "auxiliary_loss_clip": 0.01177187,
      "auxiliary_loss_mlp": 0.01027279,
      "balance_loss_clip": 1.05147123,
      "balance_loss_mlp": 1.01888657,
      "epoch": 0.6065051403835748,
      "flos": 19646584784640.0,
      "grad_norm": 35.05065569556761,
      "language_loss": 0.78119755,
      "learning_rate": 1.4158442885330865e-06,
      "loss": 0.80324221,
      "num_input_tokens_seen": 108730745,
      "step": 5044,
      "time_per_iteration": 2.5775723457336426
    },
    {
      "auxiliary_loss_clip": 0.0117249,
      "auxiliary_loss_mlp": 0.01027261,
      "balance_loss_clip": 1.04847479,
      "balance_loss_mlp": 1.01901805,
      "epoch": 0.6066253832742139,
      "flos": 23513086437120.0,
      "grad_norm": 2.550361375609033,
      "language_loss": 0.78597003,
      "learning_rate": 1.4150993286255094e-06,
      "loss": 0.80796754,
      "num_input_tokens_seen": 108749995,
      "step": 5045,
      "time_per_iteration": 2.6824095249176025
    },
    {
      "auxiliary_loss_clip": 0.01174615,
      "auxiliary_loss_mlp": 0.0102337,
      "balance_loss_clip": 1.05076361,
      "balance_loss_mlp": 1.01566887,
      "epoch": 0.6067456261648531,
      "flos": 19133708440320.0,
      "grad_norm": 2.457642903519468,
      "language_loss": 0.79921198,
      "learning_rate": 1.4143544574468993e-06,
      "loss": 0.82119191,
      "num_input_tokens_seen": 108768355,
      "step": 5046,
      "time_per_iteration": 2.5609774589538574
    },
    {
      "auxiliary_loss_clip": 0.01175889,
      "auxiliary_loss_mlp": 0.01024661,
      "balance_loss_clip": 1.05186725,
      "balance_loss_mlp": 1.01697218,
      "epoch": 0.6068658690554921,
      "flos": 20520614424960.0,
      "grad_norm": 1.761089577513431,
      "language_loss": 0.82541096,
      "learning_rate": 1.4136096751102523e-06,
      "loss": 0.84741646,
      "num_input_tokens_seen": 108786685,
      "step": 5047,
      "time_per_iteration": 2.5583958625793457
    },
    {
      "auxiliary_loss_clip": 0.01175201,
      "auxiliary_loss_mlp": 0.01026771,
      "balance_loss_clip": 1.05129135,
      "balance_loss_mlp": 1.01923108,
      "epoch": 0.6069861119461312,
      "flos": 27374560185600.0,
      "grad_norm": 2.1756200989196732,
      "language_loss": 0.83151197,
      "learning_rate": 1.4128649817285516e-06,
      "loss": 0.85353172,
      "num_input_tokens_seen": 108804820,
      "step": 5048,
      "time_per_iteration": 2.48199200630188
    },
    {
      "auxiliary_loss_clip": 0.01174124,
      "auxiliary_loss_mlp": 0.01026852,
      "balance_loss_clip": 1.05271244,
      "balance_loss_mlp": 1.01945543,
      "epoch": 0.6071063548367702,
      "flos": 25626500904960.0,
      "grad_norm": 2.8144484369507956,
      "language_loss": 0.63597637,
      "learning_rate": 1.412120377414766e-06,
      "loss": 0.65798616,
      "num_input_tokens_seen": 108825010,
      "step": 5049,
      "time_per_iteration": 2.5304322242736816
    },
    {
      "auxiliary_loss_clip": 0.01174631,
      "auxiliary_loss_mlp": 0.01029848,
      "balance_loss_clip": 1.05054307,
      "balance_loss_mlp": 1.02195668,
      "epoch": 0.6072265977274094,
      "flos": 24460517520000.0,
      "grad_norm": 2.069062293539178,
      "language_loss": 0.71653664,
      "learning_rate": 1.4113758622818522e-06,
      "loss": 0.73858142,
      "num_input_tokens_seen": 108845075,
      "step": 5050,
      "time_per_iteration": 2.5121712684631348
    },
    {
      "auxiliary_loss_clip": 0.01174096,
      "auxiliary_loss_mlp": 0.01020946,
      "balance_loss_clip": 1.05129039,
      "balance_loss_mlp": 1.01379895,
      "epoch": 0.6073468406180484,
      "flos": 18149253413760.0,
      "grad_norm": 5.35720145753227,
      "language_loss": 0.83281302,
      "learning_rate": 1.410631436442751e-06,
      "loss": 0.85476345,
      "num_input_tokens_seen": 108863870,
      "step": 5051,
      "time_per_iteration": 3.5063607692718506
    },
    {
      "auxiliary_loss_clip": 0.01176454,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.05152822,
      "balance_loss_mlp": 1.0182538,
      "epoch": 0.6074670835086875,
      "flos": 20697617669760.0,
      "grad_norm": 2.177599321286439,
      "language_loss": 0.86662215,
      "learning_rate": 1.4098871000103936e-06,
      "loss": 0.88864803,
      "num_input_tokens_seen": 108882470,
      "step": 5052,
      "time_per_iteration": 2.563506841659546
    },
    {
      "auxiliary_loss_clip": 0.0117378,
      "auxiliary_loss_mlp": 0.01022445,
      "balance_loss_clip": 1.0518657,
      "balance_loss_mlp": 1.01500058,
      "epoch": 0.6075873263993267,
      "flos": 23769955572480.0,
      "grad_norm": 1.9434473364914675,
      "language_loss": 0.82990605,
      "learning_rate": 1.409142853097693e-06,
      "loss": 0.85186833,
      "num_input_tokens_seen": 108902710,
      "step": 5053,
      "time_per_iteration": 2.6110587120056152
    },
    {
      "auxiliary_loss_clip": 0.01177206,
      "auxiliary_loss_mlp": 0.01026106,
      "balance_loss_clip": 1.05171669,
      "balance_loss_mlp": 1.01826191,
      "epoch": 0.6077075692899657,
      "flos": 24454484035200.0,
      "grad_norm": 2.3854993012407784,
      "language_loss": 0.7939778,
      "learning_rate": 1.408398695817553e-06,
      "loss": 0.81601095,
      "num_input_tokens_seen": 108919935,
      "step": 5054,
      "time_per_iteration": 3.526195526123047
    },
    {
      "auxiliary_loss_clip": 0.01179959,
      "auxiliary_loss_mlp": 0.01028943,
      "balance_loss_clip": 1.05311084,
      "balance_loss_mlp": 1.02006829,
      "epoch": 0.6078278121806048,
      "flos": 27382102041600.0,
      "grad_norm": 2.2845393115016197,
      "language_loss": 0.69882083,
      "learning_rate": 1.4076546282828593e-06,
      "loss": 0.72090983,
      "num_input_tokens_seen": 108942790,
      "step": 5055,
      "time_per_iteration": 2.6602423191070557
    },
    {
      "auxiliary_loss_clip": 0.01175765,
      "auxiliary_loss_mlp": 0.01028033,
      "balance_loss_clip": 1.0500865,
      "balance_loss_mlp": 1.01994514,
      "epoch": 0.6079480550712439,
      "flos": 38436447306240.0,
      "grad_norm": 4.393429973402587,
      "language_loss": 0.66578865,
      "learning_rate": 1.4069106506064874e-06,
      "loss": 0.68782663,
      "num_input_tokens_seen": 108964215,
      "step": 5056,
      "time_per_iteration": 2.725360155105591
    },
    {
      "auxiliary_loss_clip": 0.01173289,
      "auxiliary_loss_mlp": 0.01025395,
      "balance_loss_clip": 1.05155039,
      "balance_loss_mlp": 1.01808429,
      "epoch": 0.608068297961883,
      "flos": 25336271013120.0,
      "grad_norm": 1.9360014804409729,
      "language_loss": 0.78379416,
      "learning_rate": 1.4061667629012989e-06,
      "loss": 0.80578101,
      "num_input_tokens_seen": 108984885,
      "step": 5057,
      "time_per_iteration": 3.520887613296509
    },
    {
      "auxiliary_loss_clip": 0.01174986,
      "auxiliary_loss_mlp": 0.01024315,
      "balance_loss_clip": 1.05256045,
      "balance_loss_mlp": 1.0166229,
      "epoch": 0.608188540852522,
      "flos": 24202463235840.0,
      "grad_norm": 1.7489898984989023,
      "language_loss": 0.83280885,
      "learning_rate": 1.40542296528014e-06,
      "loss": 0.85480177,
      "num_input_tokens_seen": 109004545,
      "step": 5058,
      "time_per_iteration": 3.4829986095428467
    },
    {
      "auxiliary_loss_clip": 0.01177758,
      "auxiliary_loss_mlp": 0.01030121,
      "balance_loss_clip": 1.05236173,
      "balance_loss_mlp": 1.02124894,
      "epoch": 0.6083087837431612,
      "flos": 21284146851840.0,
      "grad_norm": 2.004638958481911,
      "language_loss": 0.75760901,
      "learning_rate": 1.4046792578558452e-06,
      "loss": 0.77968776,
      "num_input_tokens_seen": 109022440,
      "step": 5059,
      "time_per_iteration": 2.633958578109741
    },
    {
      "auxiliary_loss_clip": 0.01174016,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 1.05093265,
      "balance_loss_mlp": 1.01878834,
      "epoch": 0.6084290266338003,
      "flos": 16471435178880.0,
      "grad_norm": 2.7481888115001207,
      "language_loss": 0.7575807,
      "learning_rate": 1.4039356407412325e-06,
      "loss": 0.77958405,
      "num_input_tokens_seen": 109035680,
      "step": 5060,
      "time_per_iteration": 2.590330123901367
    },
    {
      "auxiliary_loss_clip": 0.01074526,
      "auxiliary_loss_mlp": 0.01001428,
      "balance_loss_clip": 1.01486385,
      "balance_loss_mlp": 1.00031364,
      "epoch": 0.6085492695244393,
      "flos": 66443574931200.0,
      "grad_norm": 0.7935373245692774,
      "language_loss": 0.57070643,
      "learning_rate": 1.40319211404911e-06,
      "loss": 0.59146601,
      "num_input_tokens_seen": 109090680,
      "step": 5061,
      "time_per_iteration": 3.160839557647705
    },
    {
      "auxiliary_loss_clip": 0.01175934,
      "auxiliary_loss_mlp": 0.01026319,
      "balance_loss_clip": 1.05103087,
      "balance_loss_mlp": 1.0185287,
      "epoch": 0.6086695124150785,
      "flos": 23618986709760.0,
      "grad_norm": 2.523165638785731,
      "language_loss": 0.9027878,
      "learning_rate": 1.4024486778922691e-06,
      "loss": 0.92481035,
      "num_input_tokens_seen": 109108995,
      "step": 5062,
      "time_per_iteration": 2.6703996658325195
    },
    {
      "auxiliary_loss_clip": 0.01174076,
      "auxiliary_loss_mlp": 0.01027094,
      "balance_loss_clip": 1.05035949,
      "balance_loss_mlp": 1.01924407,
      "epoch": 0.6087897553057176,
      "flos": 20157054917760.0,
      "grad_norm": 3.200423706436397,
      "language_loss": 0.77860785,
      "learning_rate": 1.4017053323834884e-06,
      "loss": 0.80061948,
      "num_input_tokens_seen": 109128825,
      "step": 5063,
      "time_per_iteration": 2.57686448097229
    },
    {
      "auxiliary_loss_clip": 0.0117154,
      "auxiliary_loss_mlp": 0.01026596,
      "balance_loss_clip": 1.04816163,
      "balance_loss_mlp": 1.01891875,
      "epoch": 0.6089099981963566,
      "flos": 25482535194240.0,
      "grad_norm": 1.8228468949641408,
      "language_loss": 0.7598325,
      "learning_rate": 1.4009620776355333e-06,
      "loss": 0.78181386,
      "num_input_tokens_seen": 109150425,
      "step": 5064,
      "time_per_iteration": 2.647151470184326
    },
    {
      "auxiliary_loss_clip": 0.01177137,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.05181432,
      "balance_loss_mlp": 1.02095032,
      "epoch": 0.6090302410869958,
      "flos": 25332895134720.0,
      "grad_norm": 2.0814147801009644,
      "language_loss": 0.79250962,
      "learning_rate": 1.4002189137611553e-06,
      "loss": 0.81457245,
      "num_input_tokens_seen": 109169765,
      "step": 5065,
      "time_per_iteration": 2.669668197631836
    },
    {
      "auxiliary_loss_clip": 0.01174593,
      "auxiliary_loss_mlp": 0.01022469,
      "balance_loss_clip": 1.05237389,
      "balance_loss_mlp": 1.01492035,
      "epoch": 0.6091504839776348,
      "flos": 23987358639360.0,
      "grad_norm": 1.8367555435292395,
      "language_loss": 0.70152354,
      "learning_rate": 1.3994758408730901e-06,
      "loss": 0.72349417,
      "num_input_tokens_seen": 109188950,
      "step": 5066,
      "time_per_iteration": 2.6243326663970947
    },
    {
      "auxiliary_loss_clip": 0.01178726,
      "auxiliary_loss_mlp": 0.01030993,
      "balance_loss_clip": 1.05196667,
      "balance_loss_mlp": 1.02254701,
      "epoch": 0.6092707268682739,
      "flos": 29643037666560.0,
      "grad_norm": 2.4492605225639092,
      "language_loss": 0.76470435,
      "learning_rate": 1.3987328590840629e-06,
      "loss": 0.78680158,
      "num_input_tokens_seen": 109209895,
      "step": 5067,
      "time_per_iteration": 2.6747100353240967
    },
    {
      "auxiliary_loss_clip": 0.01176373,
      "auxiliary_loss_mlp": 0.01028129,
      "balance_loss_clip": 1.05084145,
      "balance_loss_mlp": 1.02028489,
      "epoch": 0.609390969758913,
      "flos": 24024957200640.0,
      "grad_norm": 2.3960453713332095,
      "language_loss": 0.86338961,
      "learning_rate": 1.397989968506783e-06,
      "loss": 0.88543469,
      "num_input_tokens_seen": 109228905,
      "step": 5068,
      "time_per_iteration": 2.666837453842163
    },
    {
      "auxiliary_loss_clip": 0.01180403,
      "auxiliary_loss_mlp": 0.0102208,
      "balance_loss_clip": 1.05410647,
      "balance_loss_mlp": 1.01402152,
      "epoch": 0.6095112126495521,
      "flos": 11102143288320.0,
      "grad_norm": 4.096706021275902,
      "language_loss": 0.72456408,
      "learning_rate": 1.3972471692539458e-06,
      "loss": 0.74658889,
      "num_input_tokens_seen": 109243620,
      "step": 5069,
      "time_per_iteration": 2.5400707721710205
    },
    {
      "auxiliary_loss_clip": 0.01176214,
      "auxiliary_loss_mlp": 0.010244,
      "balance_loss_clip": 1.05268753,
      "balance_loss_mlp": 1.01678848,
      "epoch": 0.6096314555401912,
      "flos": 17265491187840.0,
      "grad_norm": 2.256076200126238,
      "language_loss": 0.75501966,
      "learning_rate": 1.3965044614382348e-06,
      "loss": 0.77702576,
      "num_input_tokens_seen": 109259070,
      "step": 5070,
      "time_per_iteration": 2.6183063983917236
    },
    {
      "auxiliary_loss_clip": 0.01179256,
      "auxiliary_loss_mlp": 0.01023769,
      "balance_loss_clip": 1.05345631,
      "balance_loss_mlp": 1.01564455,
      "epoch": 0.6097516984308303,
      "flos": 21645910679040.0,
      "grad_norm": 2.702975434023285,
      "language_loss": 0.76003909,
      "learning_rate": 1.3957618451723162e-06,
      "loss": 0.78206933,
      "num_input_tokens_seen": 109275100,
      "step": 5071,
      "time_per_iteration": 2.6500167846679688
    },
    {
      "auxiliary_loss_clip": 0.01174836,
      "auxiliary_loss_mlp": 0.01028297,
      "balance_loss_clip": 1.0508095,
      "balance_loss_mlp": 1.02022672,
      "epoch": 0.6098719413214694,
      "flos": 27199208966400.0,
      "grad_norm": 2.6845839171456567,
      "language_loss": 0.71491575,
      "learning_rate": 1.3950193205688457e-06,
      "loss": 0.73694706,
      "num_input_tokens_seen": 109294825,
      "step": 5072,
      "time_per_iteration": 2.638249158859253
    },
    {
      "auxiliary_loss_clip": 0.01175849,
      "auxiliary_loss_mlp": 0.01022335,
      "balance_loss_clip": 1.05186844,
      "balance_loss_mlp": 1.0147711,
      "epoch": 0.6099921842121084,
      "flos": 20412954385920.0,
      "grad_norm": 2.311278883515691,
      "language_loss": 0.83944547,
      "learning_rate": 1.3942768877404627e-06,
      "loss": 0.86142731,
      "num_input_tokens_seen": 109313790,
      "step": 5073,
      "time_per_iteration": 2.581040620803833
    },
    {
      "auxiliary_loss_clip": 0.01174881,
      "auxiliary_loss_mlp": 0.01026219,
      "balance_loss_clip": 1.0503056,
      "balance_loss_mlp": 1.01802373,
      "epoch": 0.6101124271027476,
      "flos": 23366139897600.0,
      "grad_norm": 1.799618107606574,
      "language_loss": 0.73704892,
      "learning_rate": 1.393534546799795e-06,
      "loss": 0.75905991,
      "num_input_tokens_seen": 109333490,
      "step": 5074,
      "time_per_iteration": 2.6881461143493652
    },
    {
      "auxiliary_loss_clip": 0.01174706,
      "auxiliary_loss_mlp": 0.01026222,
      "balance_loss_clip": 1.05196226,
      "balance_loss_mlp": 1.01827669,
      "epoch": 0.6102326699933867,
      "flos": 26687840993280.0,
      "grad_norm": 1.806467333853068,
      "language_loss": 0.68092483,
      "learning_rate": 1.3927922978594536e-06,
      "loss": 0.70293415,
      "num_input_tokens_seen": 109354575,
      "step": 5075,
      "time_per_iteration": 2.662118434906006
    },
    {
      "auxiliary_loss_clip": 0.01074283,
      "auxiliary_loss_mlp": 0.01000971,
      "balance_loss_clip": 1.0146091,
      "balance_loss_mlp": 0.9997732,
      "epoch": 0.6103529128840257,
      "flos": 60644612551680.0,
      "grad_norm": 0.7934158468409346,
      "language_loss": 0.57390624,
      "learning_rate": 1.3920501410320387e-06,
      "loss": 0.59465879,
      "num_input_tokens_seen": 109410690,
      "step": 5076,
      "time_per_iteration": 3.1567399501800537
    },
    {
      "auxiliary_loss_clip": 0.01176662,
      "auxiliary_loss_mlp": 0.01031139,
      "balance_loss_clip": 1.05045724,
      "balance_loss_mlp": 1.0233072,
      "epoch": 0.6104731557746649,
      "flos": 19021307806080.0,
      "grad_norm": 7.966960311119609,
      "language_loss": 0.7639963,
      "learning_rate": 1.3913080764301333e-06,
      "loss": 0.78607434,
      "num_input_tokens_seen": 109427650,
      "step": 5077,
      "time_per_iteration": 3.6115097999572754
    },
    {
      "auxiliary_loss_clip": 0.01178129,
      "auxiliary_loss_mlp": 0.010251,
      "balance_loss_clip": 1.0542382,
      "balance_loss_mlp": 1.01679134,
      "epoch": 0.6105933986653039,
      "flos": 23366894083200.0,
      "grad_norm": 1.9876905183807227,
      "language_loss": 0.71191722,
      "learning_rate": 1.3905661041663085e-06,
      "loss": 0.73394948,
      "num_input_tokens_seen": 109448835,
      "step": 5078,
      "time_per_iteration": 2.6445322036743164
    },
    {
      "auxiliary_loss_clip": 0.0117848,
      "auxiliary_loss_mlp": 0.0102735,
      "balance_loss_clip": 1.05099416,
      "balance_loss_mlp": 1.01889229,
      "epoch": 0.610713641555943,
      "flos": 34637565006720.0,
      "grad_norm": 2.1021399666587715,
      "language_loss": 0.6514197,
      "learning_rate": 1.389824224353122e-06,
      "loss": 0.67347801,
      "num_input_tokens_seen": 109470425,
      "step": 5079,
      "time_per_iteration": 2.744659423828125
    },
    {
      "auxiliary_loss_clip": 0.01176391,
      "auxiliary_loss_mlp": 0.01020313,
      "balance_loss_clip": 1.05242682,
      "balance_loss_mlp": 1.01337457,
      "epoch": 0.610833884446582,
      "flos": 26646471504000.0,
      "grad_norm": 1.5861693539179236,
      "language_loss": 0.77035552,
      "learning_rate": 1.389082437103115e-06,
      "loss": 0.79232252,
      "num_input_tokens_seen": 109489695,
      "step": 5080,
      "time_per_iteration": 3.628350257873535
    },
    {
      "auxiliary_loss_clip": 0.01180142,
      "auxiliary_loss_mlp": 0.0102194,
      "balance_loss_clip": 1.05386996,
      "balance_loss_mlp": 1.01375079,
      "epoch": 0.6109541273372212,
      "flos": 21215126868480.0,
      "grad_norm": 2.755305202606298,
      "language_loss": 0.78337049,
      "learning_rate": 1.3883407425288172e-06,
      "loss": 0.80539131,
      "num_input_tokens_seen": 109510030,
      "step": 5081,
      "time_per_iteration": 2.689481019973755
    },
    {
      "auxiliary_loss_clip": 0.01174185,
      "auxiliary_loss_mlp": 0.01026362,
      "balance_loss_clip": 1.04998362,
      "balance_loss_mlp": 1.0176419,
      "epoch": 0.6110743702278603,
      "flos": 20084084438400.0,
      "grad_norm": 2.3533164061855363,
      "language_loss": 0.79706156,
      "learning_rate": 1.3875991407427417e-06,
      "loss": 0.81906706,
      "num_input_tokens_seen": 109528255,
      "step": 5082,
      "time_per_iteration": 2.7318406105041504
    },
    {
      "auxiliary_loss_clip": 0.01073642,
      "auxiliary_loss_mlp": 0.01001386,
      "balance_loss_clip": 1.0141784,
      "balance_loss_mlp": 1.00020015,
      "epoch": 0.6111946131184993,
      "flos": 68302957438080.0,
      "grad_norm": 0.7701076449923462,
      "language_loss": 0.58147287,
      "learning_rate": 1.38685763185739e-06,
      "loss": 0.60222316,
      "num_input_tokens_seen": 109581915,
      "step": 5083,
      "time_per_iteration": 4.00290060043335
    },
    {
      "auxiliary_loss_clip": 0.01174228,
      "auxiliary_loss_mlp": 0.01030559,
      "balance_loss_clip": 1.05098605,
      "balance_loss_mlp": 1.02289128,
      "epoch": 0.6113148560091385,
      "flos": 19937676602880.0,
      "grad_norm": 3.4340658544548877,
      "language_loss": 0.67877692,
      "learning_rate": 1.3861162159852476e-06,
      "loss": 0.7008248,
      "num_input_tokens_seen": 109600050,
      "step": 5084,
      "time_per_iteration": 3.507936477661133
    },
    {
      "auxiliary_loss_clip": 0.01179061,
      "auxiliary_loss_mlp": 0.01029664,
      "balance_loss_clip": 1.05168676,
      "balance_loss_mlp": 1.0219183,
      "epoch": 0.6114350988997775,
      "flos": 23731854220800.0,
      "grad_norm": 2.2656325484729156,
      "language_loss": 0.79841226,
      "learning_rate": 1.3853748932387875e-06,
      "loss": 0.82049954,
      "num_input_tokens_seen": 109620690,
      "step": 5085,
      "time_per_iteration": 2.7465078830718994
    },
    {
      "auxiliary_loss_clip": 0.0117569,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.04956245,
      "balance_loss_mlp": 1.02187014,
      "epoch": 0.6115553417904166,
      "flos": 24023700224640.0,
      "grad_norm": 2.2989734190913693,
      "language_loss": 0.74989271,
      "learning_rate": 1.3846336637304671e-06,
      "loss": 0.77195227,
      "num_input_tokens_seen": 109638960,
      "step": 5086,
      "time_per_iteration": 2.594801187515259
    },
    {
      "auxiliary_loss_clip": 0.01174509,
      "auxiliary_loss_mlp": 0.01023665,
      "balance_loss_clip": 1.04968429,
      "balance_loss_mlp": 1.01624978,
      "epoch": 0.6116755846810558,
      "flos": 23733542160000.0,
      "grad_norm": 2.2036472579114306,
      "language_loss": 0.83360285,
      "learning_rate": 1.3838925275727316e-06,
      "loss": 0.85558456,
      "num_input_tokens_seen": 109659700,
      "step": 5087,
      "time_per_iteration": 2.667451858520508
    },
    {
      "auxiliary_loss_clip": 0.01179857,
      "auxiliary_loss_mlp": 0.01029257,
      "balance_loss_clip": 1.05443668,
      "balance_loss_mlp": 1.02172303,
      "epoch": 0.6117958275716948,
      "flos": 18661626967680.0,
      "grad_norm": 1.9705859037760316,
      "language_loss": 0.79575723,
      "learning_rate": 1.3831514848780089e-06,
      "loss": 0.81784838,
      "num_input_tokens_seen": 109679275,
      "step": 5088,
      "time_per_iteration": 2.6041879653930664
    },
    {
      "auxiliary_loss_clip": 0.01174246,
      "auxiliary_loss_mlp": 0.01023442,
      "balance_loss_clip": 1.0522871,
      "balance_loss_mlp": 1.01640606,
      "epoch": 0.6119160704623339,
      "flos": 16471183783680.0,
      "grad_norm": 2.486225073864357,
      "language_loss": 0.92037272,
      "learning_rate": 1.3824105357587152e-06,
      "loss": 0.94234961,
      "num_input_tokens_seen": 109696380,
      "step": 5089,
      "time_per_iteration": 2.63191556930542
    },
    {
      "auxiliary_loss_clip": 0.01174796,
      "auxiliary_loss_mlp": 0.01028218,
      "balance_loss_clip": 1.05115795,
      "balance_loss_mlp": 1.02060688,
      "epoch": 0.612036313352973,
      "flos": 23915465568000.0,
      "grad_norm": 1.5735223528698048,
      "language_loss": 0.82747513,
      "learning_rate": 1.381669680327253e-06,
      "loss": 0.84950525,
      "num_input_tokens_seen": 109718060,
      "step": 5090,
      "time_per_iteration": 2.632192611694336
    },
    {
      "auxiliary_loss_clip": 0.0117828,
      "auxiliary_loss_mlp": 0.01030231,
      "balance_loss_clip": 1.05433536,
      "balance_loss_mlp": 1.02244043,
      "epoch": 0.6121565562436121,
      "flos": 26974766833920.0,
      "grad_norm": 3.5503925484419923,
      "language_loss": 0.71198338,
      "learning_rate": 1.380928918696008e-06,
      "loss": 0.73406851,
      "num_input_tokens_seen": 109736830,
      "step": 5091,
      "time_per_iteration": 2.705899238586426
    },
    {
      "auxiliary_loss_clip": 0.01173564,
      "auxiliary_loss_mlp": 0.01025094,
      "balance_loss_clip": 1.04984999,
      "balance_loss_mlp": 1.01753867,
      "epoch": 0.6122767991342511,
      "flos": 15668867646720.0,
      "grad_norm": 2.733940806884582,
      "language_loss": 0.7174598,
      "learning_rate": 1.3801882509773548e-06,
      "loss": 0.7394464,
      "num_input_tokens_seen": 109754690,
      "step": 5092,
      "time_per_iteration": 2.6250057220458984
    },
    {
      "auxiliary_loss_clip": 0.01175479,
      "auxiliary_loss_mlp": 0.01025505,
      "balance_loss_clip": 1.05053115,
      "balance_loss_mlp": 1.01779819,
      "epoch": 0.6123970420248903,
      "flos": 27964321591680.0,
      "grad_norm": 2.0831880542796712,
      "language_loss": 0.81658238,
      "learning_rate": 1.3794476772836503e-06,
      "loss": 0.83859217,
      "num_input_tokens_seen": 109775790,
      "step": 5093,
      "time_per_iteration": 2.7234935760498047
    },
    {
      "auxiliary_loss_clip": 0.01176354,
      "auxiliary_loss_mlp": 0.0102711,
      "balance_loss_clip": 1.05054796,
      "balance_loss_mlp": 1.01872349,
      "epoch": 0.6125172849155294,
      "flos": 21468727866240.0,
      "grad_norm": 13.439617436511151,
      "language_loss": 0.848737,
      "learning_rate": 1.3787071977272402e-06,
      "loss": 0.87077159,
      "num_input_tokens_seen": 109795050,
      "step": 5094,
      "time_per_iteration": 2.616183280944824
    },
    {
      "auxiliary_loss_clip": 0.01178722,
      "auxiliary_loss_mlp": 0.01025288,
      "balance_loss_clip": 1.05181527,
      "balance_loss_mlp": 1.01786685,
      "epoch": 0.6126375278061684,
      "flos": 16248321849600.0,
      "grad_norm": 2.8793595702975643,
      "language_loss": 0.72121108,
      "learning_rate": 1.3779668124204535e-06,
      "loss": 0.7432512,
      "num_input_tokens_seen": 109811465,
      "step": 5095,
      "time_per_iteration": 2.623410224914551
    },
    {
      "auxiliary_loss_clip": 0.01173765,
      "auxiliary_loss_mlp": 0.01023872,
      "balance_loss_clip": 1.05213189,
      "balance_loss_mlp": 1.01631689,
      "epoch": 0.6127577706968076,
      "flos": 20448865008000.0,
      "grad_norm": 1.7557840091535541,
      "language_loss": 0.80774784,
      "learning_rate": 1.3772265214756074e-06,
      "loss": 0.82972419,
      "num_input_tokens_seen": 109831225,
      "step": 5096,
      "time_per_iteration": 2.7026443481445312
    },
    {
      "auxiliary_loss_clip": 0.01178775,
      "auxiliary_loss_mlp": 0.01027675,
      "balance_loss_clip": 1.05274987,
      "balance_loss_mlp": 1.0195868,
      "epoch": 0.6128780135874466,
      "flos": 18260397072000.0,
      "grad_norm": 2.270409973471735,
      "language_loss": 0.75674272,
      "learning_rate": 1.3764863250050025e-06,
      "loss": 0.77880716,
      "num_input_tokens_seen": 109849465,
      "step": 5097,
      "time_per_iteration": 2.633640766143799
    },
    {
      "auxiliary_loss_clip": 0.0117555,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.05242848,
      "balance_loss_mlp": 1.01909673,
      "epoch": 0.6129982564780857,
      "flos": 24937088192640.0,
      "grad_norm": 1.8465123141662785,
      "language_loss": 0.80331767,
      "learning_rate": 1.3757462231209272e-06,
      "loss": 0.82534134,
      "num_input_tokens_seen": 109869770,
      "step": 5098,
      "time_per_iteration": 2.6963014602661133
    },
    {
      "auxiliary_loss_clip": 0.01175721,
      "auxiliary_loss_mlp": 0.01025561,
      "balance_loss_clip": 1.05111766,
      "balance_loss_mlp": 1.0173651,
      "epoch": 0.6131184993687249,
      "flos": 22492038430080.0,
      "grad_norm": 1.9702760178730385,
      "language_loss": 0.88768101,
      "learning_rate": 1.3750062159356525e-06,
      "loss": 0.90969384,
      "num_input_tokens_seen": 109889120,
      "step": 5099,
      "time_per_iteration": 2.643908739089966
    },
    {
      "auxiliary_loss_clip": 0.01174827,
      "auxiliary_loss_mlp": 0.01023582,
      "balance_loss_clip": 1.05170894,
      "balance_loss_mlp": 1.01632774,
      "epoch": 0.6132387422593639,
      "flos": 15885839750400.0,
      "grad_norm": 1.9509934058747913,
      "language_loss": 0.83121771,
      "learning_rate": 1.3742663035614382e-06,
      "loss": 0.85320175,
      "num_input_tokens_seen": 109906490,
      "step": 5100,
      "time_per_iteration": 2.6064088344573975
    },
    {
      "auxiliary_loss_clip": 0.01178524,
      "auxiliary_loss_mlp": 0.01025376,
      "balance_loss_clip": 1.05282462,
      "balance_loss_mlp": 1.01753175,
      "epoch": 0.613358985150003,
      "flos": 25411539962880.0,
      "grad_norm": 4.552741087314072,
      "language_loss": 0.80023241,
      "learning_rate": 1.3735264861105283e-06,
      "loss": 0.82227147,
      "num_input_tokens_seen": 109927130,
      "step": 5101,
      "time_per_iteration": 2.640545606613159
    },
    {
      "auxiliary_loss_clip": 0.01172749,
      "auxiliary_loss_mlp": 0.01027721,
      "balance_loss_clip": 1.05022049,
      "balance_loss_mlp": 1.02017474,
      "epoch": 0.6134792280406421,
      "flos": 21361283308800.0,
      "grad_norm": 2.1888664546202756,
      "language_loss": 0.78161693,
      "learning_rate": 1.372786763695152e-06,
      "loss": 0.80362165,
      "num_input_tokens_seen": 109945890,
      "step": 5102,
      "time_per_iteration": 2.6140151023864746
    },
    {
      "auxiliary_loss_clip": 0.01178408,
      "auxiliary_loss_mlp": 0.01027825,
      "balance_loss_clip": 1.05317366,
      "balance_loss_mlp": 1.01950407,
      "epoch": 0.6135994709312812,
      "flos": 21211248199680.0,
      "grad_norm": 2.683829362816,
      "language_loss": 0.77535367,
      "learning_rate": 1.3720471364275257e-06,
      "loss": 0.79741597,
      "num_input_tokens_seen": 109965535,
      "step": 5103,
      "time_per_iteration": 3.520585775375366
    },
    {
      "auxiliary_loss_clip": 0.01177264,
      "auxiliary_loss_mlp": 0.01029788,
      "balance_loss_clip": 1.05244446,
      "balance_loss_mlp": 1.02197957,
      "epoch": 0.6137197138219203,
      "flos": 14794047907200.0,
      "grad_norm": 1.9961075187615835,
      "language_loss": 0.78236932,
      "learning_rate": 1.3713076044198486e-06,
      "loss": 0.80443978,
      "num_input_tokens_seen": 109982345,
      "step": 5104,
      "time_per_iteration": 2.595848798751831
    },
    {
      "auxiliary_loss_clip": 0.01175184,
      "auxiliary_loss_mlp": 0.01026328,
      "balance_loss_clip": 1.05102229,
      "balance_loss_mlp": 1.01865435,
      "epoch": 0.6138399567125594,
      "flos": 20084515401600.0,
      "grad_norm": 2.2508627964728154,
      "language_loss": 0.81300437,
      "learning_rate": 1.3705681677843086e-06,
      "loss": 0.83501947,
      "num_input_tokens_seen": 110000940,
      "step": 5105,
      "time_per_iteration": 2.7940633296966553
    },
    {
      "auxiliary_loss_clip": 0.01074174,
      "auxiliary_loss_mlp": 0.01002433,
      "balance_loss_clip": 1.01442266,
      "balance_loss_mlp": 1.00132394,
      "epoch": 0.6139601996031985,
      "flos": 60123838193280.0,
      "grad_norm": 0.7729562778671449,
      "language_loss": 0.60603535,
      "learning_rate": 1.3698288266330768e-06,
      "loss": 0.62680137,
      "num_input_tokens_seen": 110061565,
      "step": 5106,
      "time_per_iteration": 3.30122971534729
    },
    {
      "auxiliary_loss_clip": 0.01176011,
      "auxiliary_loss_mlp": 0.0102983,
      "balance_loss_clip": 1.05212808,
      "balance_loss_mlp": 1.02177191,
      "epoch": 0.6140804424938375,
      "flos": 23586703361280.0,
      "grad_norm": 2.3429903304119635,
      "language_loss": 0.72546071,
      "learning_rate": 1.3690895810783113e-06,
      "loss": 0.74751914,
      "num_input_tokens_seen": 110080360,
      "step": 5107,
      "time_per_iteration": 3.574796676635742
    },
    {
      "auxiliary_loss_clip": 0.01175024,
      "auxiliary_loss_mlp": 0.01031895,
      "balance_loss_clip": 1.04926157,
      "balance_loss_mlp": 1.0238601,
      "epoch": 0.6142006853844767,
      "flos": 21398199511680.0,
      "grad_norm": 2.0268676952825815,
      "language_loss": 0.70660961,
      "learning_rate": 1.3683504312321543e-06,
      "loss": 0.72867882,
      "num_input_tokens_seen": 110100695,
      "step": 5108,
      "time_per_iteration": 2.60750412940979
    },
    {
      "auxiliary_loss_clip": 0.01175115,
      "auxiliary_loss_mlp": 0.01023492,
      "balance_loss_clip": 1.04988623,
      "balance_loss_mlp": 1.01543319,
      "epoch": 0.6143209282751158,
      "flos": 12057367622400.0,
      "grad_norm": 1.8877894599156348,
      "language_loss": 0.80513686,
      "learning_rate": 1.3676113772067355e-06,
      "loss": 0.82712293,
      "num_input_tokens_seen": 110117750,
      "step": 5109,
      "time_per_iteration": 3.539552688598633
    },
    {
      "auxiliary_loss_clip": 0.01178906,
      "auxiliary_loss_mlp": 0.01027378,
      "balance_loss_clip": 1.05243349,
      "balance_loss_mlp": 1.01879549,
      "epoch": 0.6144411711657548,
      "flos": 25082274965760.0,
      "grad_norm": 1.8553003257375391,
      "language_loss": 0.72681439,
      "learning_rate": 1.3668724191141671e-06,
      "loss": 0.74887729,
      "num_input_tokens_seen": 110137020,
      "step": 5110,
      "time_per_iteration": 3.5980794429779053
    },
    {
      "auxiliary_loss_clip": 0.01175174,
      "auxiliary_loss_mlp": 0.01027117,
      "balance_loss_clip": 1.0521841,
      "balance_loss_mlp": 1.01928449,
      "epoch": 0.6145614140563939,
      "flos": 20114069316480.0,
      "grad_norm": 3.8614138560651954,
      "language_loss": 0.66631466,
      "learning_rate": 1.3661335570665493e-06,
      "loss": 0.68833762,
      "num_input_tokens_seen": 110154930,
      "step": 5111,
      "time_per_iteration": 2.570260524749756
    },
    {
      "auxiliary_loss_clip": 0.01180379,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.05541706,
      "balance_loss_mlp": 1.01907301,
      "epoch": 0.614681656947033,
      "flos": 16800376953600.0,
      "grad_norm": 4.021768397746524,
      "language_loss": 0.69802392,
      "learning_rate": 1.3653947911759676e-06,
      "loss": 0.72009891,
      "num_input_tokens_seen": 110172480,
      "step": 5112,
      "time_per_iteration": 2.631760835647583
    },
    {
      "auxiliary_loss_clip": 0.01180439,
      "auxiliary_loss_mlp": 0.01030529,
      "balance_loss_clip": 1.05456746,
      "balance_loss_mlp": 1.02217841,
      "epoch": 0.6148018998376721,
      "flos": 38801587011840.0,
      "grad_norm": 3.9453416034741347,
      "language_loss": 0.74584061,
      "learning_rate": 1.3646561215544904e-06,
      "loss": 0.7679503,
      "num_input_tokens_seen": 110197120,
      "step": 5113,
      "time_per_iteration": 2.7534472942352295
    },
    {
      "auxiliary_loss_clip": 0.01175501,
      "auxiliary_loss_mlp": 0.01024935,
      "balance_loss_clip": 1.05086517,
      "balance_loss_mlp": 1.01684928,
      "epoch": 0.6149221427283111,
      "flos": 23327032965120.0,
      "grad_norm": 2.0944678391062492,
      "language_loss": 0.79625821,
      "learning_rate": 1.363917548314176e-06,
      "loss": 0.81826258,
      "num_input_tokens_seen": 110216385,
      "step": 5114,
      "time_per_iteration": 2.6604340076446533
    },
    {
      "auxiliary_loss_clip": 0.01178745,
      "auxiliary_loss_mlp": 0.01024103,
      "balance_loss_clip": 1.05241406,
      "balance_loss_mlp": 1.01584518,
      "epoch": 0.6150423856189503,
      "flos": 22379494141440.0,
      "grad_norm": 3.136273803900325,
      "language_loss": 0.73386431,
      "learning_rate": 1.3631790715670626e-06,
      "loss": 0.75589281,
      "num_input_tokens_seen": 110234790,
      "step": 5115,
      "time_per_iteration": 2.619568109512329
    },
    {
      "auxiliary_loss_clip": 0.01173244,
      "auxiliary_loss_mlp": 0.01025191,
      "balance_loss_clip": 1.05216467,
      "balance_loss_mlp": 1.01779962,
      "epoch": 0.6151626285095894,
      "flos": 18692078722560.0,
      "grad_norm": 2.020135619432927,
      "language_loss": 0.85782325,
      "learning_rate": 1.3624406914251783e-06,
      "loss": 0.87980759,
      "num_input_tokens_seen": 110251910,
      "step": 5116,
      "time_per_iteration": 2.5880863666534424
    },
    {
      "auxiliary_loss_clip": 0.01174843,
      "auxiliary_loss_mlp": 0.01031246,
      "balance_loss_clip": 1.0506804,
      "balance_loss_mlp": 1.02381921,
      "epoch": 0.6152828714002284,
      "flos": 15851688894720.0,
      "grad_norm": 2.2010091654727537,
      "language_loss": 0.88345498,
      "learning_rate": 1.3617024080005335e-06,
      "loss": 0.90551591,
      "num_input_tokens_seen": 110268810,
      "step": 5117,
      "time_per_iteration": 2.836256265640259
    },
    {
      "auxiliary_loss_clip": 0.011769,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.05153191,
      "balance_loss_mlp": 1.01837516,
      "epoch": 0.6154031142908676,
      "flos": 24869792062080.0,
      "grad_norm": 1.8362434068882298,
      "language_loss": 0.74503279,
      "learning_rate": 1.3609642214051266e-06,
      "loss": 0.7670635,
      "num_input_tokens_seen": 110293035,
      "step": 5118,
      "time_per_iteration": 2.8095755577087402
    },
    {
      "auxiliary_loss_clip": 0.0117654,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.05282402,
      "balance_loss_mlp": 1.01598573,
      "epoch": 0.6155233571815066,
      "flos": 19244744357760.0,
      "grad_norm": 2.6219427274300684,
      "language_loss": 0.66081774,
      "learning_rate": 1.3602261317509385e-06,
      "loss": 0.68282717,
      "num_input_tokens_seen": 110309695,
      "step": 5119,
      "time_per_iteration": 2.6062188148498535
    },
    {
      "auxiliary_loss_clip": 0.01178912,
      "auxiliary_loss_mlp": 0.01032804,
      "balance_loss_clip": 1.05203104,
      "balance_loss_mlp": 1.02395868,
      "epoch": 0.6156436000721457,
      "flos": 18770077105920.0,
      "grad_norm": 2.7673684987148737,
      "language_loss": 0.83069855,
      "learning_rate": 1.3594881391499387e-06,
      "loss": 0.85281575,
      "num_input_tokens_seen": 110328610,
      "step": 5120,
      "time_per_iteration": 2.572354316711426
    },
    {
      "auxiliary_loss_clip": 0.01176164,
      "auxiliary_loss_mlp": 0.01024413,
      "balance_loss_clip": 1.05281091,
      "balance_loss_mlp": 1.01682854,
      "epoch": 0.6157638429627849,
      "flos": 18041198325120.0,
      "grad_norm": 4.046541288283113,
      "language_loss": 0.79151195,
      "learning_rate": 1.3587502437140778e-06,
      "loss": 0.81351769,
      "num_input_tokens_seen": 110346775,
      "step": 5121,
      "time_per_iteration": 2.620945453643799
    },
    {
      "auxiliary_loss_clip": 0.01176456,
      "auxiliary_loss_mlp": 0.01027941,
      "balance_loss_clip": 1.0502485,
      "balance_loss_mlp": 1.02029943,
      "epoch": 0.6158840858534239,
      "flos": 25556726736000.0,
      "grad_norm": 2.1012305354000764,
      "language_loss": 0.85661948,
      "learning_rate": 1.3580124455552952e-06,
      "loss": 0.87866348,
      "num_input_tokens_seen": 110366140,
      "step": 5122,
      "time_per_iteration": 2.6071510314941406
    },
    {
      "auxiliary_loss_clip": 0.01175188,
      "auxiliary_loss_mlp": 0.01033481,
      "balance_loss_clip": 1.05068731,
      "balance_loss_mlp": 1.02604532,
      "epoch": 0.616004328744063,
      "flos": 24640788902400.0,
      "grad_norm": 1.6898498131877286,
      "language_loss": 0.87324619,
      "learning_rate": 1.3572747447855148e-06,
      "loss": 0.89533287,
      "num_input_tokens_seen": 110386550,
      "step": 5123,
      "time_per_iteration": 2.6634950637817383
    },
    {
      "auxiliary_loss_clip": 0.01177291,
      "auxiliary_loss_mlp": 0.01030276,
      "balance_loss_clip": 1.05199766,
      "balance_loss_mlp": 1.02187204,
      "epoch": 0.6161245716347021,
      "flos": 21689686379520.0,
      "grad_norm": 1.782925297509805,
      "language_loss": 0.69295543,
      "learning_rate": 1.356537141516644e-06,
      "loss": 0.71503115,
      "num_input_tokens_seen": 110403970,
      "step": 5124,
      "time_per_iteration": 2.5990822315216064
    },
    {
      "auxiliary_loss_clip": 0.01177312,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.05369377,
      "balance_loss_mlp": 1.02338481,
      "epoch": 0.6162448145253412,
      "flos": 35189225061120.0,
      "grad_norm": 2.7558720310569367,
      "language_loss": 0.62197077,
      "learning_rate": 1.3557996358605775e-06,
      "loss": 0.64404947,
      "num_input_tokens_seen": 110423890,
      "step": 5125,
      "time_per_iteration": 2.7460200786590576
    },
    {
      "auxiliary_loss_clip": 0.01174016,
      "auxiliary_loss_mlp": 0.0102304,
      "balance_loss_clip": 1.05062926,
      "balance_loss_mlp": 1.01593208,
      "epoch": 0.6163650574159802,
      "flos": 21615279356160.0,
      "grad_norm": 7.828170644420449,
      "language_loss": 0.70390141,
      "learning_rate": 1.3550622279291941e-06,
      "loss": 0.72587204,
      "num_input_tokens_seen": 110442035,
      "step": 5126,
      "time_per_iteration": 2.594609260559082
    },
    {
      "auxiliary_loss_clip": 0.01173818,
      "auxiliary_loss_mlp": 0.01030068,
      "balance_loss_clip": 1.05042624,
      "balance_loss_mlp": 1.02244425,
      "epoch": 0.6164853003066194,
      "flos": 24572163968640.0,
      "grad_norm": 1.6123782023856918,
      "language_loss": 0.83196288,
      "learning_rate": 1.354324917834358e-06,
      "loss": 0.85400176,
      "num_input_tokens_seen": 110463280,
      "step": 5127,
      "time_per_iteration": 2.635284423828125
    },
    {
      "auxiliary_loss_clip": 0.01171238,
      "auxiliary_loss_mlp": 0.01025472,
      "balance_loss_clip": 1.04804802,
      "balance_loss_mlp": 1.01810765,
      "epoch": 0.6166055431972585,
      "flos": 21835986474240.0,
      "grad_norm": 2.668612320109909,
      "language_loss": 0.76499426,
      "learning_rate": 1.353587705687918e-06,
      "loss": 0.78696144,
      "num_input_tokens_seen": 110481455,
      "step": 5128,
      "time_per_iteration": 2.5756959915161133
    },
    {
      "auxiliary_loss_clip": 0.01181362,
      "auxiliary_loss_mlp": 0.01025664,
      "balance_loss_clip": 1.05453444,
      "balance_loss_mlp": 1.01688397,
      "epoch": 0.6167257860878975,
      "flos": 17785262943360.0,
      "grad_norm": 5.765180083782208,
      "language_loss": 0.71900809,
      "learning_rate": 1.3528505916017096e-06,
      "loss": 0.74107838,
      "num_input_tokens_seen": 110499155,
      "step": 5129,
      "time_per_iteration": 2.6776342391967773
    },
    {
      "auxiliary_loss_clip": 0.01179703,
      "auxiliary_loss_mlp": 0.01026432,
      "balance_loss_clip": 1.05308545,
      "balance_loss_mlp": 1.01794422,
      "epoch": 0.6168460289785367,
      "flos": 23214811898880.0,
      "grad_norm": 2.116457244650885,
      "language_loss": 0.88425505,
      "learning_rate": 1.3521135756875514e-06,
      "loss": 0.9063164,
      "num_input_tokens_seen": 110515470,
      "step": 5130,
      "time_per_iteration": 3.5586891174316406
    },
    {
      "auxiliary_loss_clip": 0.01172882,
      "auxiliary_loss_mlp": 0.01026483,
      "balance_loss_clip": 1.04958701,
      "balance_loss_mlp": 1.0192647,
      "epoch": 0.6169662718691757,
      "flos": 26213281482240.0,
      "grad_norm": 1.9630982812137097,
      "language_loss": 0.86336637,
      "learning_rate": 1.3513766580572496e-06,
      "loss": 0.88536006,
      "num_input_tokens_seen": 110538290,
      "step": 5131,
      "time_per_iteration": 2.651895046234131
    },
    {
      "auxiliary_loss_clip": 0.01171469,
      "auxiliary_loss_mlp": 0.01023033,
      "balance_loss_clip": 1.05011702,
      "balance_loss_mlp": 1.01540959,
      "epoch": 0.6170865147598148,
      "flos": 19026120228480.0,
      "grad_norm": 3.5905507027602894,
      "language_loss": 0.7730363,
      "learning_rate": 1.3506398388225924e-06,
      "loss": 0.7949813,
      "num_input_tokens_seen": 110555610,
      "step": 5132,
      "time_per_iteration": 2.557616949081421
    },
    {
      "auxiliary_loss_clip": 0.01173236,
      "auxiliary_loss_mlp": 0.01025071,
      "balance_loss_clip": 1.05040097,
      "balance_loss_mlp": 1.01806211,
      "epoch": 0.617206757650454,
      "flos": 18260361158400.0,
      "grad_norm": 3.486371346011426,
      "language_loss": 0.71618032,
      "learning_rate": 1.349903118095355e-06,
      "loss": 0.73816341,
      "num_input_tokens_seen": 110574745,
      "step": 5133,
      "time_per_iteration": 3.534257173538208
    },
    {
      "auxiliary_loss_clip": 0.01177457,
      "auxiliary_loss_mlp": 0.01027903,
      "balance_loss_clip": 1.05191767,
      "balance_loss_mlp": 1.02039015,
      "epoch": 0.617327000541093,
      "flos": 18186959715840.0,
      "grad_norm": 2.4246614899929857,
      "language_loss": 0.73231804,
      "learning_rate": 1.349166495987298e-06,
      "loss": 0.75437164,
      "num_input_tokens_seen": 110593310,
      "step": 5134,
      "time_per_iteration": 2.62148118019104
    },
    {
      "auxiliary_loss_clip": 0.01075559,
      "auxiliary_loss_mlp": 0.01005041,
      "balance_loss_clip": 1.01500213,
      "balance_loss_mlp": 1.00376594,
      "epoch": 0.6174472434317321,
      "flos": 61833796122240.0,
      "grad_norm": 0.821036341287934,
      "language_loss": 0.60853249,
      "learning_rate": 1.348429972610166e-06,
      "loss": 0.6293385,
      "num_input_tokens_seen": 110657615,
      "step": 5135,
      "time_per_iteration": 4.130438566207886
    },
    {
      "auxiliary_loss_clip": 0.01075998,
      "auxiliary_loss_mlp": 0.01004351,
      "balance_loss_clip": 1.01529431,
      "balance_loss_mlp": 1.00313473,
      "epoch": 0.6175674863223712,
      "flos": 71230970494080.0,
      "grad_norm": 0.8496655936181637,
      "language_loss": 0.57756013,
      "learning_rate": 1.3476935480756897e-06,
      "loss": 0.59836358,
      "num_input_tokens_seen": 110714365,
      "step": 5136,
      "time_per_iteration": 3.979276418685913
    },
    {
      "auxiliary_loss_clip": 0.01176315,
      "auxiliary_loss_mlp": 0.01023449,
      "balance_loss_clip": 1.05234313,
      "balance_loss_mlp": 1.01493168,
      "epoch": 0.6176877292130103,
      "flos": 21835447770240.0,
      "grad_norm": 2.2696251302675425,
      "language_loss": 0.75412965,
      "learning_rate": 1.346957222495583e-06,
      "loss": 0.77612734,
      "num_input_tokens_seen": 110732160,
      "step": 5137,
      "time_per_iteration": 2.5648934841156006
    },
    {
      "auxiliary_loss_clip": 0.01175751,
      "auxiliary_loss_mlp": 0.0102829,
      "balance_loss_clip": 1.05084944,
      "balance_loss_mlp": 1.02014828,
      "epoch": 0.6178079721036493,
      "flos": 17741738638080.0,
      "grad_norm": 3.320213792314461,
      "language_loss": 0.71279025,
      "learning_rate": 1.3462209959815466e-06,
      "loss": 0.73483068,
      "num_input_tokens_seen": 110746900,
      "step": 5138,
      "time_per_iteration": 2.6284961700439453
    },
    {
      "auxiliary_loss_clip": 0.01175745,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.05343533,
      "balance_loss_mlp": 1.02011323,
      "epoch": 0.6179282149942885,
      "flos": 22633131052800.0,
      "grad_norm": 1.813897007548846,
      "language_loss": 0.74448419,
      "learning_rate": 1.345484868645265e-06,
      "loss": 0.76651889,
      "num_input_tokens_seen": 110765710,
      "step": 5139,
      "time_per_iteration": 2.6390576362609863
    },
    {
      "auxiliary_loss_clip": 0.01173972,
      "auxiliary_loss_mlp": 0.01029148,
      "balance_loss_clip": 1.05034637,
      "balance_loss_mlp": 1.02118516,
      "epoch": 0.6180484578849276,
      "flos": 22310330503680.0,
      "grad_norm": 1.8856991804828598,
      "language_loss": 0.78610885,
      "learning_rate": 1.3447488405984088e-06,
      "loss": 0.80814016,
      "num_input_tokens_seen": 110783970,
      "step": 5140,
      "time_per_iteration": 2.7183046340942383
    },
    {
      "auxiliary_loss_clip": 0.01174305,
      "auxiliary_loss_mlp": 0.0102809,
      "balance_loss_clip": 1.04997683,
      "balance_loss_mlp": 1.02009666,
      "epoch": 0.6181687007755666,
      "flos": 35225458905600.0,
      "grad_norm": 2.7483584835345223,
      "language_loss": 0.7018829,
      "learning_rate": 1.3440129119526322e-06,
      "loss": 0.72390687,
      "num_input_tokens_seen": 110806395,
      "step": 5141,
      "time_per_iteration": 2.73899245262146
    },
    {
      "auxiliary_loss_clip": 0.01077603,
      "auxiliary_loss_mlp": 0.01001971,
      "balance_loss_clip": 1.01601171,
      "balance_loss_mlp": 1.00063002,
      "epoch": 0.6182889436662057,
      "flos": 61547370094080.0,
      "grad_norm": 0.8046020613250651,
      "language_loss": 0.51164633,
      "learning_rate": 1.3432770828195762e-06,
      "loss": 0.53244209,
      "num_input_tokens_seen": 110867380,
      "step": 5142,
      "time_per_iteration": 3.3154170513153076
    },
    {
      "auxiliary_loss_clip": 0.01174158,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.04972541,
      "balance_loss_mlp": 1.01917219,
      "epoch": 0.6184091865568448,
      "flos": 19609991804160.0,
      "grad_norm": 2.4959800803081187,
      "language_loss": 0.70473188,
      "learning_rate": 1.3425413533108635e-06,
      "loss": 0.72674257,
      "num_input_tokens_seen": 110885980,
      "step": 5143,
      "time_per_iteration": 2.5974888801574707
    },
    {
      "auxiliary_loss_clip": 0.01175118,
      "auxiliary_loss_mlp": 0.01024208,
      "balance_loss_clip": 1.05251908,
      "balance_loss_mlp": 1.01606941,
      "epoch": 0.6185294294474839,
      "flos": 23586882929280.0,
      "grad_norm": 2.398286115139533,
      "language_loss": 0.70511812,
      "learning_rate": 1.341805723538105e-06,
      "loss": 0.72711146,
      "num_input_tokens_seen": 110906085,
      "step": 5144,
      "time_per_iteration": 2.646898031234741
    },
    {
      "auxiliary_loss_clip": 0.0117534,
      "auxiliary_loss_mlp": 0.01029614,
      "balance_loss_clip": 1.05161214,
      "balance_loss_mlp": 1.02166915,
      "epoch": 0.618649672338123,
      "flos": 26762032535040.0,
      "grad_norm": 2.560708788461925,
      "language_loss": 0.77870703,
      "learning_rate": 1.3410701936128948e-06,
      "loss": 0.80075657,
      "num_input_tokens_seen": 110928865,
      "step": 5145,
      "time_per_iteration": 2.6074445247650146
    },
    {
      "auxiliary_loss_clip": 0.01176091,
      "auxiliary_loss_mlp": 0.01028774,
      "balance_loss_clip": 1.05228174,
      "balance_loss_mlp": 1.0209775,
      "epoch": 0.6187699152287621,
      "flos": 14456630522880.0,
      "grad_norm": 7.977763106402264,
      "language_loss": 0.85461211,
      "learning_rate": 1.340334763646812e-06,
      "loss": 0.87666082,
      "num_input_tokens_seen": 110943000,
      "step": 5146,
      "time_per_iteration": 2.5629332065582275
    },
    {
      "auxiliary_loss_clip": 0.01176824,
      "auxiliary_loss_mlp": 0.01027493,
      "balance_loss_clip": 1.05068302,
      "balance_loss_mlp": 1.01919627,
      "epoch": 0.6188901581194012,
      "flos": 20084766796800.0,
      "grad_norm": 1.8946262180921278,
      "language_loss": 0.74304849,
      "learning_rate": 1.3395994337514218e-06,
      "loss": 0.76509166,
      "num_input_tokens_seen": 110963170,
      "step": 5147,
      "time_per_iteration": 2.599375009536743
    },
    {
      "auxiliary_loss_clip": 0.0116973,
      "auxiliary_loss_mlp": 0.01023619,
      "balance_loss_clip": 1.04860377,
      "balance_loss_mlp": 1.01668358,
      "epoch": 0.6190104010100402,
      "flos": 25700728360320.0,
      "grad_norm": 2.134575688603786,
      "language_loss": 0.7886976,
      "learning_rate": 1.3388642040382725e-06,
      "loss": 0.81063116,
      "num_input_tokens_seen": 110983595,
      "step": 5148,
      "time_per_iteration": 2.654026508331299
    },
    {
      "auxiliary_loss_clip": 0.01175536,
      "auxiliary_loss_mlp": 0.01024719,
      "balance_loss_clip": 1.05213046,
      "balance_loss_mlp": 1.01714349,
      "epoch": 0.6191306439006794,
      "flos": 30442372974720.0,
      "grad_norm": 2.026552096143354,
      "language_loss": 0.84451187,
      "learning_rate": 1.3381290746188975e-06,
      "loss": 0.86651438,
      "num_input_tokens_seen": 111002965,
      "step": 5149,
      "time_per_iteration": 2.649789810180664
    },
    {
      "auxiliary_loss_clip": 0.01179137,
      "auxiliary_loss_mlp": 0.01029181,
      "balance_loss_clip": 1.05480075,
      "balance_loss_mlp": 1.02106929,
      "epoch": 0.6192508867913185,
      "flos": 26685793918080.0,
      "grad_norm": 1.6859705704585746,
      "language_loss": 0.67356956,
      "learning_rate": 1.3373940456048152e-06,
      "loss": 0.69565272,
      "num_input_tokens_seen": 111022990,
      "step": 5150,
      "time_per_iteration": 2.6282873153686523
    },
    {
      "auxiliary_loss_clip": 0.01173934,
      "auxiliary_loss_mlp": 0.01028734,
      "balance_loss_clip": 1.05124497,
      "balance_loss_mlp": 1.02130127,
      "epoch": 0.6193711296819575,
      "flos": 36722036090880.0,
      "grad_norm": 1.9412449101424685,
      "language_loss": 0.59738362,
      "learning_rate": 1.3366591171075299e-06,
      "loss": 0.61941028,
      "num_input_tokens_seen": 111046495,
      "step": 5151,
      "time_per_iteration": 2.7503855228424072
    },
    {
      "auxiliary_loss_clip": 0.01173456,
      "auxiliary_loss_mlp": 0.01021698,
      "balance_loss_clip": 1.04955125,
      "balance_loss_mlp": 1.01434922,
      "epoch": 0.6194913725725967,
      "flos": 25192556697600.0,
      "grad_norm": 2.2273376180619504,
      "language_loss": 0.90850002,
      "learning_rate": 1.335924289238529e-06,
      "loss": 0.93045151,
      "num_input_tokens_seen": 111065705,
      "step": 5152,
      "time_per_iteration": 2.5998282432556152
    },
    {
      "auxiliary_loss_clip": 0.01177853,
      "auxiliary_loss_mlp": 0.01027792,
      "balance_loss_clip": 1.05320621,
      "balance_loss_mlp": 1.02001357,
      "epoch": 0.6196116154632357,
      "flos": 21178821196800.0,
      "grad_norm": 1.5430428923105102,
      "language_loss": 0.76495588,
      "learning_rate": 1.3351895621092859e-06,
      "loss": 0.78701234,
      "num_input_tokens_seen": 111086050,
      "step": 5153,
      "time_per_iteration": 2.7147178649902344
    },
    {
      "auxiliary_loss_clip": 0.0116991,
      "auxiliary_loss_mlp": 0.01024871,
      "balance_loss_clip": 1.04882073,
      "balance_loss_mlp": 1.01719344,
      "epoch": 0.6197318583538748,
      "flos": 16253744803200.0,
      "grad_norm": 12.462627068857374,
      "language_loss": 0.76429212,
      "learning_rate": 1.3344549358312567e-06,
      "loss": 0.78623998,
      "num_input_tokens_seen": 111104450,
      "step": 5154,
      "time_per_iteration": 2.5869462490081787
    },
    {
      "auxiliary_loss_clip": 0.01179116,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 1.05255735,
      "balance_loss_mlp": 1.01733184,
      "epoch": 0.619852101244514,
      "flos": 24425612478720.0,
      "grad_norm": 1.883712492965628,
      "language_loss": 0.7834751,
      "learning_rate": 1.3337204105158852e-06,
      "loss": 0.80551815,
      "num_input_tokens_seen": 111123320,
      "step": 5155,
      "time_per_iteration": 2.6429827213287354
    },
    {
      "auxiliary_loss_clip": 0.01169359,
      "auxiliary_loss_mlp": 0.01021419,
      "balance_loss_clip": 1.0463953,
      "balance_loss_mlp": 1.01401627,
      "epoch": 0.619972344135153,
      "flos": 16727298733440.0,
      "grad_norm": 2.821660581934745,
      "language_loss": 0.72972751,
      "learning_rate": 1.332985986274597e-06,
      "loss": 0.75163525,
      "num_input_tokens_seen": 111140950,
      "step": 5156,
      "time_per_iteration": 2.6342880725860596
    },
    {
      "auxiliary_loss_clip": 0.01175365,
      "auxiliary_loss_mlp": 0.01025361,
      "balance_loss_clip": 1.05424356,
      "balance_loss_mlp": 1.01846123,
      "epoch": 0.6200925870257921,
      "flos": 12495190498560.0,
      "grad_norm": 2.391972586590621,
      "language_loss": 0.75443935,
      "learning_rate": 1.3322516632188047e-06,
      "loss": 0.77644658,
      "num_input_tokens_seen": 111157845,
      "step": 5157,
      "time_per_iteration": 3.7124741077423096
    },
    {
      "auxiliary_loss_clip": 0.01174511,
      "auxiliary_loss_mlp": 0.01026151,
      "balance_loss_clip": 1.05057991,
      "balance_loss_mlp": 1.01834881,
      "epoch": 0.6202128299164312,
      "flos": 26539350168960.0,
      "grad_norm": 2.705069340143011,
      "language_loss": 0.67120767,
      "learning_rate": 1.3315174414599045e-06,
      "loss": 0.6932143,
      "num_input_tokens_seen": 111179165,
      "step": 5158,
      "time_per_iteration": 2.697495698928833
    },
    {
      "auxiliary_loss_clip": 0.01174889,
      "auxiliary_loss_mlp": 0.01023999,
      "balance_loss_clip": 1.05095422,
      "balance_loss_mlp": 1.01594031,
      "epoch": 0.6203330728070703,
      "flos": 18770508069120.0,
      "grad_norm": 1.867512417895808,
      "language_loss": 0.75607902,
      "learning_rate": 1.3307833211092768e-06,
      "loss": 0.77806789,
      "num_input_tokens_seen": 111197830,
      "step": 5159,
      "time_per_iteration": 2.637561321258545
    },
    {
      "auxiliary_loss_clip": 0.01177393,
      "auxiliary_loss_mlp": 0.01027393,
      "balance_loss_clip": 1.05336177,
      "balance_loss_mlp": 1.01920331,
      "epoch": 0.6204533156977093,
      "flos": 20629782835200.0,
      "grad_norm": 3.1993269129438837,
      "language_loss": 0.75243294,
      "learning_rate": 1.3300493022782873e-06,
      "loss": 0.77448082,
      "num_input_tokens_seen": 111218400,
      "step": 5160,
      "time_per_iteration": 2.6555469036102295
    },
    {
      "auxiliary_loss_clip": 0.01172835,
      "auxiliary_loss_mlp": 0.01027233,
      "balance_loss_clip": 1.04887879,
      "balance_loss_mlp": 1.01946688,
      "epoch": 0.6205735585883485,
      "flos": 17348050598400.0,
      "grad_norm": 5.4425228312251,
      "language_loss": 0.72772515,
      "learning_rate": 1.3293153850782855e-06,
      "loss": 0.74972588,
      "num_input_tokens_seen": 111236720,
      "step": 5161,
      "time_per_iteration": 4.403026580810547
    },
    {
      "auxiliary_loss_clip": 0.01176269,
      "auxiliary_loss_mlp": 0.01026165,
      "balance_loss_clip": 1.04880404,
      "balance_loss_mlp": 1.01789749,
      "epoch": 0.6206938014789876,
      "flos": 22965017742720.0,
      "grad_norm": 2.4225584803360065,
      "language_loss": 0.71613288,
      "learning_rate": 1.3285815696206069e-06,
      "loss": 0.73815721,
      "num_input_tokens_seen": 111258265,
      "step": 5162,
      "time_per_iteration": 3.5818769931793213
    },
    {
      "auxiliary_loss_clip": 0.01176119,
      "auxiliary_loss_mlp": 0.01024401,
      "balance_loss_clip": 1.04870534,
      "balance_loss_mlp": 1.01628232,
      "epoch": 0.6208140443696266,
      "flos": 23983192661760.0,
      "grad_norm": 1.9478490063317428,
      "language_loss": 0.77030665,
      "learning_rate": 1.32784785601657e-06,
      "loss": 0.79231179,
      "num_input_tokens_seen": 111277675,
      "step": 5163,
      "time_per_iteration": 2.6449685096740723
    },
    {
      "auxiliary_loss_clip": 0.01170809,
      "auxiliary_loss_mlp": 0.01026073,
      "balance_loss_clip": 1.04791892,
      "balance_loss_mlp": 1.018152,
      "epoch": 0.6209342872602658,
      "flos": 35077291303680.0,
      "grad_norm": 2.085444689782811,
      "language_loss": 0.73822314,
      "learning_rate": 1.3271142443774798e-06,
      "loss": 0.76019192,
      "num_input_tokens_seen": 111299910,
      "step": 5164,
      "time_per_iteration": 2.6948435306549072
    },
    {
      "auxiliary_loss_clip": 0.01171995,
      "auxiliary_loss_mlp": 0.01025546,
      "balance_loss_clip": 1.04900837,
      "balance_loss_mlp": 1.01803589,
      "epoch": 0.6210545301509048,
      "flos": 26979327861120.0,
      "grad_norm": 2.4726128307657746,
      "language_loss": 0.81408286,
      "learning_rate": 1.3263807348146228e-06,
      "loss": 0.83605826,
      "num_input_tokens_seen": 111319765,
      "step": 5165,
      "time_per_iteration": 2.6607887744903564
    },
    {
      "auxiliary_loss_clip": 0.01177427,
      "auxiliary_loss_mlp": 0.01027793,
      "balance_loss_clip": 1.05263615,
      "balance_loss_mlp": 1.01931107,
      "epoch": 0.6211747730415439,
      "flos": 33618240852480.0,
      "grad_norm": 2.1395931561589494,
      "language_loss": 0.73808277,
      "learning_rate": 1.3256473274392733e-06,
      "loss": 0.76013494,
      "num_input_tokens_seen": 111341110,
      "step": 5166,
      "time_per_iteration": 2.705648899078369
    },
    {
      "auxiliary_loss_clip": 0.01175092,
      "auxiliary_loss_mlp": 0.01026345,
      "balance_loss_clip": 1.05067563,
      "balance_loss_mlp": 1.01903772,
      "epoch": 0.6212950159321831,
      "flos": 34167099646080.0,
      "grad_norm": 1.9410894532115872,
      "language_loss": 0.70204914,
      "learning_rate": 1.3249140223626873e-06,
      "loss": 0.72406352,
      "num_input_tokens_seen": 111362730,
      "step": 5167,
      "time_per_iteration": 2.7335205078125
    },
    {
      "auxiliary_loss_clip": 0.01173298,
      "auxiliary_loss_mlp": 0.01022151,
      "balance_loss_clip": 1.05141377,
      "balance_loss_mlp": 1.01428294,
      "epoch": 0.6214152588228221,
      "flos": 27965758135680.0,
      "grad_norm": 2.098772218325719,
      "language_loss": 0.75630575,
      "learning_rate": 1.3241808196961077e-06,
      "loss": 0.77826023,
      "num_input_tokens_seen": 111383855,
      "step": 5168,
      "time_per_iteration": 2.7433853149414062
    },
    {
      "auxiliary_loss_clip": 0.01172285,
      "auxiliary_loss_mlp": 0.01026103,
      "balance_loss_clip": 1.05105245,
      "balance_loss_mlp": 1.01887345,
      "epoch": 0.6215355017134612,
      "flos": 20230204965120.0,
      "grad_norm": 2.510701422239912,
      "language_loss": 0.70529425,
      "learning_rate": 1.3234477195507608e-06,
      "loss": 0.72727823,
      "num_input_tokens_seen": 111402685,
      "step": 5169,
      "time_per_iteration": 2.668038845062256
    },
    {
      "auxiliary_loss_clip": 0.01171959,
      "auxiliary_loss_mlp": 0.01028758,
      "balance_loss_clip": 1.04940355,
      "balance_loss_mlp": 1.02071702,
      "epoch": 0.6216557446041003,
      "flos": 41428129219200.0,
      "grad_norm": 2.430380332579137,
      "language_loss": 0.62944341,
      "learning_rate": 1.322714722037857e-06,
      "loss": 0.65145051,
      "num_input_tokens_seen": 111424130,
      "step": 5170,
      "time_per_iteration": 2.788257360458374
    },
    {
      "auxiliary_loss_clip": 0.01179224,
      "auxiliary_loss_mlp": 0.01034693,
      "balance_loss_clip": 1.05302405,
      "balance_loss_mlp": 1.02562761,
      "epoch": 0.6217759874947394,
      "flos": 27928770105600.0,
      "grad_norm": 2.519945967028375,
      "language_loss": 0.77391738,
      "learning_rate": 1.321981827268591e-06,
      "loss": 0.79605651,
      "num_input_tokens_seen": 111444785,
      "step": 5171,
      "time_per_iteration": 2.6870367527008057
    },
    {
      "auxiliary_loss_clip": 0.01174271,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.0508275,
      "balance_loss_mlp": 1.01749027,
      "epoch": 0.6218962303853784,
      "flos": 21765673601280.0,
      "grad_norm": 1.8043949714512408,
      "language_loss": 0.81649733,
      "learning_rate": 1.3212490353541426e-06,
      "loss": 0.83848929,
      "num_input_tokens_seen": 111467045,
      "step": 5172,
      "time_per_iteration": 2.7509939670562744
    },
    {
      "auxiliary_loss_clip": 0.01176576,
      "auxiliary_loss_mlp": 0.01028268,
      "balance_loss_clip": 1.05125391,
      "balance_loss_mlp": 1.01975608,
      "epoch": 0.6220164732760175,
      "flos": 21246260981760.0,
      "grad_norm": 2.1700971328501386,
      "language_loss": 0.80243897,
      "learning_rate": 1.3205163464056762e-06,
      "loss": 0.82448745,
      "num_input_tokens_seen": 111483650,
      "step": 5173,
      "time_per_iteration": 2.6365256309509277
    },
    {
      "auxiliary_loss_clip": 0.01172382,
      "auxiliary_loss_mlp": 0.01025563,
      "balance_loss_clip": 1.04857862,
      "balance_loss_mlp": 1.01757026,
      "epoch": 0.6221367161666567,
      "flos": 26136360506880.0,
      "grad_norm": 2.2667723202364427,
      "language_loss": 0.73358262,
      "learning_rate": 1.319783760534339e-06,
      "loss": 0.75556213,
      "num_input_tokens_seen": 111502895,
      "step": 5174,
      "time_per_iteration": 2.69812273979187
    },
    {
      "auxiliary_loss_clip": 0.01176032,
      "auxiliary_loss_mlp": 0.01028327,
      "balance_loss_clip": 1.05115867,
      "balance_loss_mlp": 1.02029884,
      "epoch": 0.6222569590572957,
      "flos": 16284196558080.0,
      "grad_norm": 2.2638829454859764,
      "language_loss": 0.75260818,
      "learning_rate": 1.319051277851266e-06,
      "loss": 0.77465177,
      "num_input_tokens_seen": 111519180,
      "step": 5175,
      "time_per_iteration": 2.608844041824341
    },
    {
      "auxiliary_loss_clip": 0.0117444,
      "auxiliary_loss_mlp": 0.01024669,
      "balance_loss_clip": 1.05165207,
      "balance_loss_mlp": 1.01795805,
      "epoch": 0.6223772019479348,
      "flos": 18223840005120.0,
      "grad_norm": 2.4771631854577807,
      "language_loss": 0.83829069,
      "learning_rate": 1.3183188984675716e-06,
      "loss": 0.86028177,
      "num_input_tokens_seen": 111537545,
      "step": 5176,
      "time_per_iteration": 2.6809792518615723
    },
    {
      "auxiliary_loss_clip": 0.01179178,
      "auxiliary_loss_mlp": 0.01034108,
      "balance_loss_clip": 1.05292296,
      "balance_loss_mlp": 1.02621663,
      "epoch": 0.6224974448385739,
      "flos": 27489797994240.0,
      "grad_norm": 2.8832489446063754,
      "language_loss": 0.70905852,
      "learning_rate": 1.3175866224943586e-06,
      "loss": 0.7311914,
      "num_input_tokens_seen": 111556265,
      "step": 5177,
      "time_per_iteration": 2.668003559112549
    },
    {
      "auxiliary_loss_clip": 0.01176229,
      "auxiliary_loss_mlp": 0.01030545,
      "balance_loss_clip": 1.05027151,
      "balance_loss_mlp": 1.02218509,
      "epoch": 0.622617687729213,
      "flos": 19791951125760.0,
      "grad_norm": 2.2322284147539193,
      "language_loss": 0.73295176,
      "learning_rate": 1.316854450042712e-06,
      "loss": 0.75501949,
      "num_input_tokens_seen": 111574205,
      "step": 5178,
      "time_per_iteration": 2.6992998123168945
    },
    {
      "auxiliary_loss_clip": 0.01179741,
      "auxiliary_loss_mlp": 0.01026602,
      "balance_loss_clip": 1.05307209,
      "balance_loss_mlp": 1.01808405,
      "epoch": 0.622737930619852,
      "flos": 23038886062080.0,
      "grad_norm": 6.1799218556161515,
      "language_loss": 0.74396515,
      "learning_rate": 1.3161223812237024e-06,
      "loss": 0.76602852,
      "num_input_tokens_seen": 111593560,
      "step": 5179,
      "time_per_iteration": 2.6287968158721924
    },
    {
      "auxiliary_loss_clip": 0.01172818,
      "auxiliary_loss_mlp": 0.01030101,
      "balance_loss_clip": 1.04803145,
      "balance_loss_mlp": 1.02207184,
      "epoch": 0.6228581735104912,
      "flos": 12634271959680.0,
      "grad_norm": 3.158975279595604,
      "language_loss": 0.85749704,
      "learning_rate": 1.3153904161483842e-06,
      "loss": 0.87952626,
      "num_input_tokens_seen": 111608860,
      "step": 5180,
      "time_per_iteration": 2.6097252368927
    },
    {
      "auxiliary_loss_clip": 0.01176814,
      "auxiliary_loss_mlp": 0.01024717,
      "balance_loss_clip": 1.04958844,
      "balance_loss_mlp": 1.01677132,
      "epoch": 0.6229784164011303,
      "flos": 23802813538560.0,
      "grad_norm": 2.0494098548442405,
      "language_loss": 0.85563779,
      "learning_rate": 1.3146585549277953e-06,
      "loss": 0.87765312,
      "num_input_tokens_seen": 111627500,
      "step": 5181,
      "time_per_iteration": 2.700728416442871
    },
    {
      "auxiliary_loss_clip": 0.01176935,
      "auxiliary_loss_mlp": 0.01032597,
      "balance_loss_clip": 1.05264866,
      "balance_loss_mlp": 1.0243597,
      "epoch": 0.6230986592917693,
      "flos": 22414219614720.0,
      "grad_norm": 2.1295154892432704,
      "language_loss": 0.78612971,
      "learning_rate": 1.3139267976729591e-06,
      "loss": 0.8082251,
      "num_input_tokens_seen": 111647690,
      "step": 5182,
      "time_per_iteration": 2.686065912246704
    },
    {
      "auxiliary_loss_clip": 0.01176173,
      "auxiliary_loss_mlp": 0.01031409,
      "balance_loss_clip": 1.05183959,
      "balance_loss_mlp": 1.02343941,
      "epoch": 0.6232189021824085,
      "flos": 34528217028480.0,
      "grad_norm": 2.5823733422308996,
      "language_loss": 0.72387648,
      "learning_rate": 1.3131951444948815e-06,
      "loss": 0.74595225,
      "num_input_tokens_seen": 111667090,
      "step": 5183,
      "time_per_iteration": 3.6921448707580566
    },
    {
      "auxiliary_loss_clip": 0.01175788,
      "auxiliary_loss_mlp": 0.01027786,
      "balance_loss_clip": 1.05241537,
      "balance_loss_mlp": 1.01949537,
      "epoch": 0.6233391450730476,
      "flos": 22237000888320.0,
      "grad_norm": 2.3521703975179618,
      "language_loss": 0.76762271,
      "learning_rate": 1.3124635955045546e-06,
      "loss": 0.78965843,
      "num_input_tokens_seen": 111686905,
      "step": 5184,
      "time_per_iteration": 2.6093709468841553
    },
    {
      "auxiliary_loss_clip": 0.01174893,
      "auxiliary_loss_mlp": 0.0102565,
      "balance_loss_clip": 1.05018079,
      "balance_loss_mlp": 1.01763272,
      "epoch": 0.6234593879636866,
      "flos": 20332693445760.0,
      "grad_norm": 1.9568989637123944,
      "language_loss": 0.84285879,
      "learning_rate": 1.3117321508129537e-06,
      "loss": 0.86486423,
      "num_input_tokens_seen": 111704985,
      "step": 5185,
      "time_per_iteration": 2.6977431774139404
    },
    {
      "auxiliary_loss_clip": 0.01180901,
      "auxiliary_loss_mlp": 0.01029561,
      "balance_loss_clip": 1.05517066,
      "balance_loss_mlp": 1.02198577,
      "epoch": 0.6235796308543258,
      "flos": 20664903358080.0,
      "grad_norm": 2.0723433168776197,
      "language_loss": 0.76625639,
      "learning_rate": 1.3110008105310388e-06,
      "loss": 0.78836101,
      "num_input_tokens_seen": 111724805,
      "step": 5186,
      "time_per_iteration": 2.6376523971557617
    },
    {
      "auxiliary_loss_clip": 0.0117629,
      "auxiliary_loss_mlp": 0.0102014,
      "balance_loss_clip": 1.0483048,
      "balance_loss_mlp": 1.01181889,
      "epoch": 0.6236998737449648,
      "flos": 26618641441920.0,
      "grad_norm": 1.754790984289786,
      "language_loss": 0.77944458,
      "learning_rate": 1.3102695747697526e-06,
      "loss": 0.80140889,
      "num_input_tokens_seen": 111747675,
      "step": 5187,
      "time_per_iteration": 3.5557403564453125
    },
    {
      "auxiliary_loss_clip": 0.01177352,
      "auxiliary_loss_mlp": 0.01030848,
      "balance_loss_clip": 1.05242014,
      "balance_loss_mlp": 1.02248001,
      "epoch": 0.6238201166356039,
      "flos": 12674599954560.0,
      "grad_norm": 4.142378256262315,
      "language_loss": 0.90466875,
      "learning_rate": 1.3095384436400237e-06,
      "loss": 0.92675078,
      "num_input_tokens_seen": 111759205,
      "step": 5188,
      "time_per_iteration": 3.4899203777313232
    },
    {
      "auxiliary_loss_clip": 0.01175098,
      "auxiliary_loss_mlp": 0.01033098,
      "balance_loss_clip": 1.05011475,
      "balance_loss_mlp": 1.02443182,
      "epoch": 0.623940359526243,
      "flos": 10452160730880.0,
      "grad_norm": 2.108017861230039,
      "language_loss": 0.82743335,
      "learning_rate": 1.3088074172527633e-06,
      "loss": 0.84951532,
      "num_input_tokens_seen": 111776335,
      "step": 5189,
      "time_per_iteration": 2.59865403175354
    },
    {
      "auxiliary_loss_clip": 0.01178052,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.04960179,
      "balance_loss_mlp": 1.01653886,
      "epoch": 0.6240606024168821,
      "flos": 29059525226880.0,
      "grad_norm": 5.2023621383830125,
      "language_loss": 0.71866518,
      "learning_rate": 1.3080764957188684e-06,
      "loss": 0.7406894,
      "num_input_tokens_seen": 111796580,
      "step": 5190,
      "time_per_iteration": 3.5522334575653076
    },
    {
      "auxiliary_loss_clip": 0.0117855,
      "auxiliary_loss_mlp": 0.01024123,
      "balance_loss_clip": 1.05177426,
      "balance_loss_mlp": 1.01609445,
      "epoch": 0.6241808453075212,
      "flos": 22018089450240.0,
      "grad_norm": 2.00942092523159,
      "language_loss": 0.7057851,
      "learning_rate": 1.3073456791492192e-06,
      "loss": 0.72781181,
      "num_input_tokens_seen": 111816290,
      "step": 5191,
      "time_per_iteration": 2.603520154953003
    },
    {
      "auxiliary_loss_clip": 0.01175628,
      "auxiliary_loss_mlp": 0.01023851,
      "balance_loss_clip": 1.04935694,
      "balance_loss_mlp": 1.01615572,
      "epoch": 0.6243010881981603,
      "flos": 21138708683520.0,
      "grad_norm": 3.2278978940225933,
      "language_loss": 0.78370011,
      "learning_rate": 1.3066149676546801e-06,
      "loss": 0.80569494,
      "num_input_tokens_seen": 111834470,
      "step": 5192,
      "time_per_iteration": 2.5771942138671875
    },
    {
      "auxiliary_loss_clip": 0.0117677,
      "auxiliary_loss_mlp": 0.01029502,
      "balance_loss_clip": 1.05251026,
      "balance_loss_mlp": 1.02190208,
      "epoch": 0.6244213310887994,
      "flos": 22344948236160.0,
      "grad_norm": 1.7531497091315642,
      "language_loss": 0.66262436,
      "learning_rate": 1.3058843613460985e-06,
      "loss": 0.68468708,
      "num_input_tokens_seen": 111852410,
      "step": 5193,
      "time_per_iteration": 2.58191180229187
    },
    {
      "auxiliary_loss_clip": 0.01176918,
      "auxiliary_loss_mlp": 0.01023705,
      "balance_loss_clip": 1.05031598,
      "balance_loss_mlp": 1.01556265,
      "epoch": 0.6245415739794384,
      "flos": 15231978524160.0,
      "grad_norm": 3.6889491747049634,
      "language_loss": 0.74836946,
      "learning_rate": 1.3051538603343075e-06,
      "loss": 0.77037573,
      "num_input_tokens_seen": 111870340,
      "step": 5194,
      "time_per_iteration": 2.6258018016815186
    },
    {
      "auxiliary_loss_clip": 0.01178592,
      "auxiliary_loss_mlp": 0.01032622,
      "balance_loss_clip": 1.05312109,
      "balance_loss_mlp": 1.02434337,
      "epoch": 0.6246618168700776,
      "flos": 18879891960960.0,
      "grad_norm": 2.01984845194897,
      "language_loss": 0.68089449,
      "learning_rate": 1.3044234647301235e-06,
      "loss": 0.70300663,
      "num_input_tokens_seen": 111888365,
      "step": 5195,
      "time_per_iteration": 2.5912246704101562
    },
    {
      "auxiliary_loss_clip": 0.01173975,
      "auxiliary_loss_mlp": 0.01027365,
      "balance_loss_clip": 1.05028105,
      "balance_loss_mlp": 1.01995015,
      "epoch": 0.6247820597607167,
      "flos": 14319201087360.0,
      "grad_norm": 2.063889477688418,
      "language_loss": 0.72566938,
      "learning_rate": 1.303693174644347e-06,
      "loss": 0.74768275,
      "num_input_tokens_seen": 111905840,
      "step": 5196,
      "time_per_iteration": 2.6334402561187744
    },
    {
      "auxiliary_loss_clip": 0.01176118,
      "auxiliary_loss_mlp": 0.01025876,
      "balance_loss_clip": 1.0504427,
      "balance_loss_mlp": 1.01793063,
      "epoch": 0.6249023026513557,
      "flos": 22637979388800.0,
      "grad_norm": 2.2319526351501553,
      "language_loss": 0.80853838,
      "learning_rate": 1.3029629901877625e-06,
      "loss": 0.8305583,
      "num_input_tokens_seen": 111925215,
      "step": 5197,
      "time_per_iteration": 2.7406351566314697
    },
    {
      "auxiliary_loss_clip": 0.01180877,
      "auxiliary_loss_mlp": 0.01027909,
      "balance_loss_clip": 1.05346084,
      "balance_loss_mlp": 1.01954651,
      "epoch": 0.6250225455419949,
      "flos": 20266690204800.0,
      "grad_norm": 3.735147918292609,
      "language_loss": 0.77391934,
      "learning_rate": 1.3022329114711376e-06,
      "loss": 0.79600716,
      "num_input_tokens_seen": 111943925,
      "step": 5198,
      "time_per_iteration": 2.5488321781158447
    },
    {
      "auxiliary_loss_clip": 0.01174136,
      "auxiliary_loss_mlp": 0.01026758,
      "balance_loss_clip": 1.0507127,
      "balance_loss_mlp": 1.01792502,
      "epoch": 0.6251427884326339,
      "flos": 23437853400960.0,
      "grad_norm": 2.263588897473636,
      "language_loss": 0.69560027,
      "learning_rate": 1.3015029386052256e-06,
      "loss": 0.71760923,
      "num_input_tokens_seen": 111964095,
      "step": 5199,
      "time_per_iteration": 2.7265567779541016
    },
    {
      "auxiliary_loss_clip": 0.01180151,
      "auxiliary_loss_mlp": 0.010284,
      "balance_loss_clip": 1.05263555,
      "balance_loss_mlp": 1.02020168,
      "epoch": 0.625263031323273,
      "flos": 31723055464320.0,
      "grad_norm": 2.017422822753896,
      "language_loss": 0.72836453,
      "learning_rate": 1.3007730717007622e-06,
      "loss": 0.75045002,
      "num_input_tokens_seen": 111984910,
      "step": 5200,
      "time_per_iteration": 2.68269419670105
    },
    {
      "auxiliary_loss_clip": 0.01180112,
      "auxiliary_loss_mlp": 0.0102511,
      "balance_loss_clip": 1.05297637,
      "balance_loss_mlp": 1.01611578,
      "epoch": 0.6253832742139122,
      "flos": 24134341092480.0,
      "grad_norm": 2.4913893349787704,
      "language_loss": 0.75308943,
      "learning_rate": 1.3000433108684676e-06,
      "loss": 0.7751416,
      "num_input_tokens_seen": 112005410,
      "step": 5201,
      "time_per_iteration": 2.6409401893615723
    },
    {
      "auxiliary_loss_clip": 0.01176155,
      "auxiliary_loss_mlp": 0.01024293,
      "balance_loss_clip": 1.05015683,
      "balance_loss_mlp": 1.016258,
      "epoch": 0.6255035171045512,
      "flos": 27668812400640.0,
      "grad_norm": 2.6819548756825706,
      "language_loss": 0.80539489,
      "learning_rate": 1.2993136562190467e-06,
      "loss": 0.82739937,
      "num_input_tokens_seen": 112024530,
      "step": 5202,
      "time_per_iteration": 2.620649576187134
    },
    {
      "auxiliary_loss_clip": 0.01180234,
      "auxiliary_loss_mlp": 0.01029624,
      "balance_loss_clip": 1.05329275,
      "balance_loss_mlp": 1.02148759,
      "epoch": 0.6256237599951903,
      "flos": 20227798753920.0,
      "grad_norm": 1.6229886965823426,
      "language_loss": 0.70458597,
      "learning_rate": 1.2985841078631871e-06,
      "loss": 0.72668457,
      "num_input_tokens_seen": 112043850,
      "step": 5203,
      "time_per_iteration": 2.612119436264038
    },
    {
      "auxiliary_loss_clip": 0.01176077,
      "auxiliary_loss_mlp": 0.01024,
      "balance_loss_clip": 1.04954243,
      "balance_loss_mlp": 1.01595318,
      "epoch": 0.6257440028858293,
      "flos": 24170574936960.0,
      "grad_norm": 1.8729620230634039,
      "language_loss": 0.78178704,
      "learning_rate": 1.2978546659115608e-06,
      "loss": 0.80378777,
      "num_input_tokens_seen": 112061930,
      "step": 5204,
      "time_per_iteration": 2.579899787902832
    },
    {
      "auxiliary_loss_clip": 0.01177447,
      "auxiliary_loss_mlp": 0.01030155,
      "balance_loss_clip": 1.05323911,
      "balance_loss_mlp": 1.02243674,
      "epoch": 0.6258642457764685,
      "flos": 15851940289920.0,
      "grad_norm": 2.30228607004877,
      "language_loss": 0.85514176,
      "learning_rate": 1.2971253304748228e-06,
      "loss": 0.87721777,
      "num_input_tokens_seen": 112079645,
      "step": 5205,
      "time_per_iteration": 2.5996217727661133
    },
    {
      "auxiliary_loss_clip": 0.01182031,
      "auxiliary_loss_mlp": 0.01026687,
      "balance_loss_clip": 1.0539763,
      "balance_loss_mlp": 1.01811028,
      "epoch": 0.6259844886671075,
      "flos": 11911354836480.0,
      "grad_norm": 2.1892029739728156,
      "language_loss": 0.74827218,
      "learning_rate": 1.296396101663614e-06,
      "loss": 0.7703594,
      "num_input_tokens_seen": 112096205,
      "step": 5206,
      "time_per_iteration": 2.5544381141662598
    },
    {
      "auxiliary_loss_clip": 0.01177512,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.05301738,
      "balance_loss_mlp": 1.02073491,
      "epoch": 0.6261047315577466,
      "flos": 15887958652800.0,
      "grad_norm": 2.3291508582340446,
      "language_loss": 0.84253597,
      "learning_rate": 1.2956669795885565e-06,
      "loss": 0.86459327,
      "num_input_tokens_seen": 112112835,
      "step": 5207,
      "time_per_iteration": 2.6064887046813965
    },
    {
      "auxiliary_loss_clip": 0.01179424,
      "auxiliary_loss_mlp": 0.01040152,
      "balance_loss_clip": 1.0534966,
      "balance_loss_mlp": 1.03091347,
      "epoch": 0.6262249744483858,
      "flos": 31248926916480.0,
      "grad_norm": 1.9370940078767092,
      "language_loss": 0.68378043,
      "learning_rate": 1.294937964360259e-06,
      "loss": 0.70597625,
      "num_input_tokens_seen": 112133105,
      "step": 5208,
      "time_per_iteration": 2.6706809997558594
    },
    {
      "auxiliary_loss_clip": 0.01176538,
      "auxiliary_loss_mlp": 0.0102753,
      "balance_loss_clip": 1.05036664,
      "balance_loss_mlp": 1.0186851,
      "epoch": 0.6263452173390248,
      "flos": 27198598435200.0,
      "grad_norm": 3.0954747867371544,
      "language_loss": 0.71050453,
      "learning_rate": 1.2942090560893108e-06,
      "loss": 0.73254514,
      "num_input_tokens_seen": 112152510,
      "step": 5209,
      "time_per_iteration": 2.7160487174987793
    },
    {
      "auxiliary_loss_clip": 0.01175832,
      "auxiliary_loss_mlp": 0.01029743,
      "balance_loss_clip": 1.05159569,
      "balance_loss_mlp": 1.0223043,
      "epoch": 0.6264654602296639,
      "flos": 37342069683840.0,
      "grad_norm": 2.758649371053003,
      "language_loss": 0.60699499,
      "learning_rate": 1.2934802548862882e-06,
      "loss": 0.62905073,
      "num_input_tokens_seen": 112175295,
      "step": 5210,
      "time_per_iteration": 3.6953752040863037
    },
    {
      "auxiliary_loss_clip": 0.01174611,
      "auxiliary_loss_mlp": 0.01021195,
      "balance_loss_clip": 1.04841638,
      "balance_loss_mlp": 1.01290369,
      "epoch": 0.626585703120303,
      "flos": 14756952136320.0,
      "grad_norm": 2.1892492153408347,
      "language_loss": 0.82316124,
      "learning_rate": 1.292751560861749e-06,
      "loss": 0.8451193,
      "num_input_tokens_seen": 112190200,
      "step": 5211,
      "time_per_iteration": 2.5837106704711914
    },
    {
      "auxiliary_loss_clip": 0.01180157,
      "auxiliary_loss_mlp": 0.01029322,
      "balance_loss_clip": 1.05288446,
      "balance_loss_mlp": 1.02066183,
      "epoch": 0.6267059460109421,
      "flos": 22347318533760.0,
      "grad_norm": 1.9002335218081503,
      "language_loss": 0.79455233,
      "learning_rate": 1.2920229741262354e-06,
      "loss": 0.81664711,
      "num_input_tokens_seen": 112208205,
      "step": 5212,
      "time_per_iteration": 2.6758499145507812
    },
    {
      "auxiliary_loss_clip": 0.0117786,
      "auxiliary_loss_mlp": 0.01026187,
      "balance_loss_clip": 1.05301666,
      "balance_loss_mlp": 1.01812863,
      "epoch": 0.6268261889015811,
      "flos": 17748813617280.0,
      "grad_norm": 2.1321787022651453,
      "language_loss": 0.75340337,
      "learning_rate": 1.2912944947902739e-06,
      "loss": 0.77544379,
      "num_input_tokens_seen": 112224690,
      "step": 5213,
      "time_per_iteration": 2.5682694911956787
    },
    {
      "auxiliary_loss_clip": 0.01179654,
      "auxiliary_loss_mlp": 0.01025565,
      "balance_loss_clip": 1.05163956,
      "balance_loss_mlp": 1.01720846,
      "epoch": 0.6269464317922203,
      "flos": 32846484211200.0,
      "grad_norm": 2.5622929402394328,
      "language_loss": 0.71899474,
      "learning_rate": 1.2905661229643742e-06,
      "loss": 0.74104691,
      "num_input_tokens_seen": 112244450,
      "step": 5214,
      "time_per_iteration": 3.6811349391937256
    },
    {
      "auxiliary_loss_clip": 0.01176983,
      "auxiliary_loss_mlp": 0.01029221,
      "balance_loss_clip": 1.04990649,
      "balance_loss_mlp": 1.02046537,
      "epoch": 0.6270666746828594,
      "flos": 17929192740480.0,
      "grad_norm": 2.3138802274667345,
      "language_loss": 0.84400558,
      "learning_rate": 1.2898378587590299e-06,
      "loss": 0.86606765,
      "num_input_tokens_seen": 112261050,
      "step": 5215,
      "time_per_iteration": 3.5509324073791504
    },
    {
      "auxiliary_loss_clip": 0.01176189,
      "auxiliary_loss_mlp": 0.01022279,
      "balance_loss_clip": 1.05038655,
      "balance_loss_mlp": 1.01440501,
      "epoch": 0.6271869175734984,
      "flos": 17457326749440.0,
      "grad_norm": 1.9045172542895294,
      "language_loss": 0.8753376,
      "learning_rate": 1.2891097022847173e-06,
      "loss": 0.8973223,
      "num_input_tokens_seen": 112278395,
      "step": 5216,
      "time_per_iteration": 2.591048002243042
    },
    {
      "auxiliary_loss_clip": 0.01181981,
      "auxiliary_loss_mlp": 0.0102981,
      "balance_loss_clip": 1.05219221,
      "balance_loss_mlp": 1.02045798,
      "epoch": 0.6273071604641376,
      "flos": 26868615166080.0,
      "grad_norm": 1.8258039643661412,
      "language_loss": 0.66573018,
      "learning_rate": 1.2883816536518978e-06,
      "loss": 0.68784803,
      "num_input_tokens_seen": 112299535,
      "step": 5217,
      "time_per_iteration": 3.6176881790161133
    },
    {
      "auxiliary_loss_clip": 0.01175071,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.05081081,
      "balance_loss_mlp": 1.01927733,
      "epoch": 0.6274274033547766,
      "flos": 26062384446720.0,
      "grad_norm": 2.0013998770977515,
      "language_loss": 0.82033467,
      "learning_rate": 1.2876537129710155e-06,
      "loss": 0.84235692,
      "num_input_tokens_seen": 112317265,
      "step": 5218,
      "time_per_iteration": 2.678267240524292
    },
    {
      "auxiliary_loss_clip": 0.01182091,
      "auxiliary_loss_mlp": 0.01029151,
      "balance_loss_clip": 1.05394936,
      "balance_loss_mlp": 1.01951289,
      "epoch": 0.6275476462454157,
      "flos": 20266259241600.0,
      "grad_norm": 1.971677143788869,
      "language_loss": 0.75274551,
      "learning_rate": 1.286925880352499e-06,
      "loss": 0.77485794,
      "num_input_tokens_seen": 112336125,
      "step": 5219,
      "time_per_iteration": 2.682825803756714
    },
    {
      "auxiliary_loss_clip": 0.01178501,
      "auxiliary_loss_mlp": 0.01027081,
      "balance_loss_clip": 1.05235338,
      "balance_loss_mlp": 1.01842654,
      "epoch": 0.6276678891360549,
      "flos": 26320402817280.0,
      "grad_norm": 4.222344474848088,
      "language_loss": 0.71142191,
      "learning_rate": 1.2861981559067592e-06,
      "loss": 0.73347771,
      "num_input_tokens_seen": 112356730,
      "step": 5220,
      "time_per_iteration": 2.6677050590515137
    },
    {
      "auxiliary_loss_clip": 0.01175654,
      "auxiliary_loss_mlp": 0.0102618,
      "balance_loss_clip": 1.05066121,
      "balance_loss_mlp": 1.01882148,
      "epoch": 0.6277881320266939,
      "flos": 13912512324480.0,
      "grad_norm": 2.947085410706905,
      "language_loss": 0.80397403,
      "learning_rate": 1.2854705397441917e-06,
      "loss": 0.82599235,
      "num_input_tokens_seen": 112372270,
      "step": 5221,
      "time_per_iteration": 2.6262905597686768
    },
    {
      "auxiliary_loss_clip": 0.01175197,
      "auxiliary_loss_mlp": 0.01026849,
      "balance_loss_clip": 1.04960215,
      "balance_loss_mlp": 1.0184567,
      "epoch": 0.627908374917333,
      "flos": 27048922462080.0,
      "grad_norm": 2.4493903172272047,
      "language_loss": 0.77660149,
      "learning_rate": 1.2847430319751747e-06,
      "loss": 0.79862195,
      "num_input_tokens_seen": 112390365,
      "step": 5222,
      "time_per_iteration": 2.7033450603485107
    },
    {
      "auxiliary_loss_clip": 0.01178169,
      "auxiliary_loss_mlp": 0.01026206,
      "balance_loss_clip": 1.05381131,
      "balance_loss_mlp": 1.01839757,
      "epoch": 0.6280286178079721,
      "flos": 23769201386880.0,
      "grad_norm": 2.4377882804372484,
      "language_loss": 0.67728424,
      "learning_rate": 1.2840156327100712e-06,
      "loss": 0.69932801,
      "num_input_tokens_seen": 112407490,
      "step": 5223,
      "time_per_iteration": 2.5861103534698486
    },
    {
      "auxiliary_loss_clip": 0.01175543,
      "auxiliary_loss_mlp": 0.01029631,
      "balance_loss_clip": 1.05123937,
      "balance_loss_mlp": 1.02104247,
      "epoch": 0.6281488606986112,
      "flos": 26359150613760.0,
      "grad_norm": 1.8335576064335966,
      "language_loss": 0.72466731,
      "learning_rate": 1.2832883420592272e-06,
      "loss": 0.74671912,
      "num_input_tokens_seen": 112426385,
      "step": 5224,
      "time_per_iteration": 2.6567587852478027
    },
    {
      "auxiliary_loss_clip": 0.01178672,
      "auxiliary_loss_mlp": 0.01031457,
      "balance_loss_clip": 1.05375433,
      "balance_loss_mlp": 1.0230999,
      "epoch": 0.6282691035892503,
      "flos": 36137194848000.0,
      "grad_norm": 14.455382170466708,
      "language_loss": 0.64175749,
      "learning_rate": 1.282561160132972e-06,
      "loss": 0.66385877,
      "num_input_tokens_seen": 112446905,
      "step": 5225,
      "time_per_iteration": 2.734158992767334
    },
    {
      "auxiliary_loss_clip": 0.0118019,
      "auxiliary_loss_mlp": 0.01026561,
      "balance_loss_clip": 1.05236554,
      "balance_loss_mlp": 1.01896095,
      "epoch": 0.6283893464798894,
      "flos": 26537231266560.0,
      "grad_norm": 1.6268675151367238,
      "language_loss": 0.80957556,
      "learning_rate": 1.2818340870416186e-06,
      "loss": 0.83164304,
      "num_input_tokens_seen": 112468040,
      "step": 5226,
      "time_per_iteration": 2.7039389610290527
    },
    {
      "auxiliary_loss_clip": 0.01179867,
      "auxiliary_loss_mlp": 0.01026428,
      "balance_loss_clip": 1.05212963,
      "balance_loss_mlp": 1.01783311,
      "epoch": 0.6285095893705285,
      "flos": 22237216369920.0,
      "grad_norm": 2.1171981409475826,
      "language_loss": 0.75732577,
      "learning_rate": 1.2811071228954626e-06,
      "loss": 0.77938873,
      "num_input_tokens_seen": 112486675,
      "step": 5227,
      "time_per_iteration": 2.6228296756744385
    },
    {
      "auxiliary_loss_clip": 0.0117653,
      "auxiliary_loss_mlp": 0.01022976,
      "balance_loss_clip": 1.05258894,
      "balance_loss_mlp": 1.01508701,
      "epoch": 0.6286298322611675,
      "flos": 26542259170560.0,
      "grad_norm": 2.292882858420029,
      "language_loss": 0.80960393,
      "learning_rate": 1.2803802678047846e-06,
      "loss": 0.831599,
      "num_input_tokens_seen": 112506825,
      "step": 5228,
      "time_per_iteration": 2.697012424468994
    },
    {
      "auxiliary_loss_clip": 0.01181151,
      "auxiliary_loss_mlp": 0.01035337,
      "balance_loss_clip": 1.05339921,
      "balance_loss_mlp": 1.02563369,
      "epoch": 0.6287500751518067,
      "flos": 21795227516160.0,
      "grad_norm": 1.7678000446415316,
      "language_loss": 0.74229378,
      "learning_rate": 1.279653521879848e-06,
      "loss": 0.76445866,
      "num_input_tokens_seen": 112526890,
      "step": 5229,
      "time_per_iteration": 2.6324164867401123
    },
    {
      "auxiliary_loss_clip": 0.01172892,
      "auxiliary_loss_mlp": 0.01024543,
      "balance_loss_clip": 1.04912698,
      "balance_loss_mlp": 1.01662755,
      "epoch": 0.6288703180424458,
      "flos": 20009605587840.0,
      "grad_norm": 2.2015842909638548,
      "language_loss": 0.83949471,
      "learning_rate": 1.2789268852308997e-06,
      "loss": 0.86146903,
      "num_input_tokens_seen": 112542100,
      "step": 5230,
      "time_per_iteration": 2.6276650428771973
    },
    {
      "auxiliary_loss_clip": 0.01179849,
      "auxiliary_loss_mlp": 0.0103061,
      "balance_loss_clip": 1.05286169,
      "balance_loss_mlp": 1.02224708,
      "epoch": 0.6289905609330848,
      "flos": 22124923476480.0,
      "grad_norm": 2.009034056794657,
      "language_loss": 0.71067166,
      "learning_rate": 1.2782003579681688e-06,
      "loss": 0.73277622,
      "num_input_tokens_seen": 112561630,
      "step": 5231,
      "time_per_iteration": 2.6425743103027344
    },
    {
      "auxiliary_loss_clip": 0.01181419,
      "auxiliary_loss_mlp": 0.01027143,
      "balance_loss_clip": 1.05384016,
      "balance_loss_mlp": 1.01900077,
      "epoch": 0.629110803823724,
      "flos": 25518481729920.0,
      "grad_norm": 1.6776064944702194,
      "language_loss": 0.74665785,
      "learning_rate": 1.2774739402018701e-06,
      "loss": 0.76874352,
      "num_input_tokens_seen": 112582465,
      "step": 5232,
      "time_per_iteration": 2.607326030731201
    },
    {
      "auxiliary_loss_clip": 0.01180226,
      "auxiliary_loss_mlp": 0.0102627,
      "balance_loss_clip": 1.05332351,
      "balance_loss_mlp": 1.01766002,
      "epoch": 0.629231046714363,
      "flos": 20886616056960.0,
      "grad_norm": 2.0060178966742477,
      "language_loss": 0.73336738,
      "learning_rate": 1.2767476320422002e-06,
      "loss": 0.75543237,
      "num_input_tokens_seen": 112602390,
      "step": 5233,
      "time_per_iteration": 2.5728089809417725
    },
    {
      "auxiliary_loss_clip": 0.01100491,
      "auxiliary_loss_mlp": 0.01011468,
      "balance_loss_clip": 1.03432572,
      "balance_loss_mlp": 1.00939405,
      "epoch": 0.6293512896050021,
      "flos": 65050027908480.0,
      "grad_norm": 0.6943126433403293,
      "language_loss": 0.5715555,
      "learning_rate": 1.2760214335993392e-06,
      "loss": 0.59267509,
      "num_input_tokens_seen": 112669035,
      "step": 5234,
      "time_per_iteration": 3.275822877883911
    },
    {
      "auxiliary_loss_clip": 0.01173217,
      "auxiliary_loss_mlp": 0.0102687,
      "balance_loss_clip": 1.04948497,
      "balance_loss_mlp": 1.01968718,
      "epoch": 0.6294715324956413,
      "flos": 34677857088000.0,
      "grad_norm": 3.859949890181047,
      "language_loss": 0.59270692,
      "learning_rate": 1.2752953449834514e-06,
      "loss": 0.61470783,
      "num_input_tokens_seen": 112691485,
      "step": 5235,
      "time_per_iteration": 2.73699688911438
    },
    {
      "auxiliary_loss_clip": 0.0117568,
      "auxiliary_loss_mlp": 0.01029024,
      "balance_loss_clip": 1.04996657,
      "balance_loss_mlp": 1.02168131,
      "epoch": 0.6295917753862803,
      "flos": 22784207656320.0,
      "grad_norm": 1.6227148729748473,
      "language_loss": 0.8046906,
      "learning_rate": 1.2745693663046836e-06,
      "loss": 0.82673764,
      "num_input_tokens_seen": 112710555,
      "step": 5236,
      "time_per_iteration": 3.560269355773926
    },
    {
      "auxiliary_loss_clip": 0.0117169,
      "auxiliary_loss_mlp": 0.01026629,
      "balance_loss_clip": 1.04894352,
      "balance_loss_mlp": 1.01923192,
      "epoch": 0.6297120182769194,
      "flos": 20850454039680.0,
      "grad_norm": 2.0529064179575185,
      "language_loss": 0.81057483,
      "learning_rate": 1.2738434976731662e-06,
      "loss": 0.83255804,
      "num_input_tokens_seen": 112728740,
      "step": 5237,
      "time_per_iteration": 2.619525671005249
    },
    {
      "auxiliary_loss_clip": 0.01182702,
      "auxiliary_loss_mlp": 0.01026413,
      "balance_loss_clip": 1.05712247,
      "balance_loss_mlp": 1.01816964,
      "epoch": 0.6298322611675584,
      "flos": 19497662997120.0,
      "grad_norm": 1.675366009866135,
      "language_loss": 0.75227469,
      "learning_rate": 1.2731177391990125e-06,
      "loss": 0.77436584,
      "num_input_tokens_seen": 112748665,
      "step": 5238,
      "time_per_iteration": 2.6456031799316406
    },
    {
      "auxiliary_loss_clip": 0.01174104,
      "auxiliary_loss_mlp": 0.01026477,
      "balance_loss_clip": 1.04935515,
      "balance_loss_mlp": 1.01838863,
      "epoch": 0.6299525040581976,
      "flos": 12604466649600.0,
      "grad_norm": 2.1093827175417785,
      "language_loss": 0.81961989,
      "learning_rate": 1.2723920909923203e-06,
      "loss": 0.84162569,
      "num_input_tokens_seen": 112764410,
      "step": 5239,
      "time_per_iteration": 2.6072537899017334
    },
    {
      "auxiliary_loss_clip": 0.01094515,
      "auxiliary_loss_mlp": 0.01003632,
      "balance_loss_clip": 1.02924585,
      "balance_loss_mlp": 1.00191545,
      "epoch": 0.6300727469488366,
      "flos": 57725685636480.0,
      "grad_norm": 0.8512407477628028,
      "language_loss": 0.60415578,
      "learning_rate": 1.2716665531631688e-06,
      "loss": 0.62513733,
      "num_input_tokens_seen": 112818695,
      "step": 5240,
      "time_per_iteration": 4.017929315567017
    },
    {
      "auxiliary_loss_clip": 0.01176135,
      "auxiliary_loss_mlp": 0.01021302,
      "balance_loss_clip": 1.04915476,
      "balance_loss_mlp": 1.01345229,
      "epoch": 0.6301929898394757,
      "flos": 22527302607360.0,
      "grad_norm": 3.6530064748169777,
      "language_loss": 0.77319062,
      "learning_rate": 1.270941125821623e-06,
      "loss": 0.79516506,
      "num_input_tokens_seen": 112839120,
      "step": 5241,
      "time_per_iteration": 3.547865867614746
    },
    {
      "auxiliary_loss_clip": 0.01173718,
      "auxiliary_loss_mlp": 0.01026183,
      "balance_loss_clip": 1.0495559,
      "balance_loss_mlp": 1.01860702,
      "epoch": 0.6303132327301149,
      "flos": 28293550675200.0,
      "grad_norm": 1.8393731138072429,
      "language_loss": 0.75016272,
      "learning_rate": 1.2702158090777278e-06,
      "loss": 0.77216172,
      "num_input_tokens_seen": 112860210,
      "step": 5242,
      "time_per_iteration": 2.7308237552642822
    },
    {
      "auxiliary_loss_clip": 0.01179402,
      "auxiliary_loss_mlp": 0.01032878,
      "balance_loss_clip": 1.05407977,
      "balance_loss_mlp": 1.02495623,
      "epoch": 0.6304334756207539,
      "flos": 25264521596160.0,
      "grad_norm": 2.4231655924616677,
      "language_loss": 0.74718261,
      "learning_rate": 1.2694906030415148e-06,
      "loss": 0.76930541,
      "num_input_tokens_seen": 112877955,
      "step": 5243,
      "time_per_iteration": 3.572737216949463
    },
    {
      "auxiliary_loss_clip": 0.01182352,
      "auxiliary_loss_mlp": 0.01030834,
      "balance_loss_clip": 1.05368841,
      "balance_loss_mlp": 1.02188158,
      "epoch": 0.630553718511393,
      "flos": 18033548728320.0,
      "grad_norm": 2.4596990191506145,
      "language_loss": 0.81902003,
      "learning_rate": 1.2687655078229958e-06,
      "loss": 0.84115195,
      "num_input_tokens_seen": 112892285,
      "step": 5244,
      "time_per_iteration": 2.565351963043213
    },
    {
      "auxiliary_loss_clip": 0.01177578,
      "auxiliary_loss_mlp": 0.01028895,
      "balance_loss_clip": 1.05103588,
      "balance_loss_mlp": 1.02104497,
      "epoch": 0.6306739614020321,
      "flos": 27304103658240.0,
      "grad_norm": 1.917589870694284,
      "language_loss": 0.68869919,
      "learning_rate": 1.2680405235321678e-06,
      "loss": 0.71076393,
      "num_input_tokens_seen": 112913620,
      "step": 5245,
      "time_per_iteration": 2.666154146194458
    },
    {
      "auxiliary_loss_clip": 0.01180397,
      "auxiliary_loss_mlp": 0.01030055,
      "balance_loss_clip": 1.05460215,
      "balance_loss_mlp": 1.02066469,
      "epoch": 0.6307942042926712,
      "flos": 15341434243200.0,
      "grad_norm": 3.752565671260404,
      "language_loss": 0.78758353,
      "learning_rate": 1.267315650279011e-06,
      "loss": 0.80968809,
      "num_input_tokens_seen": 112932090,
      "step": 5246,
      "time_per_iteration": 2.668955087661743
    },
    {
      "auxiliary_loss_clip": 0.01182239,
      "auxiliary_loss_mlp": 0.0102613,
      "balance_loss_clip": 1.0536387,
      "balance_loss_mlp": 1.01795864,
      "epoch": 0.6309144471833102,
      "flos": 19606400444160.0,
      "grad_norm": 2.1776007618614037,
      "language_loss": 0.74234283,
      "learning_rate": 1.2665908881734874e-06,
      "loss": 0.76442659,
      "num_input_tokens_seen": 112950925,
      "step": 5247,
      "time_per_iteration": 2.6717491149902344
    },
    {
      "auxiliary_loss_clip": 0.01176587,
      "auxiliary_loss_mlp": 0.01020772,
      "balance_loss_clip": 1.05222309,
      "balance_loss_mlp": 1.01254082,
      "epoch": 0.6310346900739494,
      "flos": 17493345112320.0,
      "grad_norm": 2.297677902819037,
      "language_loss": 0.85451609,
      "learning_rate": 1.2658662373255432e-06,
      "loss": 0.87648964,
      "num_input_tokens_seen": 112969315,
      "step": 5248,
      "time_per_iteration": 2.607215642929077
    },
    {
      "auxiliary_loss_clip": 0.01091379,
      "auxiliary_loss_mlp": 0.01002006,
      "balance_loss_clip": 1.02700949,
      "balance_loss_mlp": 1.00015819,
      "epoch": 0.6311549329645885,
      "flos": 55070164131840.0,
      "grad_norm": 0.7055391218785305,
      "language_loss": 0.52182567,
      "learning_rate": 1.2651416978451063e-06,
      "loss": 0.54275954,
      "num_input_tokens_seen": 113034700,
      "step": 5249,
      "time_per_iteration": 3.2874438762664795
    },
    {
      "auxiliary_loss_clip": 0.01182796,
      "auxiliary_loss_mlp": 0.01031942,
      "balance_loss_clip": 1.05469525,
      "balance_loss_mlp": 1.02304268,
      "epoch": 0.6312751758552275,
      "flos": 41902545075840.0,
      "grad_norm": 2.6471576068429568,
      "language_loss": 0.64888245,
      "learning_rate": 1.2644172698420903e-06,
      "loss": 0.67102987,
      "num_input_tokens_seen": 113056805,
      "step": 5250,
      "time_per_iteration": 2.775930404663086
    },
    {
      "auxiliary_loss_clip": 0.01176416,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.05115533,
      "balance_loss_mlp": 1.01970351,
      "epoch": 0.6313954187458667,
      "flos": 19646800266240.0,
      "grad_norm": 11.614650310626976,
      "language_loss": 0.84914488,
      "learning_rate": 1.2636929534263892e-06,
      "loss": 0.87118638,
      "num_input_tokens_seen": 113075790,
      "step": 5251,
      "time_per_iteration": 2.5973775386810303
    },
    {
      "auxiliary_loss_clip": 0.01175745,
      "auxiliary_loss_mlp": 0.01024488,
      "balance_loss_clip": 1.05140817,
      "balance_loss_mlp": 1.01620936,
      "epoch": 0.6315156616365057,
      "flos": 22894273906560.0,
      "grad_norm": 3.1072213766920047,
      "language_loss": 0.77225697,
      "learning_rate": 1.2629687487078821e-06,
      "loss": 0.79425931,
      "num_input_tokens_seen": 113094600,
      "step": 5252,
      "time_per_iteration": 2.679891347885132
    },
    {
      "auxiliary_loss_clip": 0.01176966,
      "auxiliary_loss_mlp": 0.0102772,
      "balance_loss_clip": 1.05113935,
      "balance_loss_mlp": 1.01930976,
      "epoch": 0.6316359045271448,
      "flos": 23726251699200.0,
      "grad_norm": 2.3152712148151955,
      "language_loss": 0.76603484,
      "learning_rate": 1.2622446557964293e-06,
      "loss": 0.78808171,
      "num_input_tokens_seen": 113112605,
      "step": 5253,
      "time_per_iteration": 2.663452386856079
    },
    {
      "auxiliary_loss_clip": 0.01172578,
      "auxiliary_loss_mlp": 0.01024416,
      "balance_loss_clip": 1.04887891,
      "balance_loss_mlp": 1.01690626,
      "epoch": 0.631756147417784,
      "flos": 33108417164160.0,
      "grad_norm": 1.816442949898359,
      "language_loss": 0.71662354,
      "learning_rate": 1.261520674801876e-06,
      "loss": 0.73859346,
      "num_input_tokens_seen": 113133200,
      "step": 5254,
      "time_per_iteration": 2.7623941898345947
    },
    {
      "auxiliary_loss_clip": 0.01176943,
      "auxiliary_loss_mlp": 0.01025512,
      "balance_loss_clip": 1.05072427,
      "balance_loss_mlp": 1.01673174,
      "epoch": 0.631876390308423,
      "flos": 31248424126080.0,
      "grad_norm": 2.2483225050606155,
      "language_loss": 0.72468501,
      "learning_rate": 1.2607968058340488e-06,
      "loss": 0.74670959,
      "num_input_tokens_seen": 113152895,
      "step": 5255,
      "time_per_iteration": 2.7051260471343994
    },
    {
      "auxiliary_loss_clip": 0.01173371,
      "auxiliary_loss_mlp": 0.01031148,
      "balance_loss_clip": 1.05056262,
      "balance_loss_mlp": 1.0229938,
      "epoch": 0.6319966331990621,
      "flos": 24681152810880.0,
      "grad_norm": 2.046564568330612,
      "language_loss": 0.7300548,
      "learning_rate": 1.2600730490027583e-06,
      "loss": 0.75209999,
      "num_input_tokens_seen": 113173135,
      "step": 5256,
      "time_per_iteration": 2.6967384815216064
    },
    {
      "auxiliary_loss_clip": 0.01175725,
      "auxiliary_loss_mlp": 0.01028137,
      "balance_loss_clip": 1.05210328,
      "balance_loss_mlp": 1.02011442,
      "epoch": 0.6321168760897012,
      "flos": 17491764913920.0,
      "grad_norm": 1.8722000705194926,
      "language_loss": 0.80526698,
      "learning_rate": 1.2593494044177984e-06,
      "loss": 0.82730567,
      "num_input_tokens_seen": 113191440,
      "step": 5257,
      "time_per_iteration": 2.601386547088623
    },
    {
      "auxiliary_loss_clip": 0.01178953,
      "auxiliary_loss_mlp": 0.0102792,
      "balance_loss_clip": 1.04909348,
      "balance_loss_mlp": 1.01937866,
      "epoch": 0.6322371189803403,
      "flos": 18295373940480.0,
      "grad_norm": 4.539138627533776,
      "language_loss": 0.81108141,
      "learning_rate": 1.2586258721889448e-06,
      "loss": 0.83315015,
      "num_input_tokens_seen": 113208790,
      "step": 5258,
      "time_per_iteration": 2.567509412765503
    },
    {
      "auxiliary_loss_clip": 0.01173907,
      "auxiliary_loss_mlp": 0.01026371,
      "balance_loss_clip": 1.05057669,
      "balance_loss_mlp": 1.0181458,
      "epoch": 0.6323573618709794,
      "flos": 20157270399360.0,
      "grad_norm": 5.725883634007467,
      "language_loss": 0.81365871,
      "learning_rate": 1.2579024524259573e-06,
      "loss": 0.83566147,
      "num_input_tokens_seen": 113225050,
      "step": 5259,
      "time_per_iteration": 2.6242103576660156
    },
    {
      "auxiliary_loss_clip": 0.01177099,
      "auxiliary_loss_mlp": 0.01029037,
      "balance_loss_clip": 1.05204821,
      "balance_loss_mlp": 1.02097845,
      "epoch": 0.6324776047616185,
      "flos": 20042391726720.0,
      "grad_norm": 2.4381069935099235,
      "language_loss": 0.91167557,
      "learning_rate": 1.2571791452385768e-06,
      "loss": 0.93373698,
      "num_input_tokens_seen": 113242315,
      "step": 5260,
      "time_per_iteration": 2.6239089965820312
    },
    {
      "auxiliary_loss_clip": 0.01173728,
      "auxiliary_loss_mlp": 0.01022141,
      "balance_loss_clip": 1.05221748,
      "balance_loss_mlp": 1.01483345,
      "epoch": 0.6325978476522576,
      "flos": 30848235724800.0,
      "grad_norm": 1.8253561109958885,
      "language_loss": 0.7729491,
      "learning_rate": 1.2564559507365301e-06,
      "loss": 0.79490781,
      "num_input_tokens_seen": 113264720,
      "step": 5261,
      "time_per_iteration": 2.674666166305542
    },
    {
      "auxiliary_loss_clip": 0.01179759,
      "auxiliary_loss_mlp": 0.01028065,
      "balance_loss_clip": 1.05337572,
      "balance_loss_mlp": 1.01927924,
      "epoch": 0.6327180905428966,
      "flos": 24535104111360.0,
      "grad_norm": 3.3598158034237944,
      "language_loss": 0.79316813,
      "learning_rate": 1.2557328690295244e-06,
      "loss": 0.81524634,
      "num_input_tokens_seen": 113282910,
      "step": 5262,
      "time_per_iteration": 3.5614113807678223
    },
    {
      "auxiliary_loss_clip": 0.01175974,
      "auxiliary_loss_mlp": 0.0102462,
      "balance_loss_clip": 1.05216646,
      "balance_loss_mlp": 1.01680565,
      "epoch": 0.6328383334335358,
      "flos": 21575274583680.0,
      "grad_norm": 1.7197964324992185,
      "language_loss": 0.76137352,
      "learning_rate": 1.255009900227251e-06,
      "loss": 0.7833795,
      "num_input_tokens_seen": 113301935,
      "step": 5263,
      "time_per_iteration": 2.6353230476379395
    },
    {
      "auxiliary_loss_clip": 0.01177473,
      "auxiliary_loss_mlp": 0.01026739,
      "balance_loss_clip": 1.05396318,
      "balance_loss_mlp": 1.01901746,
      "epoch": 0.6329585763241748,
      "flos": 22929861306240.0,
      "grad_norm": 1.9260793468023485,
      "language_loss": 0.79640627,
      "learning_rate": 1.254287044439383e-06,
      "loss": 0.81844836,
      "num_input_tokens_seen": 113321540,
      "step": 5264,
      "time_per_iteration": 2.6211414337158203
    },
    {
      "auxiliary_loss_clip": 0.01078228,
      "auxiliary_loss_mlp": 0.01002892,
      "balance_loss_clip": 1.01669955,
      "balance_loss_mlp": 1.00165272,
      "epoch": 0.6330788192148139,
      "flos": 70936897847040.0,
      "grad_norm": 0.7872648211098013,
      "language_loss": 0.54427469,
      "learning_rate": 1.2535643017755776e-06,
      "loss": 0.56508589,
      "num_input_tokens_seen": 113383730,
      "step": 5265,
      "time_per_iteration": 3.249950647354126
    },
    {
      "auxiliary_loss_clip": 0.01174012,
      "auxiliary_loss_mlp": 0.01022461,
      "balance_loss_clip": 1.04920483,
      "balance_loss_mlp": 1.01447392,
      "epoch": 0.6331990621054531,
      "flos": 21244501215360.0,
      "grad_norm": 5.2795828373392055,
      "language_loss": 0.72033519,
      "learning_rate": 1.2528416723454737e-06,
      "loss": 0.74229991,
      "num_input_tokens_seen": 113400400,
      "step": 5266,
      "time_per_iteration": 2.605574369430542
    },
    {
      "auxiliary_loss_clip": 0.01173506,
      "auxiliary_loss_mlp": 0.01025573,
      "balance_loss_clip": 1.05134249,
      "balance_loss_mlp": 1.01833999,
      "epoch": 0.6333193049960921,
      "flos": 34459412526720.0,
      "grad_norm": 1.4933178536161316,
      "language_loss": 0.71173352,
      "learning_rate": 1.2521191562586945e-06,
      "loss": 0.73372436,
      "num_input_tokens_seen": 113424050,
      "step": 5267,
      "time_per_iteration": 3.6642465591430664
    },
    {
      "auxiliary_loss_clip": 0.01175325,
      "auxiliary_loss_mlp": 0.01031688,
      "balance_loss_clip": 1.0506233,
      "balance_loss_mlp": 1.02391589,
      "epoch": 0.6334395478867312,
      "flos": 18329883932160.0,
      "grad_norm": 3.0740444889316754,
      "language_loss": 0.76539338,
      "learning_rate": 1.2513967536248445e-06,
      "loss": 0.78746349,
      "num_input_tokens_seen": 113440370,
      "step": 5268,
      "time_per_iteration": 3.4895126819610596
    },
    {
      "auxiliary_loss_clip": 0.01178276,
      "auxiliary_loss_mlp": 0.01025025,
      "balance_loss_clip": 1.05270708,
      "balance_loss_mlp": 1.01691866,
      "epoch": 0.6335597907773702,
      "flos": 23623152687360.0,
      "grad_norm": 1.6964169610971835,
      "language_loss": 0.81166941,
      "learning_rate": 1.2506744645535117e-06,
      "loss": 0.83370245,
      "num_input_tokens_seen": 113460800,
      "step": 5269,
      "time_per_iteration": 2.626953125
    },
    {
      "auxiliary_loss_clip": 0.01171972,
      "auxiliary_loss_mlp": 0.01022036,
      "balance_loss_clip": 1.04800153,
      "balance_loss_mlp": 1.01459765,
      "epoch": 0.6336800336680094,
      "flos": 22710913954560.0,
      "grad_norm": 13.050186200683779,
      "language_loss": 0.6071108,
      "learning_rate": 1.249952289154267e-06,
      "loss": 0.62905085,
      "num_input_tokens_seen": 113480840,
      "step": 5270,
      "time_per_iteration": 3.4847664833068848
    },
    {
      "auxiliary_loss_clip": 0.01174349,
      "auxiliary_loss_mlp": 0.01026028,
      "balance_loss_clip": 1.05282497,
      "balance_loss_mlp": 1.01862824,
      "epoch": 0.6338002765586485,
      "flos": 23622757637760.0,
      "grad_norm": 2.013009177942276,
      "language_loss": 0.76250887,
      "learning_rate": 1.2492302275366635e-06,
      "loss": 0.78451264,
      "num_input_tokens_seen": 113500515,
      "step": 5271,
      "time_per_iteration": 2.606198787689209
    },
    {
      "auxiliary_loss_clip": 0.011805,
      "auxiliary_loss_mlp": 0.01022728,
      "balance_loss_clip": 1.0529232,
      "balance_loss_mlp": 1.01470566,
      "epoch": 0.6339205194492875,
      "flos": 26505450708480.0,
      "grad_norm": 2.2839467250361287,
      "language_loss": 0.64954048,
      "learning_rate": 1.2485082798102377e-06,
      "loss": 0.67157274,
      "num_input_tokens_seen": 113520930,
      "step": 5272,
      "time_per_iteration": 2.676365613937378
    },
    {
      "auxiliary_loss_clip": 0.0117849,
      "auxiliary_loss_mlp": 0.01025886,
      "balance_loss_clip": 1.05171025,
      "balance_loss_mlp": 1.01791096,
      "epoch": 0.6340407623399267,
      "flos": 18544306170240.0,
      "grad_norm": 2.8287831644482053,
      "language_loss": 0.68409586,
      "learning_rate": 1.2477864460845084e-06,
      "loss": 0.70613962,
      "num_input_tokens_seen": 113537330,
      "step": 5273,
      "time_per_iteration": 2.597468852996826
    },
    {
      "auxiliary_loss_clip": 0.01184053,
      "auxiliary_loss_mlp": 0.01032507,
      "balance_loss_clip": 1.05574346,
      "balance_loss_mlp": 1.02390265,
      "epoch": 0.6341610052305657,
      "flos": 17712579772800.0,
      "grad_norm": 3.003565395986556,
      "language_loss": 0.73389071,
      "learning_rate": 1.2470647264689776e-06,
      "loss": 0.75605631,
      "num_input_tokens_seen": 113555810,
      "step": 5274,
      "time_per_iteration": 2.604065179824829
    },
    {
      "auxiliary_loss_clip": 0.01173998,
      "auxiliary_loss_mlp": 0.01025165,
      "balance_loss_clip": 1.0489049,
      "balance_loss_mlp": 1.01732731,
      "epoch": 0.6342812481212048,
      "flos": 23587026583680.0,
      "grad_norm": 2.038473014453964,
      "language_loss": 0.71399432,
      "learning_rate": 1.2463431210731282e-06,
      "loss": 0.73598588,
      "num_input_tokens_seen": 113575395,
      "step": 5275,
      "time_per_iteration": 2.691650867462158
    },
    {
      "auxiliary_loss_clip": 0.01177604,
      "auxiliary_loss_mlp": 0.01028375,
      "balance_loss_clip": 1.05167997,
      "balance_loss_mlp": 1.02016771,
      "epoch": 0.634401491011844,
      "flos": 17821927751040.0,
      "grad_norm": 2.4975130203994733,
      "language_loss": 0.76329911,
      "learning_rate": 1.2456216300064289e-06,
      "loss": 0.78535891,
      "num_input_tokens_seen": 113592945,
      "step": 5276,
      "time_per_iteration": 2.599181890487671
    },
    {
      "auxiliary_loss_clip": 0.01181826,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.05479705,
      "balance_loss_mlp": 1.01943445,
      "epoch": 0.634521733902483,
      "flos": 21358158825600.0,
      "grad_norm": 3.0139733465227425,
      "language_loss": 0.78593236,
      "learning_rate": 1.244900253378328e-06,
      "loss": 0.80802691,
      "num_input_tokens_seen": 113613000,
      "step": 5277,
      "time_per_iteration": 2.6671059131622314
    },
    {
      "auxiliary_loss_clip": 0.01175135,
      "auxiliary_loss_mlp": 0.01023508,
      "balance_loss_clip": 1.04953146,
      "balance_loss_mlp": 1.01600075,
      "epoch": 0.6346419767931221,
      "flos": 16545052103040.0,
      "grad_norm": 4.792364966132597,
      "language_loss": 0.6927036,
      "learning_rate": 1.2441789912982583e-06,
      "loss": 0.71469003,
      "num_input_tokens_seen": 113630085,
      "step": 5278,
      "time_per_iteration": 2.6757874488830566
    },
    {
      "auxiliary_loss_clip": 0.01177197,
      "auxiliary_loss_mlp": 0.01027424,
      "balance_loss_clip": 1.05137265,
      "balance_loss_mlp": 1.01900148,
      "epoch": 0.6347622196837612,
      "flos": 24350989973760.0,
      "grad_norm": 1.876814992411878,
      "language_loss": 0.64825737,
      "learning_rate": 1.2434578438756346e-06,
      "loss": 0.67030358,
      "num_input_tokens_seen": 113650515,
      "step": 5279,
      "time_per_iteration": 2.6966066360473633
    },
    {
      "auxiliary_loss_clip": 0.01173831,
      "auxiliary_loss_mlp": 0.01023333,
      "balance_loss_clip": 1.0493319,
      "balance_loss_mlp": 1.01535225,
      "epoch": 0.6348824625744003,
      "flos": 64523178195840.0,
      "grad_norm": 2.406000439158535,
      "language_loss": 0.78026831,
      "learning_rate": 1.242736811219855e-06,
      "loss": 0.80223989,
      "num_input_tokens_seen": 113676475,
      "step": 5280,
      "time_per_iteration": 2.972139358520508
    },
    {
      "auxiliary_loss_clip": 0.01173656,
      "auxiliary_loss_mlp": 0.01026144,
      "balance_loss_clip": 1.05061698,
      "balance_loss_mlp": 1.01860404,
      "epoch": 0.6350027054650393,
      "flos": 28622133313920.0,
      "grad_norm": 1.862888206353093,
      "language_loss": 0.8205241,
      "learning_rate": 1.2420158934402988e-06,
      "loss": 0.84252203,
      "num_input_tokens_seen": 113697090,
      "step": 5281,
      "time_per_iteration": 2.729504346847534
    },
    {
      "auxiliary_loss_clip": 0.01176383,
      "auxiliary_loss_mlp": 0.0102388,
      "balance_loss_clip": 1.0502255,
      "balance_loss_mlp": 1.01610136,
      "epoch": 0.6351229483556785,
      "flos": 23002544476800.0,
      "grad_norm": 2.2802350894288255,
      "language_loss": 0.8470552,
      "learning_rate": 1.2412950906463286e-06,
      "loss": 0.86905777,
      "num_input_tokens_seen": 113714395,
      "step": 5282,
      "time_per_iteration": 2.648725748062134
    },
    {
      "auxiliary_loss_clip": 0.01174276,
      "auxiliary_loss_mlp": 0.01027157,
      "balance_loss_clip": 1.050892,
      "balance_loss_mlp": 1.01961088,
      "epoch": 0.6352431912463176,
      "flos": 21939300967680.0,
      "grad_norm": 1.890054246283699,
      "language_loss": 0.89931619,
      "learning_rate": 1.2405744029472902e-06,
      "loss": 0.92133057,
      "num_input_tokens_seen": 113733880,
      "step": 5283,
      "time_per_iteration": 2.7168328762054443
    },
    {
      "auxiliary_loss_clip": 0.01173583,
      "auxiliary_loss_mlp": 0.01024438,
      "balance_loss_clip": 1.04983044,
      "balance_loss_mlp": 1.01661193,
      "epoch": 0.6353634341369566,
      "flos": 13735257684480.0,
      "grad_norm": 2.298652019903862,
      "language_loss": 0.76232314,
      "learning_rate": 1.2398538304525108e-06,
      "loss": 0.78430337,
      "num_input_tokens_seen": 113752505,
      "step": 5284,
      "time_per_iteration": 2.604417324066162
    },
    {
      "auxiliary_loss_clip": 0.01180466,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.05318904,
      "balance_loss_mlp": 1.01701045,
      "epoch": 0.6354836770275958,
      "flos": 19316170552320.0,
      "grad_norm": 2.1507836418518296,
      "language_loss": 0.75453657,
      "learning_rate": 1.2391333732713016e-06,
      "loss": 0.77659392,
      "num_input_tokens_seen": 113770310,
      "step": 5285,
      "time_per_iteration": 2.6465749740600586
    },
    {
      "auxiliary_loss_clip": 0.01180233,
      "auxiliary_loss_mlp": 0.01023163,
      "balance_loss_clip": 1.05241084,
      "balance_loss_mlp": 1.01475334,
      "epoch": 0.6356039199182348,
      "flos": 21613375935360.0,
      "grad_norm": 2.152505758825448,
      "language_loss": 0.78300947,
      "learning_rate": 1.2384130315129543e-06,
      "loss": 0.80504346,
      "num_input_tokens_seen": 113788635,
      "step": 5286,
      "time_per_iteration": 2.6560933589935303
    },
    {
      "auxiliary_loss_clip": 0.01173515,
      "auxiliary_loss_mlp": 0.01030155,
      "balance_loss_clip": 1.04842389,
      "balance_loss_mlp": 1.02210236,
      "epoch": 0.6357241628088739,
      "flos": 18111978074880.0,
      "grad_norm": 6.2835409228993875,
      "language_loss": 0.73500878,
      "learning_rate": 1.2376928052867447e-06,
      "loss": 0.75704551,
      "num_input_tokens_seen": 113807755,
      "step": 5287,
      "time_per_iteration": 2.694895029067993
    },
    {
      "auxiliary_loss_clip": 0.01179492,
      "auxiliary_loss_mlp": 0.01031714,
      "balance_loss_clip": 1.0534476,
      "balance_loss_mlp": 1.02384615,
      "epoch": 0.6358444056995131,
      "flos": 24935256599040.0,
      "grad_norm": 1.868464812950087,
      "language_loss": 0.77466649,
      "learning_rate": 1.2369726947019299e-06,
      "loss": 0.7967785,
      "num_input_tokens_seen": 113828230,
      "step": 5288,
      "time_per_iteration": 2.742314338684082
    },
    {
      "auxiliary_loss_clip": 0.0117103,
      "auxiliary_loss_mlp": 0.01023563,
      "balance_loss_clip": 1.04916954,
      "balance_loss_mlp": 1.01590347,
      "epoch": 0.6359646485901521,
      "flos": 23293348986240.0,
      "grad_norm": 2.2394232504881413,
      "language_loss": 0.6705336,
      "learning_rate": 1.2362526998677511e-06,
      "loss": 0.69247949,
      "num_input_tokens_seen": 113844595,
      "step": 5289,
      "time_per_iteration": 3.5539228916168213
    },
    {
      "auxiliary_loss_clip": 0.01173012,
      "auxiliary_loss_mlp": 0.01023469,
      "balance_loss_clip": 1.05098248,
      "balance_loss_mlp": 1.01630747,
      "epoch": 0.6360848914807912,
      "flos": 20887442069760.0,
      "grad_norm": 1.974780462735278,
      "language_loss": 0.84279501,
      "learning_rate": 1.2355328208934301e-06,
      "loss": 0.86475986,
      "num_input_tokens_seen": 113863470,
      "step": 5290,
      "time_per_iteration": 2.6315174102783203
    },
    {
      "auxiliary_loss_clip": 0.01173863,
      "auxiliary_loss_mlp": 0.01024147,
      "balance_loss_clip": 1.04982328,
      "balance_loss_mlp": 1.01663339,
      "epoch": 0.6362051343714303,
      "flos": 18479775386880.0,
      "grad_norm": 3.9483240267516617,
      "language_loss": 0.72337443,
      "learning_rate": 1.2348130578881728e-06,
      "loss": 0.74535453,
      "num_input_tokens_seen": 113881690,
      "step": 5291,
      "time_per_iteration": 2.5953941345214844
    },
    {
      "auxiliary_loss_clip": 0.01177214,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.05091667,
      "balance_loss_mlp": 1.01824164,
      "epoch": 0.6363253772620694,
      "flos": 24389594115840.0,
      "grad_norm": 2.3252335813124145,
      "language_loss": 0.76642454,
      "learning_rate": 1.2340934109611664e-06,
      "loss": 0.78846312,
      "num_input_tokens_seen": 113902450,
      "step": 5292,
      "time_per_iteration": 2.633091926574707
    },
    {
      "auxiliary_loss_clip": 0.01179329,
      "auxiliary_loss_mlp": 0.01030158,
      "balance_loss_clip": 1.05163598,
      "balance_loss_mlp": 1.02144384,
      "epoch": 0.6364456201527084,
      "flos": 25958243940480.0,
      "grad_norm": 2.4529033439352275,
      "language_loss": 0.68337882,
      "learning_rate": 1.2333738802215798e-06,
      "loss": 0.70547366,
      "num_input_tokens_seen": 113922670,
      "step": 5293,
      "time_per_iteration": 2.6786718368530273
    },
    {
      "auxiliary_loss_clip": 0.0117434,
      "auxiliary_loss_mlp": 0.01021609,
      "balance_loss_clip": 1.05038154,
      "balance_loss_mlp": 1.01409888,
      "epoch": 0.6365658630433476,
      "flos": 20740711011840.0,
      "grad_norm": 2.269899526884932,
      "language_loss": 0.81209326,
      "learning_rate": 1.2326544657785668e-06,
      "loss": 0.83405274,
      "num_input_tokens_seen": 113942360,
      "step": 5294,
      "time_per_iteration": 3.5761306285858154
    },
    {
      "auxiliary_loss_clip": 0.01176952,
      "auxiliary_loss_mlp": 0.01025341,
      "balance_loss_clip": 1.05327594,
      "balance_loss_mlp": 1.0169307,
      "epoch": 0.6366861059339867,
      "flos": 21434146047360.0,
      "grad_norm": 3.665823192670789,
      "language_loss": 0.74822909,
      "learning_rate": 1.2319351677412608e-06,
      "loss": 0.77025199,
      "num_input_tokens_seen": 113959405,
      "step": 5295,
      "time_per_iteration": 2.668537139892578
    },
    {
      "auxiliary_loss_clip": 0.01178102,
      "auxiliary_loss_mlp": 0.01025131,
      "balance_loss_clip": 1.052665,
      "balance_loss_mlp": 1.01710784,
      "epoch": 0.6368063488246257,
      "flos": 22267093507200.0,
      "grad_norm": 2.507621608047459,
      "language_loss": 0.74385357,
      "learning_rate": 1.2312159862187796e-06,
      "loss": 0.76588583,
      "num_input_tokens_seen": 113977815,
      "step": 5296,
      "time_per_iteration": 3.481236219406128
    },
    {
      "auxiliary_loss_clip": 0.01182327,
      "auxiliary_loss_mlp": 0.01032246,
      "balance_loss_clip": 1.05410504,
      "balance_loss_mlp": 1.02372909,
      "epoch": 0.6369265917152649,
      "flos": 22420719976320.0,
      "grad_norm": 1.4885819310879345,
      "language_loss": 0.76507097,
      "learning_rate": 1.2304969213202217e-06,
      "loss": 0.78721666,
      "num_input_tokens_seen": 113999075,
      "step": 5297,
      "time_per_iteration": 2.651262044906616
    },
    {
      "auxiliary_loss_clip": 0.01173978,
      "auxiliary_loss_mlp": 0.01023434,
      "balance_loss_clip": 1.0494504,
      "balance_loss_mlp": 1.01540494,
      "epoch": 0.6370468346059039,
      "flos": 24718176754560.0,
      "grad_norm": 2.484367285950889,
      "language_loss": 0.79196596,
      "learning_rate": 1.2297779731546692e-06,
      "loss": 0.81394005,
      "num_input_tokens_seen": 114018170,
      "step": 5298,
      "time_per_iteration": 2.646965503692627
    },
    {
      "auxiliary_loss_clip": 0.01175546,
      "auxiliary_loss_mlp": 0.01028073,
      "balance_loss_clip": 1.05140185,
      "balance_loss_mlp": 1.02020502,
      "epoch": 0.637167077496543,
      "flos": 25296589463040.0,
      "grad_norm": 1.8277537446570686,
      "language_loss": 0.77928793,
      "learning_rate": 1.2290591418311853e-06,
      "loss": 0.80132413,
      "num_input_tokens_seen": 114035565,
      "step": 5299,
      "time_per_iteration": 2.663485050201416
    },
    {
      "auxiliary_loss_clip": 0.01177046,
      "auxiliary_loss_mlp": 0.01026769,
      "balance_loss_clip": 1.05059063,
      "balance_loss_mlp": 1.01887727,
      "epoch": 0.637287320387182,
      "flos": 27671110871040.0,
      "grad_norm": 1.8571475348888455,
      "language_loss": 0.72199476,
      "learning_rate": 1.2283404274588172e-06,
      "loss": 0.74403286,
      "num_input_tokens_seen": 114054510,
      "step": 5300,
      "time_per_iteration": 2.6694326400756836
    },
    {
      "auxiliary_loss_clip": 0.01082187,
      "auxiliary_loss_mlp": 0.0100123,
      "balance_loss_clip": 1.01947463,
      "balance_loss_mlp": 0.99985945,
      "epoch": 0.6374075632778212,
      "flos": 63173406873600.0,
      "grad_norm": 0.7506111408298307,
      "language_loss": 0.52789998,
      "learning_rate": 1.227621830146592e-06,
      "loss": 0.54873419,
      "num_input_tokens_seen": 114109875,
      "step": 5301,
      "time_per_iteration": 3.071683406829834
    },
    {
      "auxiliary_loss_clip": 0.0117777,
      "auxiliary_loss_mlp": 0.01028546,
      "balance_loss_clip": 1.05187333,
      "balance_loss_mlp": 1.02043378,
      "epoch": 0.6375278061684603,
      "flos": 25558127366400.0,
      "grad_norm": 4.564574579584032,
      "language_loss": 0.79261565,
      "learning_rate": 1.2269033500035217e-06,
      "loss": 0.81467879,
      "num_input_tokens_seen": 114130010,
      "step": 5302,
      "time_per_iteration": 2.64204478263855
    },
    {
      "auxiliary_loss_clip": 0.01175136,
      "auxiliary_loss_mlp": 0.01026193,
      "balance_loss_clip": 1.05060446,
      "balance_loss_mlp": 1.0179671,
      "epoch": 0.6376480490590993,
      "flos": 25666362023040.0,
      "grad_norm": 2.1724397636344577,
      "language_loss": 0.73824424,
      "learning_rate": 1.2261849871385988e-06,
      "loss": 0.76025748,
      "num_input_tokens_seen": 114151115,
      "step": 5303,
      "time_per_iteration": 2.624051570892334
    },
    {
      "auxiliary_loss_clip": 0.01175477,
      "auxiliary_loss_mlp": 0.01026053,
      "balance_loss_clip": 1.05003452,
      "balance_loss_mlp": 1.01771986,
      "epoch": 0.6377682919497385,
      "flos": 31537684350720.0,
      "grad_norm": 2.6752671880238177,
      "language_loss": 0.62534058,
      "learning_rate": 1.2254667416607972e-06,
      "loss": 0.64735591,
      "num_input_tokens_seen": 114172715,
      "step": 5304,
      "time_per_iteration": 2.708839178085327
    },
    {
      "auxiliary_loss_clip": 0.01176214,
      "auxiliary_loss_mlp": 0.01024726,
      "balance_loss_clip": 1.05047226,
      "balance_loss_mlp": 1.01641774,
      "epoch": 0.6378885348403776,
      "flos": 23039209284480.0,
      "grad_norm": 2.5223476020417146,
      "language_loss": 0.83160305,
      "learning_rate": 1.2247486136790756e-06,
      "loss": 0.85361242,
      "num_input_tokens_seen": 114192195,
      "step": 5305,
      "time_per_iteration": 2.6099026203155518
    },
    {
      "auxiliary_loss_clip": 0.0117714,
      "auxiliary_loss_mlp": 0.01025599,
      "balance_loss_clip": 1.05231452,
      "balance_loss_mlp": 1.01805067,
      "epoch": 0.6380087777310166,
      "flos": 18697070712960.0,
      "grad_norm": 2.1750133739127935,
      "language_loss": 0.81095809,
      "learning_rate": 1.2240306033023726e-06,
      "loss": 0.83298552,
      "num_input_tokens_seen": 114210020,
      "step": 5306,
      "time_per_iteration": 2.6305720806121826
    },
    {
      "auxiliary_loss_clip": 0.01176014,
      "auxiliary_loss_mlp": 0.01026319,
      "balance_loss_clip": 1.05077434,
      "balance_loss_mlp": 1.01848125,
      "epoch": 0.6381290206216558,
      "flos": 23331558078720.0,
      "grad_norm": 1.8902435694734492,
      "language_loss": 0.72222555,
      "learning_rate": 1.223312710639611e-06,
      "loss": 0.74424887,
      "num_input_tokens_seen": 114228740,
      "step": 5307,
      "time_per_iteration": 2.618774890899658
    },
    {
      "auxiliary_loss_clip": 0.01179137,
      "auxiliary_loss_mlp": 0.01027136,
      "balance_loss_clip": 1.05347466,
      "balance_loss_mlp": 1.0185287,
      "epoch": 0.6382492635122948,
      "flos": 18880466578560.0,
      "grad_norm": 22.790805367187165,
      "language_loss": 0.86893934,
      "learning_rate": 1.2225949357996928e-06,
      "loss": 0.89100206,
      "num_input_tokens_seen": 114246865,
      "step": 5308,
      "time_per_iteration": 2.612675905227661
    },
    {
      "auxiliary_loss_clip": 0.01173579,
      "auxiliary_loss_mlp": 0.01026338,
      "balance_loss_clip": 1.05082464,
      "balance_loss_mlp": 1.01884842,
      "epoch": 0.6383695064029339,
      "flos": 27819134818560.0,
      "grad_norm": 1.6210686038668363,
      "language_loss": 0.80298376,
      "learning_rate": 1.221877278891505e-06,
      "loss": 0.82498288,
      "num_input_tokens_seen": 114266120,
      "step": 5309,
      "time_per_iteration": 2.6602895259857178
    },
    {
      "auxiliary_loss_clip": 0.01177155,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.0514853,
      "balance_loss_mlp": 1.01896977,
      "epoch": 0.638489749293573,
      "flos": 26395635853440.0,
      "grad_norm": 3.6133076228940157,
      "language_loss": 0.72148621,
      "learning_rate": 1.221159740023915e-06,
      "loss": 0.74353802,
      "num_input_tokens_seen": 114285950,
      "step": 5310,
      "time_per_iteration": 2.7029473781585693
    },
    {
      "auxiliary_loss_clip": 0.01180662,
      "auxiliary_loss_mlp": 0.0102733,
      "balance_loss_clip": 1.05411291,
      "balance_loss_mlp": 1.0186758,
      "epoch": 0.6386099921842121,
      "flos": 23988328306560.0,
      "grad_norm": 2.6646858969372493,
      "language_loss": 0.7281605,
      "learning_rate": 1.2204423193057735e-06,
      "loss": 0.75024045,
      "num_input_tokens_seen": 114304780,
      "step": 5311,
      "time_per_iteration": 2.6514508724212646
    },
    {
      "auxiliary_loss_clip": 0.01079881,
      "auxiliary_loss_mlp": 0.01001674,
      "balance_loss_clip": 1.01781178,
      "balance_loss_mlp": 1.0003922,
      "epoch": 0.6387302350748512,
      "flos": 71731169337600.0,
      "grad_norm": 0.8478353346476726,
      "language_loss": 0.63337207,
      "learning_rate": 1.2197250168459122e-06,
      "loss": 0.65418756,
      "num_input_tokens_seen": 114361180,
      "step": 5312,
      "time_per_iteration": 3.2264761924743652
    },
    {
      "auxiliary_loss_clip": 0.01172966,
      "auxiliary_loss_mlp": 0.01026986,
      "balance_loss_clip": 1.04993725,
      "balance_loss_mlp": 1.01942492,
      "epoch": 0.6388504779654903,
      "flos": 14535778141440.0,
      "grad_norm": 1.9477329164303807,
      "language_loss": 0.74433231,
      "learning_rate": 1.2190078327531454e-06,
      "loss": 0.76633179,
      "num_input_tokens_seen": 114377425,
      "step": 5313,
      "time_per_iteration": 2.580641031265259
    },
    {
      "auxiliary_loss_clip": 0.01172109,
      "auxiliary_loss_mlp": 0.01026862,
      "balance_loss_clip": 1.0496428,
      "balance_loss_mlp": 1.01866698,
      "epoch": 0.6389707208561294,
      "flos": 22346133384960.0,
      "grad_norm": 1.7456676839935676,
      "language_loss": 0.72458971,
      "learning_rate": 1.2182907671362697e-06,
      "loss": 0.74657941,
      "num_input_tokens_seen": 114398120,
      "step": 5314,
      "time_per_iteration": 2.634221076965332
    },
    {
      "auxiliary_loss_clip": 0.01176637,
      "auxiliary_loss_mlp": 0.01019645,
      "balance_loss_clip": 1.05195415,
      "balance_loss_mlp": 1.01223016,
      "epoch": 0.6390909637467684,
      "flos": 19426883247360.0,
      "grad_norm": 4.877661909660184,
      "language_loss": 0.78945005,
      "learning_rate": 1.2175738201040626e-06,
      "loss": 0.81141287,
      "num_input_tokens_seen": 114415160,
      "step": 5315,
      "time_per_iteration": 3.628424644470215
    },
    {
      "auxiliary_loss_clip": 0.01173584,
      "auxiliary_loss_mlp": 0.01025258,
      "balance_loss_clip": 1.04910183,
      "balance_loss_mlp": 1.01745844,
      "epoch": 0.6392112066374076,
      "flos": 24090852700800.0,
      "grad_norm": 1.9506940322748882,
      "language_loss": 0.78873318,
      "learning_rate": 1.2168569917652855e-06,
      "loss": 0.81072164,
      "num_input_tokens_seen": 114435015,
      "step": 5316,
      "time_per_iteration": 2.6360442638397217
    },
    {
      "auxiliary_loss_clip": 0.01175256,
      "auxiliary_loss_mlp": 0.01026224,
      "balance_loss_clip": 1.05096006,
      "balance_loss_mlp": 1.01864207,
      "epoch": 0.6393314495280467,
      "flos": 26795141896320.0,
      "grad_norm": 1.8443307876148565,
      "language_loss": 0.6420784,
      "learning_rate": 1.2161402822286797e-06,
      "loss": 0.6640932,
      "num_input_tokens_seen": 114455700,
      "step": 5317,
      "time_per_iteration": 2.6609067916870117
    },
    {
      "auxiliary_loss_clip": 0.0117706,
      "auxiliary_loss_mlp": 0.01023963,
      "balance_loss_clip": 1.05131578,
      "balance_loss_mlp": 1.01573169,
      "epoch": 0.6394516924186857,
      "flos": 20260692633600.0,
      "grad_norm": 2.3265641654343843,
      "language_loss": 0.787871,
      "learning_rate": 1.2154236916029703e-06,
      "loss": 0.80988121,
      "num_input_tokens_seen": 114473675,
      "step": 5318,
      "time_per_iteration": 2.629699468612671
    },
    {
      "auxiliary_loss_clip": 0.01174969,
      "auxiliary_loss_mlp": 0.01024533,
      "balance_loss_clip": 1.0510447,
      "balance_loss_mlp": 1.01664186,
      "epoch": 0.6395719353093249,
      "flos": 18368847210240.0,
      "grad_norm": 2.649716249106914,
      "language_loss": 0.73750961,
      "learning_rate": 1.2147072199968627e-06,
      "loss": 0.75950468,
      "num_input_tokens_seen": 114492310,
      "step": 5319,
      "time_per_iteration": 2.578798770904541
    },
    {
      "auxiliary_loss_clip": 0.01172896,
      "auxiliary_loss_mlp": 0.01024089,
      "balance_loss_clip": 1.04980183,
      "balance_loss_mlp": 1.01668024,
      "epoch": 0.6396921781999639,
      "flos": 17566315591680.0,
      "grad_norm": 1.9587548984282648,
      "language_loss": 0.71614581,
      "learning_rate": 1.2139908675190454e-06,
      "loss": 0.73811567,
      "num_input_tokens_seen": 114511520,
      "step": 5320,
      "time_per_iteration": 3.4679880142211914
    },
    {
      "auxiliary_loss_clip": 0.01173995,
      "auxiliary_loss_mlp": 0.01024115,
      "balance_loss_clip": 1.05056894,
      "balance_loss_mlp": 1.01632178,
      "epoch": 0.639812421090603,
      "flos": 21251252972160.0,
      "grad_norm": 2.184555773013878,
      "language_loss": 0.75224376,
      "learning_rate": 1.2132746342781883e-06,
      "loss": 0.77422488,
      "num_input_tokens_seen": 114532680,
      "step": 5321,
      "time_per_iteration": 3.5552635192871094
    },
    {
      "auxiliary_loss_clip": 0.01176822,
      "auxiliary_loss_mlp": 0.01023258,
      "balance_loss_clip": 1.05143845,
      "balance_loss_mlp": 1.01528275,
      "epoch": 0.6399326639812422,
      "flos": 11180967684480.0,
      "grad_norm": 3.6530900355650004,
      "language_loss": 0.80254757,
      "learning_rate": 1.2125585203829442e-06,
      "loss": 0.82454836,
      "num_input_tokens_seen": 114548320,
      "step": 5322,
      "time_per_iteration": 3.5307564735412598
    },
    {
      "auxiliary_loss_clip": 0.01175894,
      "auxiliary_loss_mlp": 0.01028962,
      "balance_loss_clip": 1.05157495,
      "balance_loss_mlp": 1.0210228,
      "epoch": 0.6400529068718812,
      "flos": 23911048195200.0,
      "grad_norm": 1.8431398442584424,
      "language_loss": 0.74599183,
      "learning_rate": 1.211842525941946e-06,
      "loss": 0.76804036,
      "num_input_tokens_seen": 114568115,
      "step": 5323,
      "time_per_iteration": 2.677520275115967
    },
    {
      "auxiliary_loss_clip": 0.01172985,
      "auxiliary_loss_mlp": 0.01020306,
      "balance_loss_clip": 1.05003166,
      "balance_loss_mlp": 1.01258123,
      "epoch": 0.6401731497625203,
      "flos": 44018724890880.0,
      "grad_norm": 1.6836421087377387,
      "language_loss": 0.78703523,
      "learning_rate": 1.2111266510638105e-06,
      "loss": 0.80896807,
      "num_input_tokens_seen": 114591040,
      "step": 5324,
      "time_per_iteration": 2.798940658569336
    },
    {
      "auxiliary_loss_clip": 0.01178561,
      "auxiliary_loss_mlp": 0.01026776,
      "balance_loss_clip": 1.05428338,
      "balance_loss_mlp": 1.01832414,
      "epoch": 0.6402933926531594,
      "flos": 20662209838080.0,
      "grad_norm": 2.890857934636382,
      "language_loss": 0.80147493,
      "learning_rate": 1.2104108958571346e-06,
      "loss": 0.82352829,
      "num_input_tokens_seen": 114609310,
      "step": 5325,
      "time_per_iteration": 2.6252002716064453
    },
    {
      "auxiliary_loss_clip": 0.01172991,
      "auxiliary_loss_mlp": 0.01024627,
      "balance_loss_clip": 1.05141234,
      "balance_loss_mlp": 1.01718211,
      "epoch": 0.6404136355437985,
      "flos": 24863327614080.0,
      "grad_norm": 1.4901651491270036,
      "language_loss": 0.75986201,
      "learning_rate": 1.2096952604304975e-06,
      "loss": 0.78183818,
      "num_input_tokens_seen": 114629740,
      "step": 5326,
      "time_per_iteration": 2.640078067779541
    },
    {
      "auxiliary_loss_clip": 0.01176594,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04936862,
      "balance_loss_mlp": 1.01554453,
      "epoch": 0.6405338784344375,
      "flos": 40479548901120.0,
      "grad_norm": 2.53618854034909,
      "language_loss": 0.7026974,
      "learning_rate": 1.2089797448924616e-06,
      "loss": 0.72470802,
      "num_input_tokens_seen": 114653615,
      "step": 5327,
      "time_per_iteration": 2.777557611465454
    },
    {
      "auxiliary_loss_clip": 0.01177945,
      "auxiliary_loss_mlp": 0.0103058,
      "balance_loss_clip": 1.05261159,
      "balance_loss_mlp": 1.02277458,
      "epoch": 0.6406541213250767,
      "flos": 20886041439360.0,
      "grad_norm": 2.071036427622821,
      "language_loss": 0.65935552,
      "learning_rate": 1.2082643493515692e-06,
      "loss": 0.68144077,
      "num_input_tokens_seen": 114671935,
      "step": 5328,
      "time_per_iteration": 2.6043450832366943
    },
    {
      "auxiliary_loss_clip": 0.01174655,
      "auxiliary_loss_mlp": 0.01027844,
      "balance_loss_clip": 1.05070424,
      "balance_loss_mlp": 1.02051616,
      "epoch": 0.6407743642157158,
      "flos": 23295970679040.0,
      "grad_norm": 4.59843145695927,
      "language_loss": 0.81752497,
      "learning_rate": 1.207549073916346e-06,
      "loss": 0.8395499,
      "num_input_tokens_seen": 114692870,
      "step": 5329,
      "time_per_iteration": 2.64225172996521
    },
    {
      "auxiliary_loss_clip": 0.0117517,
      "auxiliary_loss_mlp": 0.01024011,
      "balance_loss_clip": 1.05093491,
      "balance_loss_mlp": 1.01669133,
      "epoch": 0.6408946071063548,
      "flos": 15012636122880.0,
      "grad_norm": 2.2011571556022345,
      "language_loss": 0.77663177,
      "learning_rate": 1.2068339186952976e-06,
      "loss": 0.79862356,
      "num_input_tokens_seen": 114710410,
      "step": 5330,
      "time_per_iteration": 2.6363344192504883
    },
    {
      "auxiliary_loss_clip": 0.01176776,
      "auxiliary_loss_mlp": 0.01028457,
      "balance_loss_clip": 1.05151236,
      "balance_loss_mlp": 1.02042842,
      "epoch": 0.6410148499969939,
      "flos": 22528595496960.0,
      "grad_norm": 2.69857633044905,
      "language_loss": 0.73780507,
      "learning_rate": 1.2061188837969136e-06,
      "loss": 0.75985742,
      "num_input_tokens_seen": 114730020,
      "step": 5331,
      "time_per_iteration": 2.6236908435821533
    },
    {
      "auxiliary_loss_clip": 0.0117683,
      "auxiliary_loss_mlp": 0.0103159,
      "balance_loss_clip": 1.05005479,
      "balance_loss_mlp": 1.02316761,
      "epoch": 0.641135092887633,
      "flos": 12422004537600.0,
      "grad_norm": 2.6517357223717313,
      "language_loss": 0.84670305,
      "learning_rate": 1.2054039693296631e-06,
      "loss": 0.86878723,
      "num_input_tokens_seen": 114748015,
      "step": 5332,
      "time_per_iteration": 2.6280922889709473
    },
    {
      "auxiliary_loss_clip": 0.01174041,
      "auxiliary_loss_mlp": 0.01022675,
      "balance_loss_clip": 1.05013859,
      "balance_loss_mlp": 1.01539421,
      "epoch": 0.6412553357782721,
      "flos": 22127329687680.0,
      "grad_norm": 1.843556788677032,
      "language_loss": 0.81274205,
      "learning_rate": 1.2046891754019992e-06,
      "loss": 0.83470917,
      "num_input_tokens_seen": 114768625,
      "step": 5333,
      "time_per_iteration": 2.6701927185058594
    },
    {
      "auxiliary_loss_clip": 0.01175759,
      "auxiliary_loss_mlp": 0.0102591,
      "balance_loss_clip": 1.05203569,
      "balance_loss_mlp": 1.01853108,
      "epoch": 0.6413755786689112,
      "flos": 15888605097600.0,
      "grad_norm": 1.8565902632681628,
      "language_loss": 0.82764006,
      "learning_rate": 1.2039745021223548e-06,
      "loss": 0.8496567,
      "num_input_tokens_seen": 114786045,
      "step": 5334,
      "time_per_iteration": 2.564795732498169
    },
    {
      "auxiliary_loss_clip": 0.01079589,
      "auxiliary_loss_mlp": 0.01001757,
      "balance_loss_clip": 1.01770794,
      "balance_loss_mlp": 1.00046396,
      "epoch": 0.6414958215595503,
      "flos": 68039159955840.0,
      "grad_norm": 0.7891640954038279,
      "language_loss": 0.57039034,
      "learning_rate": 1.2032599495991456e-06,
      "loss": 0.59120381,
      "num_input_tokens_seen": 114850785,
      "step": 5335,
      "time_per_iteration": 3.292754650115967
    },
    {
      "auxiliary_loss_clip": 0.01179252,
      "auxiliary_loss_mlp": 0.01028415,
      "balance_loss_clip": 1.05318356,
      "balance_loss_mlp": 1.01991558,
      "epoch": 0.6416160644501894,
      "flos": 44091300320640.0,
      "grad_norm": 1.6363426064528308,
      "language_loss": 0.69829202,
      "learning_rate": 1.2025455179407685e-06,
      "loss": 0.72036874,
      "num_input_tokens_seen": 114871945,
      "step": 5336,
      "time_per_iteration": 2.853607416152954
    },
    {
      "auxiliary_loss_clip": 0.01176611,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.05105209,
      "balance_loss_mlp": 1.0161283,
      "epoch": 0.6417363073408284,
      "flos": 20959837931520.0,
      "grad_norm": 1.8873071837851467,
      "language_loss": 0.73613095,
      "learning_rate": 1.2018312072556022e-06,
      "loss": 0.75814247,
      "num_input_tokens_seen": 114890445,
      "step": 5337,
      "time_per_iteration": 2.641524314880371
    },
    {
      "auxiliary_loss_clip": 0.01173957,
      "auxiliary_loss_mlp": 0.01019786,
      "balance_loss_clip": 1.04993296,
      "balance_loss_mlp": 1.0121212,
      "epoch": 0.6418565502314676,
      "flos": 22455122227200.0,
      "grad_norm": 2.307459849970182,
      "language_loss": 0.74287283,
      "learning_rate": 1.2011170176520077e-06,
      "loss": 0.76481026,
      "num_input_tokens_seen": 114911360,
      "step": 5338,
      "time_per_iteration": 2.657931089401245
    },
    {
      "auxiliary_loss_clip": 0.01172084,
      "auxiliary_loss_mlp": 0.01024043,
      "balance_loss_clip": 1.04936552,
      "balance_loss_mlp": 1.0166223,
      "epoch": 0.6419767931221066,
      "flos": 25045502417280.0,
      "grad_norm": 1.5424208581751073,
      "language_loss": 0.81272829,
      "learning_rate": 1.2004029492383256e-06,
      "loss": 0.83468956,
      "num_input_tokens_seen": 114932700,
      "step": 5339,
      "time_per_iteration": 2.672362804412842
    },
    {
      "auxiliary_loss_clip": 0.01175645,
      "auxiliary_loss_mlp": 0.01027789,
      "balance_loss_clip": 1.05208552,
      "balance_loss_mlp": 1.02012944,
      "epoch": 0.6420970360127457,
      "flos": 19463691709440.0,
      "grad_norm": 2.335373903267891,
      "language_loss": 0.73439038,
      "learning_rate": 1.1996890021228814e-06,
      "loss": 0.75642473,
      "num_input_tokens_seen": 114949475,
      "step": 5340,
      "time_per_iteration": 2.6418306827545166
    },
    {
      "auxiliary_loss_clip": 0.01173221,
      "auxiliary_loss_mlp": 0.01027141,
      "balance_loss_clip": 1.05065978,
      "balance_loss_mlp": 1.01988101,
      "epoch": 0.6422172789033849,
      "flos": 40406147458560.0,
      "grad_norm": 2.40555326920079,
      "language_loss": 0.70251518,
      "learning_rate": 1.1989751764139785e-06,
      "loss": 0.72451878,
      "num_input_tokens_seen": 114973125,
      "step": 5341,
      "time_per_iteration": 2.768808364868164
    },
    {
      "auxiliary_loss_clip": 0.01177016,
      "auxiliary_loss_mlp": 0.01025384,
      "balance_loss_clip": 1.05094898,
      "balance_loss_mlp": 1.0168817,
      "epoch": 0.6423375217940239,
      "flos": 27672870637440.0,
      "grad_norm": 1.541060181117498,
      "language_loss": 0.83285618,
      "learning_rate": 1.1982614722199044e-06,
      "loss": 0.85488015,
      "num_input_tokens_seen": 114994300,
      "step": 5342,
      "time_per_iteration": 3.659270763397217
    },
    {
      "auxiliary_loss_clip": 0.01172268,
      "auxiliary_loss_mlp": 0.01026004,
      "balance_loss_clip": 1.04904258,
      "balance_loss_mlp": 1.01823127,
      "epoch": 0.642457764684663,
      "flos": 18369242259840.0,
      "grad_norm": 2.306173680489509,
      "language_loss": 0.77838737,
      "learning_rate": 1.1975478896489276e-06,
      "loss": 0.8003701,
      "num_input_tokens_seen": 115012135,
      "step": 5343,
      "time_per_iteration": 2.669156074523926
    },
    {
      "auxiliary_loss_clip": 0.01172543,
      "auxiliary_loss_mlp": 0.01026266,
      "balance_loss_clip": 1.04905128,
      "balance_loss_mlp": 1.01889598,
      "epoch": 0.6425780075753021,
      "flos": 19750509809280.0,
      "grad_norm": 2.1814172429674796,
      "language_loss": 0.76617658,
      "learning_rate": 1.1968344288092981e-06,
      "loss": 0.78816473,
      "num_input_tokens_seen": 115028715,
      "step": 5344,
      "time_per_iteration": 2.6077663898468018
    },
    {
      "auxiliary_loss_clip": 0.01175449,
      "auxiliary_loss_mlp": 0.01027095,
      "balance_loss_clip": 1.05190945,
      "balance_loss_mlp": 1.01901221,
      "epoch": 0.6426982504659412,
      "flos": 20558536208640.0,
      "grad_norm": 2.4856735170547304,
      "language_loss": 0.64592189,
      "learning_rate": 1.1961210898092468e-06,
      "loss": 0.66794729,
      "num_input_tokens_seen": 115047665,
      "step": 5345,
      "time_per_iteration": 2.6514055728912354
    },
    {
      "auxiliary_loss_clip": 0.01176529,
      "auxiliary_loss_mlp": 0.01028219,
      "balance_loss_clip": 1.05155277,
      "balance_loss_mlp": 1.02030945,
      "epoch": 0.6428184933565803,
      "flos": 17851984456320.0,
      "grad_norm": 2.4929084963248327,
      "language_loss": 0.79340023,
      "learning_rate": 1.1954078727569874e-06,
      "loss": 0.81544775,
      "num_input_tokens_seen": 115064965,
      "step": 5346,
      "time_per_iteration": 2.584108829498291
    },
    {
      "auxiliary_loss_clip": 0.01175922,
      "auxiliary_loss_mlp": 0.01028925,
      "balance_loss_clip": 1.05242443,
      "balance_loss_mlp": 1.02061641,
      "epoch": 0.6429387362472194,
      "flos": 22456953820800.0,
      "grad_norm": 2.737245791604055,
      "language_loss": 0.77968919,
      "learning_rate": 1.1946947777607141e-06,
      "loss": 0.80173761,
      "num_input_tokens_seen": 115086100,
      "step": 5347,
      "time_per_iteration": 4.467098236083984
    },
    {
      "auxiliary_loss_clip": 0.01177154,
      "auxiliary_loss_mlp": 0.01028477,
      "balance_loss_clip": 1.05001378,
      "balance_loss_mlp": 1.02019823,
      "epoch": 0.6430589791378585,
      "flos": 24752579005440.0,
      "grad_norm": 2.524392893599534,
      "language_loss": 0.79901493,
      "learning_rate": 1.1939818049286024e-06,
      "loss": 0.82107127,
      "num_input_tokens_seen": 115104260,
      "step": 5348,
      "time_per_iteration": 2.6581063270568848
    },
    {
      "auxiliary_loss_clip": 0.01176976,
      "auxiliary_loss_mlp": 0.01026304,
      "balance_loss_clip": 1.05223632,
      "balance_loss_mlp": 1.01860285,
      "epoch": 0.6431792220284975,
      "flos": 24901249397760.0,
      "grad_norm": 2.2175102647432365,
      "language_loss": 0.75485164,
      "learning_rate": 1.1932689543688101e-06,
      "loss": 0.77688438,
      "num_input_tokens_seen": 115125365,
      "step": 5349,
      "time_per_iteration": 3.5448760986328125
    },
    {
      "auxiliary_loss_clip": 0.01178348,
      "auxiliary_loss_mlp": 0.01025712,
      "balance_loss_clip": 1.05271196,
      "balance_loss_mlp": 1.01801658,
      "epoch": 0.6432994649191367,
      "flos": 21032305620480.0,
      "grad_norm": 1.9012698814530682,
      "language_loss": 0.72601473,
      "learning_rate": 1.1925562261894756e-06,
      "loss": 0.74805534,
      "num_input_tokens_seen": 115144445,
      "step": 5350,
      "time_per_iteration": 2.6100611686706543
    },
    {
      "auxiliary_loss_clip": 0.01172852,
      "auxiliary_loss_mlp": 0.01027067,
      "balance_loss_clip": 1.04971886,
      "balance_loss_mlp": 1.0198369,
      "epoch": 0.6434197078097758,
      "flos": 30884433655680.0,
      "grad_norm": 1.8524821414625698,
      "language_loss": 0.77486408,
      "learning_rate": 1.1918436204987207e-06,
      "loss": 0.79686326,
      "num_input_tokens_seen": 115166305,
      "step": 5351,
      "time_per_iteration": 2.651644468307495
    },
    {
      "auxiliary_loss_clip": 0.01173807,
      "auxiliary_loss_mlp": 0.01025957,
      "balance_loss_clip": 1.05026698,
      "balance_loss_mlp": 1.01829815,
      "epoch": 0.6435399507004148,
      "flos": 15012492468480.0,
      "grad_norm": 2.8659050566857815,
      "language_loss": 0.82014179,
      "learning_rate": 1.191131137404645e-06,
      "loss": 0.84213936,
      "num_input_tokens_seen": 115183045,
      "step": 5352,
      "time_per_iteration": 2.6165778636932373
    },
    {
      "auxiliary_loss_clip": 0.01179761,
      "auxiliary_loss_mlp": 0.01029217,
      "balance_loss_clip": 1.05334198,
      "balance_loss_mlp": 1.02162361,
      "epoch": 0.643660193591054,
      "flos": 19901981462400.0,
      "grad_norm": 2.153274541070452,
      "language_loss": 0.77232921,
      "learning_rate": 1.190418777015333e-06,
      "loss": 0.79441899,
      "num_input_tokens_seen": 115201955,
      "step": 5353,
      "time_per_iteration": 2.6108293533325195
    },
    {
      "auxiliary_loss_clip": 0.01172828,
      "auxiliary_loss_mlp": 0.01030678,
      "balance_loss_clip": 1.0493083,
      "balance_loss_mlp": 1.02316177,
      "epoch": 0.643780436481693,
      "flos": 24133622820480.0,
      "grad_norm": 2.1871444656447614,
      "language_loss": 0.73869717,
      "learning_rate": 1.1897065394388487e-06,
      "loss": 0.76073229,
      "num_input_tokens_seen": 115222395,
      "step": 5354,
      "time_per_iteration": 2.706484794616699
    },
    {
      "auxiliary_loss_clip": 0.01175234,
      "auxiliary_loss_mlp": 0.01029737,
      "balance_loss_clip": 1.05219269,
      "balance_loss_mlp": 1.02203631,
      "epoch": 0.6439006793723321,
      "flos": 23148808657920.0,
      "grad_norm": 2.171795567585254,
      "language_loss": 0.76530766,
      "learning_rate": 1.1889944247832385e-06,
      "loss": 0.78735733,
      "num_input_tokens_seen": 115242635,
      "step": 5355,
      "time_per_iteration": 2.5984268188476562
    },
    {
      "auxiliary_loss_clip": 0.01176008,
      "auxiliary_loss_mlp": 0.01025813,
      "balance_loss_clip": 1.05068409,
      "balance_loss_mlp": 1.01776636,
      "epoch": 0.6440209222629713,
      "flos": 23617909301760.0,
      "grad_norm": 7.197213740904694,
      "language_loss": 0.70764542,
      "learning_rate": 1.1882824331565283e-06,
      "loss": 0.72966367,
      "num_input_tokens_seen": 115262095,
      "step": 5356,
      "time_per_iteration": 2.6888232231140137
    },
    {
      "auxiliary_loss_clip": 0.01172721,
      "auxiliary_loss_mlp": 0.01025837,
      "balance_loss_clip": 1.05068135,
      "balance_loss_mlp": 1.01833928,
      "epoch": 0.6441411651536103,
      "flos": 16544872535040.0,
      "grad_norm": 2.285775578544807,
      "language_loss": 0.88902479,
      "learning_rate": 1.1875705646667287e-06,
      "loss": 0.91101038,
      "num_input_tokens_seen": 115279985,
      "step": 5357,
      "time_per_iteration": 2.6090500354766846
    },
    {
      "auxiliary_loss_clip": 0.0117358,
      "auxiliary_loss_mlp": 0.0102246,
      "balance_loss_clip": 1.04969811,
      "balance_loss_mlp": 1.01494408,
      "epoch": 0.6442614080442494,
      "flos": 25410965345280.0,
      "grad_norm": 4.241375101257576,
      "language_loss": 0.75652665,
      "learning_rate": 1.1868588194218282e-06,
      "loss": 0.77848709,
      "num_input_tokens_seen": 115300365,
      "step": 5358,
      "time_per_iteration": 2.6342227458953857
    },
    {
      "auxiliary_loss_clip": 0.01177687,
      "auxiliary_loss_mlp": 0.01026737,
      "balance_loss_clip": 1.05225337,
      "balance_loss_mlp": 1.01822579,
      "epoch": 0.6443816509348885,
      "flos": 28294017552000.0,
      "grad_norm": 1.934943131889981,
      "language_loss": 0.7428807,
      "learning_rate": 1.1861471975297979e-06,
      "loss": 0.764925,
      "num_input_tokens_seen": 115322060,
      "step": 5359,
      "time_per_iteration": 2.727104663848877
    },
    {
      "auxiliary_loss_clip": 0.01177703,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 1.05138111,
      "balance_loss_mlp": 1.01997852,
      "epoch": 0.6445018938255276,
      "flos": 36690075964800.0,
      "grad_norm": 5.866546127478425,
      "language_loss": 0.71095353,
      "learning_rate": 1.185435699098591e-06,
      "loss": 0.73301136,
      "num_input_tokens_seen": 115348255,
      "step": 5360,
      "time_per_iteration": 2.726167917251587
    },
    {
      "auxiliary_loss_clip": 0.01179149,
      "auxiliary_loss_mlp": 0.01027208,
      "balance_loss_clip": 1.0531683,
      "balance_loss_mlp": 1.01838672,
      "epoch": 0.6446221367161666,
      "flos": 14501411804160.0,
      "grad_norm": 2.442399835841918,
      "language_loss": 0.78110272,
      "learning_rate": 1.1847243242361403e-06,
      "loss": 0.80316627,
      "num_input_tokens_seen": 115366845,
      "step": 5361,
      "time_per_iteration": 2.6547603607177734
    },
    {
      "auxiliary_loss_clip": 0.01174542,
      "auxiliary_loss_mlp": 0.01023743,
      "balance_loss_clip": 1.04937661,
      "balance_loss_mlp": 1.01602399,
      "epoch": 0.6447423796068057,
      "flos": 24609367480320.0,
      "grad_norm": 2.2107606343738135,
      "language_loss": 0.78118062,
      "learning_rate": 1.1840130730503624e-06,
      "loss": 0.80316341,
      "num_input_tokens_seen": 115388125,
      "step": 5362,
      "time_per_iteration": 2.6826579570770264
    },
    {
      "auxiliary_loss_clip": 0.0117585,
      "auxiliary_loss_mlp": 0.01022279,
      "balance_loss_clip": 1.05098879,
      "balance_loss_mlp": 1.0148108,
      "epoch": 0.6448626224974449,
      "flos": 25047298097280.0,
      "grad_norm": 4.138287667227883,
      "language_loss": 0.74656588,
      "learning_rate": 1.1833019456491518e-06,
      "loss": 0.76854718,
      "num_input_tokens_seen": 115409655,
      "step": 5363,
      "time_per_iteration": 2.646881580352783
    },
    {
      "auxiliary_loss_clip": 0.01176183,
      "auxiliary_loss_mlp": 0.01022146,
      "balance_loss_clip": 1.05069399,
      "balance_loss_mlp": 1.01390314,
      "epoch": 0.6449828653880839,
      "flos": 22530355263360.0,
      "grad_norm": 2.0533143134860623,
      "language_loss": 0.78755498,
      "learning_rate": 1.1825909421403871e-06,
      "loss": 0.80953825,
      "num_input_tokens_seen": 115428750,
      "step": 5364,
      "time_per_iteration": 2.61936616897583
    },
    {
      "auxiliary_loss_clip": 0.01173076,
      "auxiliary_loss_mlp": 0.01025028,
      "balance_loss_clip": 1.05010509,
      "balance_loss_mlp": 1.01757395,
      "epoch": 0.645103108278723,
      "flos": 25695736369920.0,
      "grad_norm": 2.311484963254829,
      "language_loss": 0.76331389,
      "learning_rate": 1.181880062631926e-06,
      "loss": 0.78529489,
      "num_input_tokens_seen": 115448085,
      "step": 5365,
      "time_per_iteration": 2.65228009223938
    },
    {
      "auxiliary_loss_clip": 0.01178329,
      "auxiliary_loss_mlp": 0.01024598,
      "balance_loss_clip": 1.05293274,
      "balance_loss_mlp": 1.01689076,
      "epoch": 0.6452233511693621,
      "flos": 27450331925760.0,
      "grad_norm": 2.148065348291934,
      "language_loss": 0.84936243,
      "learning_rate": 1.1811693072316093e-06,
      "loss": 0.87139165,
      "num_input_tokens_seen": 115465765,
      "step": 5366,
      "time_per_iteration": 2.576075792312622
    },
    {
      "auxiliary_loss_clip": 0.01174207,
      "auxiliary_loss_mlp": 0.01021352,
      "balance_loss_clip": 1.04915869,
      "balance_loss_mlp": 1.01335955,
      "epoch": 0.6453435940600012,
      "flos": 19208618254080.0,
      "grad_norm": 2.2611259515558695,
      "language_loss": 0.83801407,
      "learning_rate": 1.1804586760472574e-06,
      "loss": 0.85996968,
      "num_input_tokens_seen": 115482230,
      "step": 5367,
      "time_per_iteration": 2.598121166229248
    },
    {
      "auxiliary_loss_clip": 0.01174368,
      "auxiliary_loss_mlp": 0.01022462,
      "balance_loss_clip": 1.04960847,
      "balance_loss_mlp": 1.01488614,
      "epoch": 0.6454638369506402,
      "flos": 25737680476800.0,
      "grad_norm": 2.6529684033740444,
      "language_loss": 0.80308777,
      "learning_rate": 1.1797481691866736e-06,
      "loss": 0.82505608,
      "num_input_tokens_seen": 115499455,
      "step": 5368,
      "time_per_iteration": 3.6216378211975098
    },
    {
      "auxiliary_loss_clip": 0.01174022,
      "auxiliary_loss_mlp": 0.01029539,
      "balance_loss_clip": 1.05187106,
      "balance_loss_mlp": 1.02222848,
      "epoch": 0.6455840798412794,
      "flos": 20989176364800.0,
      "grad_norm": 2.470352676458114,
      "language_loss": 0.83364409,
      "learning_rate": 1.1790377867576393e-06,
      "loss": 0.85567969,
      "num_input_tokens_seen": 115517205,
      "step": 5369,
      "time_per_iteration": 2.6352591514587402
    },
    {
      "auxiliary_loss_clip": 0.01174783,
      "auxiliary_loss_mlp": 0.01025604,
      "balance_loss_clip": 1.05032992,
      "balance_loss_mlp": 1.01767302,
      "epoch": 0.6457043227319185,
      "flos": 26067556005120.0,
      "grad_norm": 1.988134562893504,
      "language_loss": 0.76443398,
      "learning_rate": 1.1783275288679203e-06,
      "loss": 0.78643787,
      "num_input_tokens_seen": 115534370,
      "step": 5370,
      "time_per_iteration": 2.7268593311309814
    },
    {
      "auxiliary_loss_clip": 0.01081798,
      "auxiliary_loss_mlp": 0.0100551,
      "balance_loss_clip": 1.01923466,
      "balance_loss_mlp": 1.00387692,
      "epoch": 0.6458245656225575,
      "flos": 60370831088640.0,
      "grad_norm": 0.8561050631821066,
      "language_loss": 0.5707339,
      "learning_rate": 1.177617395625262e-06,
      "loss": 0.59160697,
      "num_input_tokens_seen": 115592345,
      "step": 5371,
      "time_per_iteration": 3.150517225265503
    },
    {
      "auxiliary_loss_clip": 0.01174039,
      "auxiliary_loss_mlp": 0.01024467,
      "balance_loss_clip": 1.04993558,
      "balance_loss_mlp": 1.01671827,
      "epoch": 0.6459448085131967,
      "flos": 23076771932160.0,
      "grad_norm": 1.8054236965295114,
      "language_loss": 0.75712997,
      "learning_rate": 1.1769073871373908e-06,
      "loss": 0.77911496,
      "num_input_tokens_seen": 115612550,
      "step": 5372,
      "time_per_iteration": 2.6353518962860107
    },
    {
      "auxiliary_loss_clip": 0.01174183,
      "auxiliary_loss_mlp": 0.01025205,
      "balance_loss_clip": 1.04939413,
      "balance_loss_mlp": 1.01784682,
      "epoch": 0.6460650514038357,
      "flos": 22598190097920.0,
      "grad_norm": 1.726599873885758,
      "language_loss": 0.837852,
      "learning_rate": 1.176197503512015e-06,
      "loss": 0.85984594,
      "num_input_tokens_seen": 115632265,
      "step": 5373,
      "time_per_iteration": 3.6285910606384277
    },
    {
      "auxiliary_loss_clip": 0.01177652,
      "auxiliary_loss_mlp": 0.01026809,
      "balance_loss_clip": 1.05473745,
      "balance_loss_mlp": 1.0199368,
      "epoch": 0.6461852942944748,
      "flos": 20266726118400.0,
      "grad_norm": 2.2216206901533746,
      "language_loss": 0.82412016,
      "learning_rate": 1.1754877448568223e-06,
      "loss": 0.84616482,
      "num_input_tokens_seen": 115651720,
      "step": 5374,
      "time_per_iteration": 3.511188268661499
    },
    {
      "auxiliary_loss_clip": 0.0117446,
      "auxiliary_loss_mlp": 0.01019686,
      "balance_loss_clip": 1.05079734,
      "balance_loss_mlp": 1.01233673,
      "epoch": 0.646305537185114,
      "flos": 23367109564800.0,
      "grad_norm": 3.9793969679540764,
      "language_loss": 0.90218532,
      "learning_rate": 1.1747781112794837e-06,
      "loss": 0.92412686,
      "num_input_tokens_seen": 115668215,
      "step": 5375,
      "time_per_iteration": 3.593773365020752
    },
    {
      "auxiliary_loss_clip": 0.0117837,
      "auxiliary_loss_mlp": 0.01030027,
      "balance_loss_clip": 1.05439186,
      "balance_loss_mlp": 1.02265418,
      "epoch": 0.646425780075753,
      "flos": 24277480790400.0,
      "grad_norm": 1.825868068770928,
      "language_loss": 0.83065379,
      "learning_rate": 1.1740686028876487e-06,
      "loss": 0.85273778,
      "num_input_tokens_seen": 115687080,
      "step": 5376,
      "time_per_iteration": 2.667872428894043
    },
    {
      "auxiliary_loss_clip": 0.01174737,
      "auxiliary_loss_mlp": 0.01029204,
      "balance_loss_clip": 1.05181754,
      "balance_loss_mlp": 1.02143764,
      "epoch": 0.6465460229663921,
      "flos": 20813968800000.0,
      "grad_norm": 2.7088037035290466,
      "language_loss": 0.75168568,
      "learning_rate": 1.1733592197889507e-06,
      "loss": 0.77372503,
      "num_input_tokens_seen": 115703990,
      "step": 5377,
      "time_per_iteration": 2.671510934829712
    },
    {
      "auxiliary_loss_clip": 0.01171451,
      "auxiliary_loss_mlp": 0.01024119,
      "balance_loss_clip": 1.04979539,
      "balance_loss_mlp": 1.0166564,
      "epoch": 0.6466662658570312,
      "flos": 22853299466880.0,
      "grad_norm": 1.9815537460697492,
      "language_loss": 0.72718859,
      "learning_rate": 1.1726499620910014e-06,
      "loss": 0.74914432,
      "num_input_tokens_seen": 115724270,
      "step": 5378,
      "time_per_iteration": 2.6225178241729736
    },
    {
      "auxiliary_loss_clip": 0.0117579,
      "auxiliary_loss_mlp": 0.01024739,
      "balance_loss_clip": 1.04958892,
      "balance_loss_mlp": 1.01678133,
      "epoch": 0.6467865087476703,
      "flos": 15304553953920.0,
      "grad_norm": 2.1636732170031725,
      "language_loss": 0.78535265,
      "learning_rate": 1.1719408299013955e-06,
      "loss": 0.80735791,
      "num_input_tokens_seen": 115742995,
      "step": 5379,
      "time_per_iteration": 2.557655096054077
    },
    {
      "auxiliary_loss_clip": 0.01173862,
      "auxiliary_loss_mlp": 0.01022653,
      "balance_loss_clip": 1.05121565,
      "balance_loss_mlp": 1.01513672,
      "epoch": 0.6469067516383094,
      "flos": 19573650218880.0,
      "grad_norm": 3.472748759634805,
      "language_loss": 0.75807291,
      "learning_rate": 1.1712318233277067e-06,
      "loss": 0.78003812,
      "num_input_tokens_seen": 115762015,
      "step": 5380,
      "time_per_iteration": 2.6593868732452393
    },
    {
      "auxiliary_loss_clip": 0.0108068,
      "auxiliary_loss_mlp": 0.01002424,
      "balance_loss_clip": 1.01837587,
      "balance_loss_mlp": 1.00099325,
      "epoch": 0.6470269945289485,
      "flos": 65098002522240.0,
      "grad_norm": 0.764806347837756,
      "language_loss": 0.57854748,
      "learning_rate": 1.1705229424774916e-06,
      "loss": 0.59937859,
      "num_input_tokens_seen": 115816285,
      "step": 5381,
      "time_per_iteration": 3.136270761489868
    },
    {
      "auxiliary_loss_clip": 0.01172986,
      "auxiliary_loss_mlp": 0.01025594,
      "balance_loss_clip": 1.04974818,
      "balance_loss_mlp": 1.01813102,
      "epoch": 0.6471472374195876,
      "flos": 30696943639680.0,
      "grad_norm": 7.56903587568447,
      "language_loss": 0.64391422,
      "learning_rate": 1.1698141874582867e-06,
      "loss": 0.66590005,
      "num_input_tokens_seen": 115837330,
      "step": 5382,
      "time_per_iteration": 2.7020115852355957
    },
    {
      "auxiliary_loss_clip": 0.01171441,
      "auxiliary_loss_mlp": 0.01024326,
      "balance_loss_clip": 1.05057001,
      "balance_loss_mlp": 1.01692331,
      "epoch": 0.6472674803102266,
      "flos": 20521835487360.0,
      "grad_norm": 2.0898899278126937,
      "language_loss": 0.72242475,
      "learning_rate": 1.169105558377609e-06,
      "loss": 0.74438238,
      "num_input_tokens_seen": 115857420,
      "step": 5383,
      "time_per_iteration": 2.6455795764923096
    },
    {
      "auxiliary_loss_clip": 0.01174298,
      "auxiliary_loss_mlp": 0.01025148,
      "balance_loss_clip": 1.05277812,
      "balance_loss_mlp": 1.0175066,
      "epoch": 0.6473877232008658,
      "flos": 24715447320960.0,
      "grad_norm": 1.8685062240181036,
      "language_loss": 0.78653079,
      "learning_rate": 1.1683970553429587e-06,
      "loss": 0.8085252,
      "num_input_tokens_seen": 115878875,
      "step": 5384,
      "time_per_iteration": 2.728572130203247
    },
    {
      "auxiliary_loss_clip": 0.01175726,
      "auxiliary_loss_mlp": 0.01026469,
      "balance_loss_clip": 1.0515908,
      "balance_loss_mlp": 1.01833916,
      "epoch": 0.6475079660915048,
      "flos": 15885552441600.0,
      "grad_norm": 2.1651449857919705,
      "language_loss": 0.82605982,
      "learning_rate": 1.1676886784618128e-06,
      "loss": 0.84808183,
      "num_input_tokens_seen": 115895540,
      "step": 5385,
      "time_per_iteration": 2.5694756507873535
    },
    {
      "auxiliary_loss_clip": 0.01177524,
      "auxiliary_loss_mlp": 0.01023786,
      "balance_loss_clip": 1.05065465,
      "balance_loss_mlp": 1.01532173,
      "epoch": 0.6476282089821439,
      "flos": 17381590922880.0,
      "grad_norm": 2.5941143442056376,
      "language_loss": 0.8374505,
      "learning_rate": 1.1669804278416332e-06,
      "loss": 0.85946357,
      "num_input_tokens_seen": 115910265,
      "step": 5386,
      "time_per_iteration": 2.6223809719085693
    },
    {
      "auxiliary_loss_clip": 0.01177437,
      "auxiliary_loss_mlp": 0.01026115,
      "balance_loss_clip": 1.05214286,
      "balance_loss_mlp": 1.01822329,
      "epoch": 0.6477484518727831,
      "flos": 20194078861440.0,
      "grad_norm": 1.8963661515838681,
      "language_loss": 0.71560204,
      "learning_rate": 1.1662723035898602e-06,
      "loss": 0.73763758,
      "num_input_tokens_seen": 115930025,
      "step": 5387,
      "time_per_iteration": 2.6486220359802246
    },
    {
      "auxiliary_loss_clip": 0.01175554,
      "auxiliary_loss_mlp": 0.01020604,
      "balance_loss_clip": 1.05078638,
      "balance_loss_mlp": 1.01257515,
      "epoch": 0.6478686947634221,
      "flos": 25410426641280.0,
      "grad_norm": 1.7222280635747256,
      "language_loss": 0.8202334,
      "learning_rate": 1.165564305813915e-06,
      "loss": 0.84219497,
      "num_input_tokens_seen": 115949025,
      "step": 5388,
      "time_per_iteration": 2.724452495574951
    },
    {
      "auxiliary_loss_clip": 0.01172313,
      "auxiliary_loss_mlp": 0.01024013,
      "balance_loss_clip": 1.05066204,
      "balance_loss_mlp": 1.0168221,
      "epoch": 0.6479889376540612,
      "flos": 20083581648000.0,
      "grad_norm": 1.812609271757373,
      "language_loss": 0.81602699,
      "learning_rate": 1.1648564346212019e-06,
      "loss": 0.83799028,
      "num_input_tokens_seen": 115968145,
      "step": 5389,
      "time_per_iteration": 2.841592788696289
    },
    {
      "auxiliary_loss_clip": 0.01171735,
      "auxiliary_loss_mlp": 0.01023979,
      "balance_loss_clip": 1.05020189,
      "balance_loss_mlp": 1.01699066,
      "epoch": 0.6481091805447003,
      "flos": 26758082039040.0,
      "grad_norm": 3.228205650741951,
      "language_loss": 0.76471514,
      "learning_rate": 1.164148690119104e-06,
      "loss": 0.78667229,
      "num_input_tokens_seen": 115989425,
      "step": 5390,
      "time_per_iteration": 2.6359682083129883
    },
    {
      "auxiliary_loss_clip": 0.0117128,
      "auxiliary_loss_mlp": 0.01024497,
      "balance_loss_clip": 1.04850292,
      "balance_loss_mlp": 1.01700509,
      "epoch": 0.6482294234353394,
      "flos": 23952094462080.0,
      "grad_norm": 1.7943967881648584,
      "language_loss": 0.74426162,
      "learning_rate": 1.163441072414985e-06,
      "loss": 0.76621938,
      "num_input_tokens_seen": 116009630,
      "step": 5391,
      "time_per_iteration": 2.6819589138031006
    },
    {
      "auxiliary_loss_clip": 0.01172019,
      "auxiliary_loss_mlp": 0.0102433,
      "balance_loss_clip": 1.05093789,
      "balance_loss_mlp": 1.01677847,
      "epoch": 0.6483496663259785,
      "flos": 26209833776640.0,
      "grad_norm": 1.9106057257182327,
      "language_loss": 0.69907665,
      "learning_rate": 1.16273358161619e-06,
      "loss": 0.72104013,
      "num_input_tokens_seen": 116029965,
      "step": 5392,
      "time_per_iteration": 2.668813467025757
    },
    {
      "auxiliary_loss_clip": 0.01176464,
      "auxiliary_loss_mlp": 0.0102428,
      "balance_loss_clip": 1.05145538,
      "balance_loss_mlp": 1.01641846,
      "epoch": 0.6484699092166175,
      "flos": 20922239370240.0,
      "grad_norm": 2.2442140062586544,
      "language_loss": 0.83908498,
      "learning_rate": 1.1620262178300446e-06,
      "loss": 0.86109245,
      "num_input_tokens_seen": 116048580,
      "step": 5393,
      "time_per_iteration": 2.6196963787078857
    },
    {
      "auxiliary_loss_clip": 0.0117378,
      "auxiliary_loss_mlp": 0.01019701,
      "balance_loss_clip": 1.05133832,
      "balance_loss_mlp": 1.01223803,
      "epoch": 0.6485901521072567,
      "flos": 33072865678080.0,
      "grad_norm": 2.0500537276813335,
      "language_loss": 0.75979269,
      "learning_rate": 1.1613189811638563e-06,
      "loss": 0.78172755,
      "num_input_tokens_seen": 116070305,
      "step": 5394,
      "time_per_iteration": 2.659989356994629
    },
    {
      "auxiliary_loss_clip": 0.011754,
      "auxiliary_loss_mlp": 0.01028754,
      "balance_loss_clip": 1.05452394,
      "balance_loss_mlp": 1.02153921,
      "epoch": 0.6487103949978957,
      "flos": 22274060745600.0,
      "grad_norm": 1.8327009056158752,
      "language_loss": 0.78359669,
      "learning_rate": 1.1606118717249117e-06,
      "loss": 0.80563819,
      "num_input_tokens_seen": 116090405,
      "step": 5395,
      "time_per_iteration": 3.49552321434021
    },
    {
      "auxiliary_loss_clip": 0.01177483,
      "auxiliary_loss_mlp": 0.01028875,
      "balance_loss_clip": 1.05039489,
      "balance_loss_mlp": 1.02064347,
      "epoch": 0.6488306378885348,
      "flos": 22930400010240.0,
      "grad_norm": 1.8370400318474074,
      "language_loss": 0.68145895,
      "learning_rate": 1.1599048896204787e-06,
      "loss": 0.7035225,
      "num_input_tokens_seen": 116110285,
      "step": 5396,
      "time_per_iteration": 2.6623332500457764
    },
    {
      "auxiliary_loss_clip": 0.01174817,
      "auxiliary_loss_mlp": 0.01024636,
      "balance_loss_clip": 1.05070627,
      "balance_loss_mlp": 1.01722765,
      "epoch": 0.648950880779174,
      "flos": 20376110010240.0,
      "grad_norm": 2.0048968314214037,
      "language_loss": 0.80826765,
      "learning_rate": 1.1591980349578061e-06,
      "loss": 0.83026218,
      "num_input_tokens_seen": 116128955,
      "step": 5397,
      "time_per_iteration": 2.7018685340881348
    },
    {
      "auxiliary_loss_clip": 0.01077381,
      "auxiliary_loss_mlp": 0.01001603,
      "balance_loss_clip": 1.01598263,
      "balance_loss_mlp": 1.00024974,
      "epoch": 0.649071123669813,
      "flos": 59930889310080.0,
      "grad_norm": 0.7315913557107455,
      "language_loss": 0.54257172,
      "learning_rate": 1.158491307844123e-06,
      "loss": 0.56336159,
      "num_input_tokens_seen": 116188875,
      "step": 5398,
      "time_per_iteration": 3.18929386138916
    },
    {
      "auxiliary_loss_clip": 0.01172916,
      "auxiliary_loss_mlp": 0.01022036,
      "balance_loss_clip": 1.05096483,
      "balance_loss_mlp": 1.01455545,
      "epoch": 0.6491913665604521,
      "flos": 20446566537600.0,
      "grad_norm": 1.9530575726707309,
      "language_loss": 0.83954084,
      "learning_rate": 1.1577847083866387e-06,
      "loss": 0.86149037,
      "num_input_tokens_seen": 116207910,
      "step": 5399,
      "time_per_iteration": 3.5655710697174072
    },
    {
      "auxiliary_loss_clip": 0.01172856,
      "auxiliary_loss_mlp": 0.01024821,
      "balance_loss_clip": 1.04921722,
      "balance_loss_mlp": 1.01689982,
      "epoch": 0.6493116094510912,
      "flos": 16946820702720.0,
      "grad_norm": 1.880270722824935,
      "language_loss": 0.7197628,
      "learning_rate": 1.1570782366925453e-06,
      "loss": 0.74173963,
      "num_input_tokens_seen": 116226425,
      "step": 5400,
      "time_per_iteration": 2.6749463081359863
    },
    {
      "auxiliary_loss_clip": 0.01173031,
      "auxiliary_loss_mlp": 0.01020773,
      "balance_loss_clip": 1.04794741,
      "balance_loss_mlp": 1.01282811,
      "epoch": 0.6494318523417303,
      "flos": 18802935072000.0,
      "grad_norm": 2.1137960285205204,
      "language_loss": 0.75674248,
      "learning_rate": 1.1563718928690132e-06,
      "loss": 0.77868056,
      "num_input_tokens_seen": 116243860,
      "step": 5401,
      "time_per_iteration": 2.669339179992676
    },
    {
      "auxiliary_loss_clip": 0.01174111,
      "auxiliary_loss_mlp": 0.01029682,
      "balance_loss_clip": 1.050318,
      "balance_loss_mlp": 1.02155232,
      "epoch": 0.6495520952323693,
      "flos": 18982847318400.0,
      "grad_norm": 3.152513236706009,
      "language_loss": 0.71236151,
      "learning_rate": 1.1556656770231942e-06,
      "loss": 0.73439944,
      "num_input_tokens_seen": 116260055,
      "step": 5402,
      "time_per_iteration": 3.5056982040405273
    },
    {
      "auxiliary_loss_clip": 0.01173162,
      "auxiliary_loss_mlp": 0.01025884,
      "balance_loss_clip": 1.05018687,
      "balance_loss_mlp": 1.01851106,
      "epoch": 0.6496723381230085,
      "flos": 22745388032640.0,
      "grad_norm": 1.9133799193011698,
      "language_loss": 0.76048517,
      "learning_rate": 1.1549595892622207e-06,
      "loss": 0.78247559,
      "num_input_tokens_seen": 116278825,
      "step": 5403,
      "time_per_iteration": 2.692844867706299
    },
    {
      "auxiliary_loss_clip": 0.01076544,
      "auxiliary_loss_mlp": 0.01003071,
      "balance_loss_clip": 1.01541662,
      "balance_loss_mlp": 1.00160491,
      "epoch": 0.6497925810136476,
      "flos": 62145283887360.0,
      "grad_norm": 0.8135880636066298,
      "language_loss": 0.59000415,
      "learning_rate": 1.1542536296932047e-06,
      "loss": 0.61080027,
      "num_input_tokens_seen": 116342360,
      "step": 5404,
      "time_per_iteration": 3.155388593673706
    },
    {
      "auxiliary_loss_clip": 0.01178979,
      "auxiliary_loss_mlp": 0.01027329,
      "balance_loss_clip": 1.05307388,
      "balance_loss_mlp": 1.01939893,
      "epoch": 0.6499128239042866,
      "flos": 20156731695360.0,
      "grad_norm": 6.057535354823579,
      "language_loss": 0.70147604,
      "learning_rate": 1.1535477984232414e-06,
      "loss": 0.72353911,
      "num_input_tokens_seen": 116362235,
      "step": 5405,
      "time_per_iteration": 2.6462297439575195
    },
    {
      "auxiliary_loss_clip": 0.01172902,
      "auxiliary_loss_mlp": 0.01028058,
      "balance_loss_clip": 1.04946208,
      "balance_loss_mlp": 1.02027404,
      "epoch": 0.6500330667949258,
      "flos": 24462420940800.0,
      "grad_norm": 1.9779948544518555,
      "language_loss": 0.77140605,
      "learning_rate": 1.152842095559404e-06,
      "loss": 0.79341567,
      "num_input_tokens_seen": 116382895,
      "step": 5406,
      "time_per_iteration": 2.635765552520752
    },
    {
      "auxiliary_loss_clip": 0.0117131,
      "auxiliary_loss_mlp": 0.01024264,
      "balance_loss_clip": 1.04861903,
      "balance_loss_mlp": 1.01687574,
      "epoch": 0.6501533096855648,
      "flos": 25477399549440.0,
      "grad_norm": 1.7644688237364987,
      "language_loss": 0.7671327,
      "learning_rate": 1.1521365212087474e-06,
      "loss": 0.78908849,
      "num_input_tokens_seen": 116402880,
      "step": 5407,
      "time_per_iteration": 2.7143502235412598
    },
    {
      "auxiliary_loss_clip": 0.01173977,
      "auxiliary_loss_mlp": 0.01031478,
      "balance_loss_clip": 1.04802418,
      "balance_loss_mlp": 1.02316952,
      "epoch": 0.6502735525762039,
      "flos": 44819245347840.0,
      "grad_norm": 2.2348044732520664,
      "language_loss": 0.70815992,
      "learning_rate": 1.1514310754783062e-06,
      "loss": 0.73021448,
      "num_input_tokens_seen": 116425830,
      "step": 5408,
      "time_per_iteration": 2.8057801723480225
    },
    {
      "auxiliary_loss_clip": 0.01174551,
      "auxiliary_loss_mlp": 0.01022393,
      "balance_loss_clip": 1.05169439,
      "balance_loss_mlp": 1.01533878,
      "epoch": 0.6503937954668431,
      "flos": 28658546726400.0,
      "grad_norm": 5.049416060588733,
      "language_loss": 0.73581016,
      "learning_rate": 1.1507257584750964e-06,
      "loss": 0.75777954,
      "num_input_tokens_seen": 116446010,
      "step": 5409,
      "time_per_iteration": 2.71881103515625
    },
    {
      "auxiliary_loss_clip": 0.01174976,
      "auxiliary_loss_mlp": 0.01023897,
      "balance_loss_clip": 1.05111384,
      "balance_loss_mlp": 1.01604676,
      "epoch": 0.6505140383574821,
      "flos": 20922562592640.0,
      "grad_norm": 2.488238458180619,
      "language_loss": 0.77989054,
      "learning_rate": 1.150020570306113e-06,
      "loss": 0.80187929,
      "num_input_tokens_seen": 116465150,
      "step": 5410,
      "time_per_iteration": 2.6606407165527344
    },
    {
      "auxiliary_loss_clip": 0.01176315,
      "auxiliary_loss_mlp": 0.01022255,
      "balance_loss_clip": 1.05075908,
      "balance_loss_mlp": 1.01456308,
      "epoch": 0.6506342812481212,
      "flos": 20595236929920.0,
      "grad_norm": 4.483851781058544,
      "language_loss": 0.74829566,
      "learning_rate": 1.1493155110783338e-06,
      "loss": 0.77028137,
      "num_input_tokens_seen": 116483675,
      "step": 5411,
      "time_per_iteration": 2.6546790599823
    },
    {
      "auxiliary_loss_clip": 0.01176217,
      "auxiliary_loss_mlp": 0.01027236,
      "balance_loss_clip": 1.05132008,
      "balance_loss_mlp": 1.01908207,
      "epoch": 0.6507545241387603,
      "flos": 30226478279040.0,
      "grad_norm": 2.0389812833129817,
      "language_loss": 0.70745379,
      "learning_rate": 1.1486105808987155e-06,
      "loss": 0.72948825,
      "num_input_tokens_seen": 116505165,
      "step": 5412,
      "time_per_iteration": 2.790682554244995
    },
    {
      "auxiliary_loss_clip": 0.01175179,
      "auxiliary_loss_mlp": 0.01028433,
      "balance_loss_clip": 1.05117154,
      "balance_loss_mlp": 1.02045739,
      "epoch": 0.6508747670293994,
      "flos": 17128241320320.0,
      "grad_norm": 2.9899683518280775,
      "language_loss": 0.81138158,
      "learning_rate": 1.1479057798741947e-06,
      "loss": 0.83341771,
      "num_input_tokens_seen": 116523220,
      "step": 5413,
      "time_per_iteration": 2.6292307376861572
    },
    {
      "auxiliary_loss_clip": 0.01074235,
      "auxiliary_loss_mlp": 0.01002046,
      "balance_loss_clip": 1.01373518,
      "balance_loss_mlp": 1.00068092,
      "epoch": 0.6509950099200384,
      "flos": 68559826573440.0,
      "grad_norm": 0.7871832584935101,
      "language_loss": 0.53306329,
      "learning_rate": 1.14720110811169e-06,
      "loss": 0.55382609,
      "num_input_tokens_seen": 116580450,
      "step": 5414,
      "time_per_iteration": 3.201322078704834
    },
    {
      "auxiliary_loss_clip": 0.01180462,
      "auxiliary_loss_mlp": 0.01028125,
      "balance_loss_clip": 1.05398679,
      "balance_loss_mlp": 1.0198158,
      "epoch": 0.6511152528106776,
      "flos": 22347462188160.0,
      "grad_norm": 1.844370058790848,
      "language_loss": 0.7745893,
      "learning_rate": 1.146496565718098e-06,
      "loss": 0.79667521,
      "num_input_tokens_seen": 116601020,
      "step": 5415,
      "time_per_iteration": 2.60471248626709
    },
    {
      "auxiliary_loss_clip": 0.01175746,
      "auxiliary_loss_mlp": 0.0102114,
      "balance_loss_clip": 1.04991889,
      "balance_loss_mlp": 1.01292682,
      "epoch": 0.6512354957013167,
      "flos": 20522158709760.0,
      "grad_norm": 2.294004127643003,
      "language_loss": 0.75867581,
      "learning_rate": 1.1457921528002996e-06,
      "loss": 0.78064466,
      "num_input_tokens_seen": 116619455,
      "step": 5416,
      "time_per_iteration": 2.6934683322906494
    },
    {
      "auxiliary_loss_clip": 0.01176142,
      "auxiliary_loss_mlp": 0.01029021,
      "balance_loss_clip": 1.05118346,
      "balance_loss_mlp": 1.02109373,
      "epoch": 0.6513557385919557,
      "flos": 32337342881280.0,
      "grad_norm": 2.4979052713088055,
      "language_loss": 0.72349286,
      "learning_rate": 1.1450878694651522e-06,
      "loss": 0.74554455,
      "num_input_tokens_seen": 116640020,
      "step": 5417,
      "time_per_iteration": 2.684713840484619
    },
    {
      "auxiliary_loss_clip": 0.01172754,
      "auxiliary_loss_mlp": 0.01025991,
      "balance_loss_clip": 1.04766321,
      "balance_loss_mlp": 1.01793242,
      "epoch": 0.6514759814825949,
      "flos": 12093206417280.0,
      "grad_norm": 3.03546089748703,
      "language_loss": 0.62576318,
      "learning_rate": 1.1443837158194954e-06,
      "loss": 0.64775068,
      "num_input_tokens_seen": 116655165,
      "step": 5418,
      "time_per_iteration": 2.5997133255004883
    },
    {
      "auxiliary_loss_clip": 0.01177771,
      "auxiliary_loss_mlp": 0.01033109,
      "balance_loss_clip": 1.05213642,
      "balance_loss_mlp": 1.02548587,
      "epoch": 0.651596224373234,
      "flos": 22526907557760.0,
      "grad_norm": 1.620061402340739,
      "language_loss": 0.74436343,
      "learning_rate": 1.1436796919701484e-06,
      "loss": 0.76647222,
      "num_input_tokens_seen": 116673880,
      "step": 5419,
      "time_per_iteration": 2.6694657802581787
    },
    {
      "auxiliary_loss_clip": 0.01172804,
      "auxiliary_loss_mlp": 0.0102822,
      "balance_loss_clip": 1.0482564,
      "balance_loss_mlp": 1.02029526,
      "epoch": 0.651716467263873,
      "flos": 27818955250560.0,
      "grad_norm": 1.8897086567778287,
      "language_loss": 0.61917365,
      "learning_rate": 1.1429757980239115e-06,
      "loss": 0.64118385,
      "num_input_tokens_seen": 116694305,
      "step": 5420,
      "time_per_iteration": 2.7014541625976562
    },
    {
      "auxiliary_loss_clip": 0.0117568,
      "auxiliary_loss_mlp": 0.01030139,
      "balance_loss_clip": 1.05043912,
      "balance_loss_mlp": 1.02119875,
      "epoch": 0.6518367101545122,
      "flos": 24316300414080.0,
      "grad_norm": 2.8227628759368124,
      "language_loss": 0.81747103,
      "learning_rate": 1.1422720340875636e-06,
      "loss": 0.83952916,
      "num_input_tokens_seen": 116713055,
      "step": 5421,
      "time_per_iteration": 3.6717231273651123
    },
    {
      "auxiliary_loss_clip": 0.01177214,
      "auxiliary_loss_mlp": 0.01029006,
      "balance_loss_clip": 1.05132174,
      "balance_loss_mlp": 1.02049768,
      "epoch": 0.6519569530451512,
      "flos": 20011939971840.0,
      "grad_norm": 1.983814934548183,
      "language_loss": 0.79087317,
      "learning_rate": 1.1415684002678671e-06,
      "loss": 0.81293541,
      "num_input_tokens_seen": 116731815,
      "step": 5422,
      "time_per_iteration": 2.59065580368042
    },
    {
      "auxiliary_loss_clip": 0.01181778,
      "auxiliary_loss_mlp": 0.01024508,
      "balance_loss_clip": 1.05490017,
      "balance_loss_mlp": 1.0155679,
      "epoch": 0.6520771959357903,
      "flos": 21576064682880.0,
      "grad_norm": 3.777887957484419,
      "language_loss": 0.77351695,
      "learning_rate": 1.1408648966715617e-06,
      "loss": 0.79557979,
      "num_input_tokens_seen": 116749335,
      "step": 5423,
      "time_per_iteration": 2.648587942123413
    },
    {
      "auxiliary_loss_clip": 0.01176238,
      "auxiliary_loss_mlp": 0.01028104,
      "balance_loss_clip": 1.048702,
      "balance_loss_mlp": 1.01952124,
      "epoch": 0.6521974388264293,
      "flos": 22711021695360.0,
      "grad_norm": 5.899480168560915,
      "language_loss": 0.72879887,
      "learning_rate": 1.1401615234053683e-06,
      "loss": 0.75084233,
      "num_input_tokens_seen": 116768155,
      "step": 5424,
      "time_per_iteration": 2.6271488666534424
    },
    {
      "auxiliary_loss_clip": 0.01175911,
      "auxiliary_loss_mlp": 0.01029487,
      "balance_loss_clip": 1.04963255,
      "balance_loss_mlp": 1.02099586,
      "epoch": 0.6523176817170685,
      "flos": 23002939526400.0,
      "grad_norm": 2.374732237610932,
      "language_loss": 0.75901693,
      "learning_rate": 1.1394582805759885e-06,
      "loss": 0.78107095,
      "num_input_tokens_seen": 116787435,
      "step": 5425,
      "time_per_iteration": 2.6880338191986084
    },
    {
      "auxiliary_loss_clip": 0.01177741,
      "auxiliary_loss_mlp": 0.01027638,
      "balance_loss_clip": 1.0528971,
      "balance_loss_mlp": 1.01993454,
      "epoch": 0.6524379246077076,
      "flos": 21688249835520.0,
      "grad_norm": 2.0160818286443036,
      "language_loss": 0.75629151,
      "learning_rate": 1.1387551682901022e-06,
      "loss": 0.77834529,
      "num_input_tokens_seen": 116808040,
      "step": 5426,
      "time_per_iteration": 4.5678887367248535
    },
    {
      "auxiliary_loss_clip": 0.01174351,
      "auxiliary_loss_mlp": 0.01031629,
      "balance_loss_clip": 1.0502336,
      "balance_loss_mlp": 1.02339792,
      "epoch": 0.6525581674983466,
      "flos": 19390936711680.0,
      "grad_norm": 2.765150566126996,
      "language_loss": 0.70769095,
      "learning_rate": 1.138052186654373e-06,
      "loss": 0.72975075,
      "num_input_tokens_seen": 116825510,
      "step": 5427,
      "time_per_iteration": 2.5581939220428467
    },
    {
      "auxiliary_loss_clip": 0.01179775,
      "auxiliary_loss_mlp": 0.0102886,
      "balance_loss_clip": 1.05387235,
      "balance_loss_mlp": 1.0205152,
      "epoch": 0.6526784103889858,
      "flos": 17165444832000.0,
      "grad_norm": 2.4736060158868063,
      "language_loss": 0.88466406,
      "learning_rate": 1.1373493357754417e-06,
      "loss": 0.90675038,
      "num_input_tokens_seen": 116844415,
      "step": 5428,
      "time_per_iteration": 2.5583717823028564
    },
    {
      "auxiliary_loss_clip": 0.01173641,
      "auxiliary_loss_mlp": 0.01025115,
      "balance_loss_clip": 1.04977059,
      "balance_loss_mlp": 1.01689267,
      "epoch": 0.6527986532796248,
      "flos": 18989168112000.0,
      "grad_norm": 1.8904826117398204,
      "language_loss": 0.77018082,
      "learning_rate": 1.1366466157599303e-06,
      "loss": 0.79216838,
      "num_input_tokens_seen": 116863690,
      "step": 5429,
      "time_per_iteration": 3.4947361946105957
    },
    {
      "auxiliary_loss_clip": 0.01176952,
      "auxiliary_loss_mlp": 0.01026673,
      "balance_loss_clip": 1.05110919,
      "balance_loss_mlp": 1.01803601,
      "epoch": 0.6529188961702639,
      "flos": 14238581011200.0,
      "grad_norm": 2.6447164944317754,
      "language_loss": 0.76370144,
      "learning_rate": 1.1359440267144412e-06,
      "loss": 0.78573763,
      "num_input_tokens_seen": 116881145,
      "step": 5430,
      "time_per_iteration": 2.5627996921539307
    },
    {
      "auxiliary_loss_clip": 0.01175261,
      "auxiliary_loss_mlp": 0.01025718,
      "balance_loss_clip": 1.05164981,
      "balance_loss_mlp": 1.0181129,
      "epoch": 0.653039139060903,
      "flos": 36682929158400.0,
      "grad_norm": 1.8051629781372405,
      "language_loss": 0.74318266,
      "learning_rate": 1.1352415687455556e-06,
      "loss": 0.76519245,
      "num_input_tokens_seen": 116902405,
      "step": 5431,
      "time_per_iteration": 2.693225145339966
    },
    {
      "auxiliary_loss_clip": 0.01176996,
      "auxiliary_loss_mlp": 0.01024546,
      "balance_loss_clip": 1.0529623,
      "balance_loss_mlp": 1.01579046,
      "epoch": 0.6531593819515421,
      "flos": 25376275785600.0,
      "grad_norm": 2.7787649866559887,
      "language_loss": 0.64169371,
      "learning_rate": 1.1345392419598362e-06,
      "loss": 0.66370916,
      "num_input_tokens_seen": 116921285,
      "step": 5432,
      "time_per_iteration": 2.8865792751312256
    },
    {
      "auxiliary_loss_clip": 0.01174145,
      "auxiliary_loss_mlp": 0.01027204,
      "balance_loss_clip": 1.05076766,
      "balance_loss_mlp": 1.01921654,
      "epoch": 0.6532796248421812,
      "flos": 21178533888000.0,
      "grad_norm": 1.654716183180113,
      "language_loss": 0.7164306,
      "learning_rate": 1.1338370464638263e-06,
      "loss": 0.73844409,
      "num_input_tokens_seen": 116940685,
      "step": 5433,
      "time_per_iteration": 2.679722547531128
    },
    {
      "auxiliary_loss_clip": 0.01173908,
      "auxiliary_loss_mlp": 0.0102547,
      "balance_loss_clip": 1.0487473,
      "balance_loss_mlp": 1.01763809,
      "epoch": 0.6533998677328203,
      "flos": 17675950878720.0,
      "grad_norm": 2.1343595824291612,
      "language_loss": 0.64572507,
      "learning_rate": 1.1331349823640474e-06,
      "loss": 0.66771883,
      "num_input_tokens_seen": 116958115,
      "step": 5434,
      "time_per_iteration": 2.579855442047119
    },
    {
      "auxiliary_loss_clip": 0.01173394,
      "auxiliary_loss_mlp": 0.01023985,
      "balance_loss_clip": 1.05048752,
      "balance_loss_mlp": 1.01677275,
      "epoch": 0.6535201106234594,
      "flos": 28400384701440.0,
      "grad_norm": 4.2716064484945955,
      "language_loss": 0.77855325,
      "learning_rate": 1.132433049767003e-06,
      "loss": 0.8005271,
      "num_input_tokens_seen": 116976030,
      "step": 5435,
      "time_per_iteration": 2.69272780418396
    },
    {
      "auxiliary_loss_clip": 0.01172265,
      "auxiliary_loss_mlp": 0.01025428,
      "balance_loss_clip": 1.05013216,
      "balance_loss_mlp": 1.01807296,
      "epoch": 0.6536403535140984,
      "flos": 23586667447680.0,
      "grad_norm": 1.53751841627574,
      "language_loss": 0.81059122,
      "learning_rate": 1.1317312487791748e-06,
      "loss": 0.83256817,
      "num_input_tokens_seen": 116997680,
      "step": 5436,
      "time_per_iteration": 2.606569290161133
    },
    {
      "auxiliary_loss_clip": 0.01173618,
      "auxiliary_loss_mlp": 0.01027019,
      "balance_loss_clip": 1.04851532,
      "balance_loss_mlp": 1.01922858,
      "epoch": 0.6537605964047376,
      "flos": 21579476474880.0,
      "grad_norm": 1.8561385812606936,
      "language_loss": 0.732059,
      "learning_rate": 1.1310295795070253e-06,
      "loss": 0.75406539,
      "num_input_tokens_seen": 117017620,
      "step": 5437,
      "time_per_iteration": 2.66574764251709
    },
    {
      "auxiliary_loss_clip": 0.0117502,
      "auxiliary_loss_mlp": 0.01031027,
      "balance_loss_clip": 1.05170727,
      "balance_loss_mlp": 1.02369595,
      "epoch": 0.6538808392953767,
      "flos": 26833997433600.0,
      "grad_norm": 2.986854732464837,
      "language_loss": 0.8085807,
      "learning_rate": 1.1303280420569982e-06,
      "loss": 0.83064115,
      "num_input_tokens_seen": 117039505,
      "step": 5438,
      "time_per_iteration": 2.6344292163848877
    },
    {
      "auxiliary_loss_clip": 0.01172901,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.04874885,
      "balance_loss_mlp": 1.01854241,
      "epoch": 0.6540010821860157,
      "flos": 30738241301760.0,
      "grad_norm": 1.7312885700952545,
      "language_loss": 0.77368462,
      "learning_rate": 1.1296266365355158e-06,
      "loss": 0.79567909,
      "num_input_tokens_seen": 117062890,
      "step": 5439,
      "time_per_iteration": 2.6975197792053223
    },
    {
      "auxiliary_loss_clip": 0.01173125,
      "auxiliary_loss_mlp": 0.01027257,
      "balance_loss_clip": 1.04851604,
      "balance_loss_mlp": 1.0192939,
      "epoch": 0.6541213250766549,
      "flos": 26907147480960.0,
      "grad_norm": 2.6219022776969547,
      "language_loss": 0.73778886,
      "learning_rate": 1.1289253630489806e-06,
      "loss": 0.75979269,
      "num_input_tokens_seen": 117083940,
      "step": 5440,
      "time_per_iteration": 2.6811885833740234
    },
    {
      "auxiliary_loss_clip": 0.01178786,
      "auxiliary_loss_mlp": 0.01019337,
      "balance_loss_clip": 1.05015576,
      "balance_loss_mlp": 1.0113616,
      "epoch": 0.6542415679672939,
      "flos": 19172384409600.0,
      "grad_norm": 2.175162878995957,
      "language_loss": 0.72438604,
      "learning_rate": 1.1282242217037753e-06,
      "loss": 0.74636722,
      "num_input_tokens_seen": 117101440,
      "step": 5441,
      "time_per_iteration": 2.676939010620117
    },
    {
      "auxiliary_loss_clip": 0.01178501,
      "auxiliary_loss_mlp": 0.01027102,
      "balance_loss_clip": 1.05022418,
      "balance_loss_mlp": 1.01793504,
      "epoch": 0.654361810857933,
      "flos": 48173517100800.0,
      "grad_norm": 2.27340027377834,
      "language_loss": 0.61533237,
      "learning_rate": 1.127523212606262e-06,
      "loss": 0.63738835,
      "num_input_tokens_seen": 117124265,
      "step": 5442,
      "time_per_iteration": 2.8138577938079834
    },
    {
      "auxiliary_loss_clip": 0.01173696,
      "auxiliary_loss_mlp": 0.01027194,
      "balance_loss_clip": 1.04889894,
      "balance_loss_mlp": 1.01909399,
      "epoch": 0.6544820537485722,
      "flos": 26943165843840.0,
      "grad_norm": 2.2421303615626864,
      "language_loss": 0.73471618,
      "learning_rate": 1.1268223358627835e-06,
      "loss": 0.75672507,
      "num_input_tokens_seen": 117146755,
      "step": 5443,
      "time_per_iteration": 2.7227702140808105
    },
    {
      "auxiliary_loss_clip": 0.01173493,
      "auxiliary_loss_mlp": 0.01025822,
      "balance_loss_clip": 1.04817986,
      "balance_loss_mlp": 1.01792789,
      "epoch": 0.6546022966392112,
      "flos": 20886328748160.0,
      "grad_norm": 2.378267639646405,
      "language_loss": 0.71954954,
      "learning_rate": 1.126121591579663e-06,
      "loss": 0.7415427,
      "num_input_tokens_seen": 117165960,
      "step": 5444,
      "time_per_iteration": 2.636916399002075
    },
    {
      "auxiliary_loss_clip": 0.01173175,
      "auxiliary_loss_mlp": 0.01026762,
      "balance_loss_clip": 1.05190241,
      "balance_loss_mlp": 1.01940107,
      "epoch": 0.6547225395298503,
      "flos": 24936693143040.0,
      "grad_norm": 2.8348046358675725,
      "language_loss": 0.69496167,
      "learning_rate": 1.1254209798632018e-06,
      "loss": 0.71696103,
      "num_input_tokens_seen": 117186980,
      "step": 5445,
      "time_per_iteration": 2.7215232849121094
    },
    {
      "auxiliary_loss_clip": 0.01176278,
      "auxiliary_loss_mlp": 0.01027806,
      "balance_loss_clip": 1.05206895,
      "balance_loss_mlp": 1.02015901,
      "epoch": 0.6548427824204894,
      "flos": 22565942663040.0,
      "grad_norm": 2.052321400101288,
      "language_loss": 0.84531307,
      "learning_rate": 1.124720500819683e-06,
      "loss": 0.86735392,
      "num_input_tokens_seen": 117205135,
      "step": 5446,
      "time_per_iteration": 2.623981237411499
    },
    {
      "auxiliary_loss_clip": 0.01174031,
      "auxiliary_loss_mlp": 0.01028182,
      "balance_loss_clip": 1.04937553,
      "balance_loss_mlp": 1.0198369,
      "epoch": 0.6549630253111285,
      "flos": 18442500048000.0,
      "grad_norm": 1.8660701806989777,
      "language_loss": 0.81963313,
      "learning_rate": 1.1240201545553682e-06,
      "loss": 0.84165525,
      "num_input_tokens_seen": 117222935,
      "step": 5447,
      "time_per_iteration": 2.578752279281616
    },
    {
      "auxiliary_loss_clip": 0.01174589,
      "auxiliary_loss_mlp": 0.01021238,
      "balance_loss_clip": 1.05072641,
      "balance_loss_mlp": 1.01300037,
      "epoch": 0.6550832682017675,
      "flos": 25187313312000.0,
      "grad_norm": 1.8960403334642615,
      "language_loss": 0.73514396,
      "learning_rate": 1.1233199411764987e-06,
      "loss": 0.75710219,
      "num_input_tokens_seen": 117242370,
      "step": 5448,
      "time_per_iteration": 3.598907470703125
    },
    {
      "auxiliary_loss_clip": 0.01173235,
      "auxiliary_loss_mlp": 0.01020222,
      "balance_loss_clip": 1.04964304,
      "balance_loss_mlp": 1.01289654,
      "epoch": 0.6552035110924067,
      "flos": 22748153379840.0,
      "grad_norm": 2.175163325453958,
      "language_loss": 0.6909163,
      "learning_rate": 1.1226198607892978e-06,
      "loss": 0.71285093,
      "num_input_tokens_seen": 117262930,
      "step": 5449,
      "time_per_iteration": 2.619903802871704
    },
    {
      "auxiliary_loss_clip": 0.01175577,
      "auxiliary_loss_mlp": 0.01021579,
      "balance_loss_clip": 1.05103052,
      "balance_loss_mlp": 1.01469183,
      "epoch": 0.6553237539830458,
      "flos": 21799178012160.0,
      "grad_norm": 2.4662430059266462,
      "language_loss": 0.79991865,
      "learning_rate": 1.1219199134999664e-06,
      "loss": 0.82189018,
      "num_input_tokens_seen": 117281430,
      "step": 5450,
      "time_per_iteration": 2.678790330886841
    },
    {
      "auxiliary_loss_clip": 0.01176238,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.05046415,
      "balance_loss_mlp": 1.01916826,
      "epoch": 0.6554439968736848,
      "flos": 20887226588160.0,
      "grad_norm": 2.7046428826142264,
      "language_loss": 0.78859293,
      "learning_rate": 1.1212200994146863e-06,
      "loss": 0.81062746,
      "num_input_tokens_seen": 117299185,
      "step": 5451,
      "time_per_iteration": 2.6224639415740967
    },
    {
      "auxiliary_loss_clip": 0.0116927,
      "auxiliary_loss_mlp": 0.0102511,
      "balance_loss_clip": 1.04647541,
      "balance_loss_mlp": 1.01716471,
      "epoch": 0.655564239764324,
      "flos": 16139045698560.0,
      "grad_norm": 2.340820145110017,
      "language_loss": 0.75795698,
      "learning_rate": 1.120520418639618e-06,
      "loss": 0.77990079,
      "num_input_tokens_seen": 117317720,
      "step": 5452,
      "time_per_iteration": 2.6529712677001953
    },
    {
      "auxiliary_loss_clip": 0.01175081,
      "auxiliary_loss_mlp": 0.01026077,
      "balance_loss_clip": 1.05137396,
      "balance_loss_mlp": 1.01845682,
      "epoch": 0.655684482654963,
      "flos": 29570354496000.0,
      "grad_norm": 2.2850829309439162,
      "language_loss": 0.83451331,
      "learning_rate": 1.119820871280903e-06,
      "loss": 0.85652488,
      "num_input_tokens_seen": 117338795,
      "step": 5453,
      "time_per_iteration": 3.73313307762146
    },
    {
      "auxiliary_loss_clip": 0.01174814,
      "auxiliary_loss_mlp": 0.0102699,
      "balance_loss_clip": 1.05114555,
      "balance_loss_mlp": 1.01941454,
      "epoch": 0.6558047255456021,
      "flos": 29789409588480.0,
      "grad_norm": 2.5880781247322338,
      "language_loss": 0.73368382,
      "learning_rate": 1.1191214574446614e-06,
      "loss": 0.7557019,
      "num_input_tokens_seen": 117359040,
      "step": 5454,
      "time_per_iteration": 2.7372584342956543
    },
    {
      "auxiliary_loss_clip": 0.01174068,
      "auxiliary_loss_mlp": 0.01027494,
      "balance_loss_clip": 1.04776382,
      "balance_loss_mlp": 1.01995659,
      "epoch": 0.6559249684362413,
      "flos": 29059166090880.0,
      "grad_norm": 1.6750171612050653,
      "language_loss": 0.80293345,
      "learning_rate": 1.118422177236995e-06,
      "loss": 0.82494909,
      "num_input_tokens_seen": 117380865,
      "step": 5455,
      "time_per_iteration": 3.624546766281128
    },
    {
      "auxiliary_loss_clip": 0.01172732,
      "auxiliary_loss_mlp": 0.0102817,
      "balance_loss_clip": 1.04966021,
      "balance_loss_mlp": 1.01961708,
      "epoch": 0.6560452113268803,
      "flos": 20225464369920.0,
      "grad_norm": 2.0910324112023138,
      "language_loss": 0.85877061,
      "learning_rate": 1.1177230307639835e-06,
      "loss": 0.88077962,
      "num_input_tokens_seen": 117398405,
      "step": 5456,
      "time_per_iteration": 2.625819206237793
    },
    {
      "auxiliary_loss_clip": 0.01175736,
      "auxiliary_loss_mlp": 0.01031943,
      "balance_loss_clip": 1.0508194,
      "balance_loss_mlp": 1.02411079,
      "epoch": 0.6561654542175194,
      "flos": 25045538330880.0,
      "grad_norm": 2.964982466083652,
      "language_loss": 0.78734183,
      "learning_rate": 1.1170240181316865e-06,
      "loss": 0.80941868,
      "num_input_tokens_seen": 117419850,
      "step": 5457,
      "time_per_iteration": 2.870455741882324
    },
    {
      "auxiliary_loss_clip": 0.01173357,
      "auxiliary_loss_mlp": 0.01023693,
      "balance_loss_clip": 1.04785287,
      "balance_loss_mlp": 1.01558137,
      "epoch": 0.6562856971081584,
      "flos": 22856711258880.0,
      "grad_norm": 3.5620976308701096,
      "language_loss": 0.7971859,
      "learning_rate": 1.1163251394461442e-06,
      "loss": 0.81915635,
      "num_input_tokens_seen": 117438330,
      "step": 5458,
      "time_per_iteration": 2.618063449859619
    },
    {
      "auxiliary_loss_clip": 0.01176098,
      "auxiliary_loss_mlp": 0.01029153,
      "balance_loss_clip": 1.05182648,
      "balance_loss_mlp": 1.02089214,
      "epoch": 0.6564059399987976,
      "flos": 18872565586560.0,
      "grad_norm": 2.486687475851736,
      "language_loss": 0.82591861,
      "learning_rate": 1.1156263948133746e-06,
      "loss": 0.8479712,
      "num_input_tokens_seen": 117454985,
      "step": 5459,
      "time_per_iteration": 2.6759941577911377
    },
    {
      "auxiliary_loss_clip": 0.01174908,
      "auxiliary_loss_mlp": 0.0102664,
      "balance_loss_clip": 1.04919875,
      "balance_loss_mlp": 1.01847982,
      "epoch": 0.6565261828894366,
      "flos": 25484187219840.0,
      "grad_norm": 1.6534294427601286,
      "language_loss": 0.77539289,
      "learning_rate": 1.1149277843393787e-06,
      "loss": 0.7974084,
      "num_input_tokens_seen": 117476145,
      "step": 5460,
      "time_per_iteration": 2.7442617416381836
    },
    {
      "auxiliary_loss_clip": 0.01173165,
      "auxiliary_loss_mlp": 0.01029043,
      "balance_loss_clip": 1.04926276,
      "balance_loss_mlp": 1.02109814,
      "epoch": 0.6566464257800757,
      "flos": 19683500987520.0,
      "grad_norm": 2.3053359216894767,
      "language_loss": 0.63279271,
      "learning_rate": 1.1142293081301342e-06,
      "loss": 0.65481484,
      "num_input_tokens_seen": 117494025,
      "step": 5461,
      "time_per_iteration": 2.657526969909668
    },
    {
      "auxiliary_loss_clip": 0.01171445,
      "auxiliary_loss_mlp": 0.01023761,
      "balance_loss_clip": 1.04948616,
      "balance_loss_mlp": 1.01667082,
      "epoch": 0.6567666686707149,
      "flos": 23514127931520.0,
      "grad_norm": 1.7971707536996928,
      "language_loss": 0.68360484,
      "learning_rate": 1.1135309662915995e-06,
      "loss": 0.70555693,
      "num_input_tokens_seen": 117514190,
      "step": 5462,
      "time_per_iteration": 2.8263370990753174
    },
    {
      "auxiliary_loss_clip": 0.01171743,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.04684997,
      "balance_loss_mlp": 1.01727688,
      "epoch": 0.6568869115613539,
      "flos": 32781342896640.0,
      "grad_norm": 2.1992548310536106,
      "language_loss": 0.60710359,
      "learning_rate": 1.112832758929712e-06,
      "loss": 0.62907302,
      "num_input_tokens_seen": 117536800,
      "step": 5463,
      "time_per_iteration": 2.897972345352173
    },
    {
      "auxiliary_loss_clip": 0.01173436,
      "auxiliary_loss_mlp": 0.010261,
      "balance_loss_clip": 1.0479387,
      "balance_loss_mlp": 1.01800561,
      "epoch": 0.657007154451993,
      "flos": 18442428220800.0,
      "grad_norm": 2.164299611039736,
      "language_loss": 0.75013918,
      "learning_rate": 1.11213468615039e-06,
      "loss": 0.77213454,
      "num_input_tokens_seen": 117556230,
      "step": 5464,
      "time_per_iteration": 2.6049811840057373
    },
    {
      "auxiliary_loss_clip": 0.0116993,
      "auxiliary_loss_mlp": 0.01025792,
      "balance_loss_clip": 1.05056238,
      "balance_loss_mlp": 1.01900339,
      "epoch": 0.6571273973426321,
      "flos": 25156717902720.0,
      "grad_norm": 1.8834954203832075,
      "language_loss": 0.75065905,
      "learning_rate": 1.1114367480595292e-06,
      "loss": 0.77261627,
      "num_input_tokens_seen": 117577310,
      "step": 5465,
      "time_per_iteration": 2.6373252868652344
    },
    {
      "auxiliary_loss_clip": 0.01177353,
      "auxiliary_loss_mlp": 0.01022118,
      "balance_loss_clip": 1.04961181,
      "balance_loss_mlp": 1.013726,
      "epoch": 0.6572476402332712,
      "flos": 17529830352000.0,
      "grad_norm": 2.0098233205677496,
      "language_loss": 0.81315106,
      "learning_rate": 1.1107389447630086e-06,
      "loss": 0.83514577,
      "num_input_tokens_seen": 117596010,
      "step": 5466,
      "time_per_iteration": 2.6374568939208984
    },
    {
      "auxiliary_loss_clip": 0.01172205,
      "auxiliary_loss_mlp": 0.01023908,
      "balance_loss_clip": 1.04959607,
      "balance_loss_mlp": 1.01619184,
      "epoch": 0.6573678831239103,
      "flos": 17014260487680.0,
      "grad_norm": 2.544634089528715,
      "language_loss": 0.78317821,
      "learning_rate": 1.1100412763666818e-06,
      "loss": 0.8051393,
      "num_input_tokens_seen": 117611270,
      "step": 5467,
      "time_per_iteration": 2.6043670177459717
    },
    {
      "auxiliary_loss_clip": 0.01176435,
      "auxiliary_loss_mlp": 0.01030649,
      "balance_loss_clip": 1.05154872,
      "balance_loss_mlp": 1.02223897,
      "epoch": 0.6574881260145494,
      "flos": 23910078528000.0,
      "grad_norm": 2.592217955660974,
      "language_loss": 0.80178124,
      "learning_rate": 1.1093437429763865e-06,
      "loss": 0.82385206,
      "num_input_tokens_seen": 117631535,
      "step": 5468,
      "time_per_iteration": 2.644686460494995
    },
    {
      "auxiliary_loss_clip": 0.01171027,
      "auxiliary_loss_mlp": 0.01023883,
      "balance_loss_clip": 1.04935861,
      "balance_loss_mlp": 1.01658475,
      "epoch": 0.6576083689051885,
      "flos": 11218458504960.0,
      "grad_norm": 2.6621801907643508,
      "language_loss": 0.73606789,
      "learning_rate": 1.1086463446979361e-06,
      "loss": 0.758017,
      "num_input_tokens_seen": 117649885,
      "step": 5469,
      "time_per_iteration": 2.651737928390503
    },
    {
      "auxiliary_loss_clip": 0.01180899,
      "auxiliary_loss_mlp": 0.01029691,
      "balance_loss_clip": 1.05546665,
      "balance_loss_mlp": 1.02103591,
      "epoch": 0.6577286117958275,
      "flos": 22455553190400.0,
      "grad_norm": 2.7488734458902098,
      "language_loss": 0.77342165,
      "learning_rate": 1.1079490816371277e-06,
      "loss": 0.79552752,
      "num_input_tokens_seen": 117669650,
      "step": 5470,
      "time_per_iteration": 2.6113009452819824
    },
    {
      "auxiliary_loss_clip": 0.0117259,
      "auxiliary_loss_mlp": 0.01025522,
      "balance_loss_clip": 1.04779017,
      "balance_loss_mlp": 1.018116,
      "epoch": 0.6578488546864667,
      "flos": 21872184405120.0,
      "grad_norm": 2.619201660856418,
      "language_loss": 0.74745923,
      "learning_rate": 1.1072519538997352e-06,
      "loss": 0.76944035,
      "num_input_tokens_seen": 117688790,
      "step": 5471,
      "time_per_iteration": 2.5877721309661865
    },
    {
      "auxiliary_loss_clip": 0.01172302,
      "auxiliary_loss_mlp": 0.01021039,
      "balance_loss_clip": 1.04756951,
      "balance_loss_mlp": 1.0123961,
      "epoch": 0.6579690975771058,
      "flos": 23543753673600.0,
      "grad_norm": 2.012236333908745,
      "language_loss": 0.82343984,
      "learning_rate": 1.1065549615915095e-06,
      "loss": 0.84537327,
      "num_input_tokens_seen": 117708620,
      "step": 5472,
      "time_per_iteration": 2.6243932247161865
    },
    {
      "auxiliary_loss_clip": 0.01175822,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.05237103,
      "balance_loss_mlp": 1.01746774,
      "epoch": 0.6580893404677448,
      "flos": 32743995730560.0,
      "grad_norm": 4.424860601091857,
      "language_loss": 0.7828185,
      "learning_rate": 1.105858104818187e-06,
      "loss": 0.804838,
      "num_input_tokens_seen": 117729775,
      "step": 5473,
      "time_per_iteration": 2.716383457183838
    },
    {
      "auxiliary_loss_clip": 0.01177079,
      "auxiliary_loss_mlp": 0.01025498,
      "balance_loss_clip": 1.04999876,
      "balance_loss_mlp": 1.01731408,
      "epoch": 0.658209583358384,
      "flos": 15888138220800.0,
      "grad_norm": 2.3626085606869487,
      "language_loss": 0.75445563,
      "learning_rate": 1.105161383685478e-06,
      "loss": 0.77648139,
      "num_input_tokens_seen": 117746160,
      "step": 5474,
      "time_per_iteration": 3.5748631954193115
    },
    {
      "auxiliary_loss_clip": 0.01075017,
      "auxiliary_loss_mlp": 0.01002644,
      "balance_loss_clip": 1.01464844,
      "balance_loss_mlp": 1.0013926,
      "epoch": 0.658329826249023,
      "flos": 62695902447360.0,
      "grad_norm": 0.7292715837264343,
      "language_loss": 0.56280386,
      "learning_rate": 1.1044647982990771e-06,
      "loss": 0.58358049,
      "num_input_tokens_seen": 117808045,
      "step": 5475,
      "time_per_iteration": 3.1984200477600098
    },
    {
      "auxiliary_loss_clip": 0.01178259,
      "auxiliary_loss_mlp": 0.01025516,
      "balance_loss_clip": 1.05246341,
      "balance_loss_mlp": 1.01744509,
      "epoch": 0.6584500691396621,
      "flos": 31722624501120.0,
      "grad_norm": 2.6010564822171793,
      "language_loss": 0.6444456,
      "learning_rate": 1.1037683487646536e-06,
      "loss": 0.6664834,
      "num_input_tokens_seen": 117828330,
      "step": 5476,
      "time_per_iteration": 2.6414873600006104
    },
    {
      "auxiliary_loss_clip": 0.01174643,
      "auxiliary_loss_mlp": 0.01027209,
      "balance_loss_clip": 1.05232239,
      "balance_loss_mlp": 1.01913857,
      "epoch": 0.6585703120303013,
      "flos": 18406086635520.0,
      "grad_norm": 1.7787185343208964,
      "language_loss": 0.76785898,
      "learning_rate": 1.1030720351878583e-06,
      "loss": 0.78987753,
      "num_input_tokens_seen": 117846450,
      "step": 5477,
      "time_per_iteration": 2.758305549621582
    },
    {
      "auxiliary_loss_clip": 0.01074457,
      "auxiliary_loss_mlp": 0.01001435,
      "balance_loss_clip": 1.01462555,
      "balance_loss_mlp": 1.00015306,
      "epoch": 0.6586905549209403,
      "flos": 58309880434560.0,
      "grad_norm": 0.8139404506808526,
      "language_loss": 0.57558054,
      "learning_rate": 1.102375857674323e-06,
      "loss": 0.5963394,
      "num_input_tokens_seen": 117908365,
      "step": 5478,
      "time_per_iteration": 3.1930830478668213
    },
    {
      "auxiliary_loss_clip": 0.01173371,
      "auxiliary_loss_mlp": 0.01026515,
      "balance_loss_clip": 1.04999256,
      "balance_loss_mlp": 1.01917791,
      "epoch": 0.6588107978115794,
      "flos": 22782627457920.0,
      "grad_norm": 1.7516880540998447,
      "language_loss": 0.90468985,
      "learning_rate": 1.1016798163296561e-06,
      "loss": 0.92668867,
      "num_input_tokens_seen": 117927565,
      "step": 5479,
      "time_per_iteration": 2.75030517578125
    },
    {
      "auxiliary_loss_clip": 0.01171842,
      "auxiliary_loss_mlp": 0.01022741,
      "balance_loss_clip": 1.04953253,
      "balance_loss_mlp": 1.01515877,
      "epoch": 0.6589310407022185,
      "flos": 20667525050880.0,
      "grad_norm": 1.9732811416561267,
      "language_loss": 0.65997255,
      "learning_rate": 1.1009839112594471e-06,
      "loss": 0.68191838,
      "num_input_tokens_seen": 117945590,
      "step": 5480,
      "time_per_iteration": 4.644946336746216
    },
    {
      "auxiliary_loss_clip": 0.01174774,
      "auxiliary_loss_mlp": 0.01027354,
      "balance_loss_clip": 1.05030549,
      "balance_loss_mlp": 1.01900947,
      "epoch": 0.6590512835928576,
      "flos": 25630595055360.0,
      "grad_norm": 2.569036351442232,
      "language_loss": 0.72339606,
      "learning_rate": 1.1002881425692638e-06,
      "loss": 0.74541736,
      "num_input_tokens_seen": 117966020,
      "step": 5481,
      "time_per_iteration": 3.537109613418579
    },
    {
      "auxiliary_loss_clip": 0.01172711,
      "auxiliary_loss_mlp": 0.01025087,
      "balance_loss_clip": 1.04845476,
      "balance_loss_mlp": 1.01694465,
      "epoch": 0.6591715264834966,
      "flos": 23726108044800.0,
      "grad_norm": 2.2222892299018264,
      "language_loss": 0.75451493,
      "learning_rate": 1.0995925103646532e-06,
      "loss": 0.77649289,
      "num_input_tokens_seen": 117984620,
      "step": 5482,
      "time_per_iteration": 2.6413984298706055
    },
    {
      "auxiliary_loss_clip": 0.01175247,
      "auxiliary_loss_mlp": 0.01028002,
      "balance_loss_clip": 1.0511868,
      "balance_loss_mlp": 1.02050996,
      "epoch": 0.6592917693741358,
      "flos": 35773850822400.0,
      "grad_norm": 1.7135609538896348,
      "language_loss": 0.66964054,
      "learning_rate": 1.0988970147511437e-06,
      "loss": 0.69167304,
      "num_input_tokens_seen": 118006500,
      "step": 5483,
      "time_per_iteration": 2.765021324157715
    },
    {
      "auxiliary_loss_clip": 0.01175222,
      "auxiliary_loss_mlp": 0.01025622,
      "balance_loss_clip": 1.05173993,
      "balance_loss_mlp": 1.01734304,
      "epoch": 0.6594120122647749,
      "flos": 21396834794880.0,
      "grad_norm": 3.8613132964787935,
      "language_loss": 0.80719608,
      "learning_rate": 1.0982016558342405e-06,
      "loss": 0.82920456,
      "num_input_tokens_seen": 118025470,
      "step": 5484,
      "time_per_iteration": 2.628161907196045
    },
    {
      "auxiliary_loss_clip": 0.01175889,
      "auxiliary_loss_mlp": 0.01024794,
      "balance_loss_clip": 1.05120778,
      "balance_loss_mlp": 1.01646757,
      "epoch": 0.6595322551554139,
      "flos": 19351829779200.0,
      "grad_norm": 1.898691026703364,
      "language_loss": 0.70906883,
      "learning_rate": 1.0975064337194291e-06,
      "loss": 0.73107564,
      "num_input_tokens_seen": 118043515,
      "step": 5485,
      "time_per_iteration": 2.6148324012756348
    },
    {
      "auxiliary_loss_clip": 0.01175587,
      "auxiliary_loss_mlp": 0.01026949,
      "balance_loss_clip": 1.05169904,
      "balance_loss_mlp": 1.01898551,
      "epoch": 0.6596524980460531,
      "flos": 16837113588480.0,
      "grad_norm": 1.6362865024174809,
      "language_loss": 0.70406735,
      "learning_rate": 1.0968113485121743e-06,
      "loss": 0.72609282,
      "num_input_tokens_seen": 118063105,
      "step": 5486,
      "time_per_iteration": 2.6281795501708984
    },
    {
      "auxiliary_loss_clip": 0.01174204,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.04941392,
      "balance_loss_mlp": 1.02216196,
      "epoch": 0.6597727409366921,
      "flos": 21798567480960.0,
      "grad_norm": 2.6428690125445824,
      "language_loss": 0.80161023,
      "learning_rate": 1.0961164003179185e-06,
      "loss": 0.82365692,
      "num_input_tokens_seen": 118081615,
      "step": 5487,
      "time_per_iteration": 2.6600232124328613
    },
    {
      "auxiliary_loss_clip": 0.01175924,
      "auxiliary_loss_mlp": 0.01024842,
      "balance_loss_clip": 1.0507443,
      "balance_loss_mlp": 1.01645589,
      "epoch": 0.6598929838273312,
      "flos": 23730704985600.0,
      "grad_norm": 2.832168879005816,
      "language_loss": 0.84259081,
      "learning_rate": 1.0954215892420884e-06,
      "loss": 0.86459845,
      "num_input_tokens_seen": 118102315,
      "step": 5488,
      "time_per_iteration": 2.768961191177368
    },
    {
      "auxiliary_loss_clip": 0.01180364,
      "auxiliary_loss_mlp": 0.01028484,
      "balance_loss_clip": 1.05385852,
      "balance_loss_mlp": 1.02001452,
      "epoch": 0.6600132267179702,
      "flos": 19974520978560.0,
      "grad_norm": 1.8014697798310724,
      "language_loss": 0.70617712,
      "learning_rate": 1.094726915390082e-06,
      "loss": 0.72826564,
      "num_input_tokens_seen": 118120650,
      "step": 5489,
      "time_per_iteration": 2.8766186237335205
    },
    {
      "auxiliary_loss_clip": 0.01175,
      "auxiliary_loss_mlp": 0.01024095,
      "balance_loss_clip": 1.05032158,
      "balance_loss_mlp": 1.01603627,
      "epoch": 0.6601334696086094,
      "flos": 22342649765760.0,
      "grad_norm": 2.1204387259525883,
      "language_loss": 0.69475418,
      "learning_rate": 1.0940323788672836e-06,
      "loss": 0.71674514,
      "num_input_tokens_seen": 118139825,
      "step": 5490,
      "time_per_iteration": 2.7118213176727295
    },
    {
      "auxiliary_loss_clip": 0.01171537,
      "auxiliary_loss_mlp": 0.01020723,
      "balance_loss_clip": 1.04947543,
      "balance_loss_mlp": 1.01342177,
      "epoch": 0.6602537124992485,
      "flos": 25703098657920.0,
      "grad_norm": 1.6923341215280299,
      "language_loss": 0.73777795,
      "learning_rate": 1.0933379797790522e-06,
      "loss": 0.7597006,
      "num_input_tokens_seen": 118159240,
      "step": 5491,
      "time_per_iteration": 2.6951446533203125
    },
    {
      "auxiliary_loss_clip": 0.01172654,
      "auxiliary_loss_mlp": 0.01027328,
      "balance_loss_clip": 1.04862309,
      "balance_loss_mlp": 1.01957977,
      "epoch": 0.6603739553898875,
      "flos": 25848572739840.0,
      "grad_norm": 73.82673083638,
      "language_loss": 0.71513283,
      "learning_rate": 1.0926437182307293e-06,
      "loss": 0.73713273,
      "num_input_tokens_seen": 118178050,
      "step": 5492,
      "time_per_iteration": 2.694972276687622
    },
    {
      "auxiliary_loss_clip": 0.0117419,
      "auxiliary_loss_mlp": 0.01029011,
      "balance_loss_clip": 1.04926527,
      "balance_loss_mlp": 1.02026105,
      "epoch": 0.6604941982805267,
      "flos": 24570296461440.0,
      "grad_norm": 4.78465569531055,
      "language_loss": 0.78072548,
      "learning_rate": 1.0919495943276338e-06,
      "loss": 0.80275744,
      "num_input_tokens_seen": 118199070,
      "step": 5493,
      "time_per_iteration": 2.642343282699585
    },
    {
      "auxiliary_loss_clip": 0.01178734,
      "auxiliary_loss_mlp": 0.01030332,
      "balance_loss_clip": 1.05123353,
      "balance_loss_mlp": 1.02228522,
      "epoch": 0.6606144411711657,
      "flos": 13261775581440.0,
      "grad_norm": 2.7229373420488954,
      "language_loss": 0.76394308,
      "learning_rate": 1.0912556081750611e-06,
      "loss": 0.78603375,
      "num_input_tokens_seen": 118217000,
      "step": 5494,
      "time_per_iteration": 2.5935754776000977
    },
    {
      "auxiliary_loss_clip": 0.01176268,
      "auxiliary_loss_mlp": 0.01024508,
      "balance_loss_clip": 1.05110347,
      "balance_loss_mlp": 1.01655054,
      "epoch": 0.6607346840618048,
      "flos": 25155281358720.0,
      "grad_norm": 2.3602087836509926,
      "language_loss": 0.76595551,
      "learning_rate": 1.0905617598782909e-06,
      "loss": 0.78796327,
      "num_input_tokens_seen": 118237205,
      "step": 5495,
      "time_per_iteration": 2.6185598373413086
    },
    {
      "auxiliary_loss_clip": 0.01174482,
      "auxiliary_loss_mlp": 0.01026021,
      "balance_loss_clip": 1.05117571,
      "balance_loss_mlp": 1.01868105,
      "epoch": 0.660854926952444,
      "flos": 17638029095040.0,
      "grad_norm": 3.5028314505006217,
      "language_loss": 0.81413931,
      "learning_rate": 1.0898680495425775e-06,
      "loss": 0.83614433,
      "num_input_tokens_seen": 118255495,
      "step": 5496,
      "time_per_iteration": 2.6178839206695557
    },
    {
      "auxiliary_loss_clip": 0.01176048,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.05076385,
      "balance_loss_mlp": 1.0172298,
      "epoch": 0.660975169843083,
      "flos": 16836000266880.0,
      "grad_norm": 4.5507720556706746,
      "language_loss": 0.8039552,
      "learning_rate": 1.0891744772731594e-06,
      "loss": 0.82596338,
      "num_input_tokens_seen": 118273310,
      "step": 5497,
      "time_per_iteration": 2.5520763397216797
    },
    {
      "auxiliary_loss_clip": 0.01172474,
      "auxiliary_loss_mlp": 0.01024962,
      "balance_loss_clip": 1.04980361,
      "balance_loss_mlp": 1.0168618,
      "epoch": 0.6610954127337221,
      "flos": 26870410846080.0,
      "grad_norm": 1.7687501950370228,
      "language_loss": 0.662058,
      "learning_rate": 1.088481043175248e-06,
      "loss": 0.68403232,
      "num_input_tokens_seen": 118293880,
      "step": 5498,
      "time_per_iteration": 2.6554174423217773
    },
    {
      "auxiliary_loss_clip": 0.0117254,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.0499022,
      "balance_loss_mlp": 1.01780224,
      "epoch": 0.6612156556243612,
      "flos": 26465697331200.0,
      "grad_norm": 1.7063590816218623,
      "language_loss": 0.7568469,
      "learning_rate": 1.0877877473540368e-06,
      "loss": 0.77882624,
      "num_input_tokens_seen": 118314465,
      "step": 5499,
      "time_per_iteration": 2.638768196105957
    },
    {
      "auxiliary_loss_clip": 0.01176775,
      "auxiliary_loss_mlp": 0.01026837,
      "balance_loss_clip": 1.05017233,
      "balance_loss_mlp": 1.01875496,
      "epoch": 0.6613358985150003,
      "flos": 19791915212160.0,
      "grad_norm": 1.8342598660941138,
      "language_loss": 0.72615278,
      "learning_rate": 1.0870945899147002e-06,
      "loss": 0.74818891,
      "num_input_tokens_seen": 118331110,
      "step": 5500,
      "time_per_iteration": 3.6745309829711914
    },
    {
      "auxiliary_loss_clip": 0.01172472,
      "auxiliary_loss_mlp": 0.0102412,
      "balance_loss_clip": 1.05007911,
      "balance_loss_mlp": 1.01635337,
      "epoch": 0.6614561414056394,
      "flos": 26831627136000.0,
      "grad_norm": 3.6628219717087878,
      "language_loss": 0.7628687,
      "learning_rate": 1.0864015709623879e-06,
      "loss": 0.78483462,
      "num_input_tokens_seen": 118351980,
      "step": 5501,
      "time_per_iteration": 2.9293556213378906
    },
    {
      "auxiliary_loss_clip": 0.01173506,
      "auxiliary_loss_mlp": 0.01021014,
      "balance_loss_clip": 1.04986012,
      "balance_loss_mlp": 1.01306844,
      "epoch": 0.6615763842962785,
      "flos": 22894597128960.0,
      "grad_norm": 9.49963957038067,
      "language_loss": 0.80018389,
      "learning_rate": 1.0857086906022313e-06,
      "loss": 0.82212913,
      "num_input_tokens_seen": 118370315,
      "step": 5502,
      "time_per_iteration": 2.599094867706299
    },
    {
      "auxiliary_loss_clip": 0.01172631,
      "auxiliary_loss_mlp": 0.01025454,
      "balance_loss_clip": 1.04888415,
      "balance_loss_mlp": 1.01706743,
      "epoch": 0.6616966271869176,
      "flos": 24790321221120.0,
      "grad_norm": 1.9564996255644183,
      "language_loss": 0.7296958,
      "learning_rate": 1.0850159489393388e-06,
      "loss": 0.75167662,
      "num_input_tokens_seen": 118389575,
      "step": 5503,
      "time_per_iteration": 2.713484287261963
    },
    {
      "auxiliary_loss_clip": 0.01172183,
      "auxiliary_loss_mlp": 0.01025357,
      "balance_loss_clip": 1.04853642,
      "balance_loss_mlp": 1.0178318,
      "epoch": 0.6618168700775566,
      "flos": 17202109639680.0,
      "grad_norm": 1.8648280495270322,
      "language_loss": 0.82391,
      "learning_rate": 1.0843233460787992e-06,
      "loss": 0.84588534,
      "num_input_tokens_seen": 118406790,
      "step": 5504,
      "time_per_iteration": 2.5686659812927246
    },
    {
      "auxiliary_loss_clip": 0.01174548,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.05228949,
      "balance_loss_mlp": 1.0220629,
      "epoch": 0.6619371129681958,
      "flos": 25447091448960.0,
      "grad_norm": 1.8043762090385014,
      "language_loss": 0.77800989,
      "learning_rate": 1.0836308821256805e-06,
      "loss": 0.8000468,
      "num_input_tokens_seen": 118427590,
      "step": 5505,
      "time_per_iteration": 2.6482977867126465
    },
    {
      "auxiliary_loss_clip": 0.01172988,
      "auxiliary_loss_mlp": 0.01021935,
      "balance_loss_clip": 1.05046248,
      "balance_loss_mlp": 1.01474667,
      "epoch": 0.6620573558588349,
      "flos": 18040444139520.0,
      "grad_norm": 2.313214311611348,
      "language_loss": 0.78303337,
      "learning_rate": 1.0829385571850282e-06,
      "loss": 0.80498254,
      "num_input_tokens_seen": 118444570,
      "step": 5506,
      "time_per_iteration": 2.5985264778137207
    },
    {
      "auxiliary_loss_clip": 0.01180005,
      "auxiliary_loss_mlp": 0.01030752,
      "balance_loss_clip": 1.05191648,
      "balance_loss_mlp": 1.02210879,
      "epoch": 0.6621775987494739,
      "flos": 17785586165760.0,
      "grad_norm": 3.805772717495438,
      "language_loss": 0.83582449,
      "learning_rate": 1.0822463713618679e-06,
      "loss": 0.85793209,
      "num_input_tokens_seen": 118461425,
      "step": 5507,
      "time_per_iteration": 4.394238233566284
    },
    {
      "auxiliary_loss_clip": 0.01174372,
      "auxiliary_loss_mlp": 0.01025413,
      "balance_loss_clip": 1.05033386,
      "balance_loss_mlp": 1.01776814,
      "epoch": 0.6622978416401131,
      "flos": 17492590926720.0,
      "grad_norm": 2.4352348256853227,
      "language_loss": 0.85048711,
      "learning_rate": 1.0815543247612034e-06,
      "loss": 0.87248492,
      "num_input_tokens_seen": 118478495,
      "step": 5508,
      "time_per_iteration": 3.5855295658111572
    },
    {
      "auxiliary_loss_clip": 0.01173371,
      "auxiliary_loss_mlp": 0.01022786,
      "balance_loss_clip": 1.04964721,
      "balance_loss_mlp": 1.01488209,
      "epoch": 0.6624180845307521,
      "flos": 21648352803840.0,
      "grad_norm": 2.5616330769725058,
      "language_loss": 0.83411205,
      "learning_rate": 1.0808624174880168e-06,
      "loss": 0.85607356,
      "num_input_tokens_seen": 118499145,
      "step": 5509,
      "time_per_iteration": 2.653473138809204
    },
    {
      "auxiliary_loss_clip": 0.01173215,
      "auxiliary_loss_mlp": 0.01029677,
      "balance_loss_clip": 1.05120707,
      "balance_loss_mlp": 1.02242303,
      "epoch": 0.6625383274213912,
      "flos": 23805902108160.0,
      "grad_norm": 2.594071628224776,
      "language_loss": 0.80051261,
      "learning_rate": 1.080170649647272e-06,
      "loss": 0.82254153,
      "num_input_tokens_seen": 118518950,
      "step": 5510,
      "time_per_iteration": 2.7052996158599854
    },
    {
      "auxiliary_loss_clip": 0.01173512,
      "auxiliary_loss_mlp": 0.01026108,
      "balance_loss_clip": 1.05077147,
      "balance_loss_mlp": 1.01776946,
      "epoch": 0.6626585703120303,
      "flos": 33262941473280.0,
      "grad_norm": 1.6164226879217762,
      "language_loss": 0.67202234,
      "learning_rate": 1.0794790213439068e-06,
      "loss": 0.69401854,
      "num_input_tokens_seen": 118545850,
      "step": 5511,
      "time_per_iteration": 2.8361659049987793
    },
    {
      "auxiliary_loss_clip": 0.0117898,
      "auxiliary_loss_mlp": 0.01023985,
      "balance_loss_clip": 1.05265367,
      "balance_loss_mlp": 1.01527143,
      "epoch": 0.6627788132026694,
      "flos": 22085780630400.0,
      "grad_norm": 2.076899875884723,
      "language_loss": 0.78588438,
      "learning_rate": 1.078787532682843e-06,
      "loss": 0.80791408,
      "num_input_tokens_seen": 118563325,
      "step": 5512,
      "time_per_iteration": 2.6014256477355957
    },
    {
      "auxiliary_loss_clip": 0.01172594,
      "auxiliary_loss_mlp": 0.01026633,
      "balance_loss_clip": 1.04879248,
      "balance_loss_mlp": 1.01880097,
      "epoch": 0.6628990560933085,
      "flos": 36173608260480.0,
      "grad_norm": 2.2280048310270537,
      "language_loss": 0.75671512,
      "learning_rate": 1.0780961837689773e-06,
      "loss": 0.77870739,
      "num_input_tokens_seen": 118582835,
      "step": 5513,
      "time_per_iteration": 2.7703585624694824
    },
    {
      "auxiliary_loss_clip": 0.01175335,
      "auxiliary_loss_mlp": 0.01021662,
      "balance_loss_clip": 1.052055,
      "balance_loss_mlp": 1.01454544,
      "epoch": 0.6630192989839476,
      "flos": 18513567106560.0,
      "grad_norm": 1.548105079749875,
      "language_loss": 0.69934332,
      "learning_rate": 1.0774049747071883e-06,
      "loss": 0.7213133,
      "num_input_tokens_seen": 118600715,
      "step": 5514,
      "time_per_iteration": 2.6353116035461426
    },
    {
      "auxiliary_loss_clip": 0.01175392,
      "auxiliary_loss_mlp": 0.01030829,
      "balance_loss_clip": 1.05265868,
      "balance_loss_mlp": 1.0229311,
      "epoch": 0.6631395418745867,
      "flos": 35809510049280.0,
      "grad_norm": 1.8138666890366584,
      "language_loss": 0.68243778,
      "learning_rate": 1.076713905602332e-06,
      "loss": 0.70449996,
      "num_input_tokens_seen": 118621290,
      "step": 5515,
      "time_per_iteration": 2.81588077545166
    },
    {
      "auxiliary_loss_clip": 0.01177529,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.05304742,
      "balance_loss_mlp": 1.0168153,
      "epoch": 0.6632597847652257,
      "flos": 20047742853120.0,
      "grad_norm": 1.8000105800997839,
      "language_loss": 0.81378907,
      "learning_rate": 1.07602297655924e-06,
      "loss": 0.8358103,
      "num_input_tokens_seen": 118639610,
      "step": 5516,
      "time_per_iteration": 2.6373071670532227
    },
    {
      "auxiliary_loss_clip": 0.01173825,
      "auxiliary_loss_mlp": 0.01024362,
      "balance_loss_clip": 1.0512104,
      "balance_loss_mlp": 1.01699519,
      "epoch": 0.6633800276558649,
      "flos": 21214480423680.0,
      "grad_norm": 1.9113977787463878,
      "language_loss": 0.80988532,
      "learning_rate": 1.0753321876827292e-06,
      "loss": 0.83186716,
      "num_input_tokens_seen": 118658895,
      "step": 5517,
      "time_per_iteration": 2.632366418838501
    },
    {
      "auxiliary_loss_clip": 0.01172225,
      "auxiliary_loss_mlp": 0.01026519,
      "balance_loss_clip": 1.04799807,
      "balance_loss_mlp": 1.01848483,
      "epoch": 0.663500270546504,
      "flos": 23987753688960.0,
      "grad_norm": 2.3565975561345063,
      "language_loss": 0.73745012,
      "learning_rate": 1.0746415390775893e-06,
      "loss": 0.75943756,
      "num_input_tokens_seen": 118677025,
      "step": 5518,
      "time_per_iteration": 2.8257577419281006
    },
    {
      "auxiliary_loss_clip": 0.01175973,
      "auxiliary_loss_mlp": 0.01026547,
      "balance_loss_clip": 1.05290985,
      "balance_loss_mlp": 1.01904309,
      "epoch": 0.663620513437143,
      "flos": 17932389050880.0,
      "grad_norm": 2.141578976694863,
      "language_loss": 0.76703441,
      "learning_rate": 1.0739510308485939e-06,
      "loss": 0.78905958,
      "num_input_tokens_seen": 118694240,
      "step": 5519,
      "time_per_iteration": 2.6934759616851807
    },
    {
      "auxiliary_loss_clip": 0.01073744,
      "auxiliary_loss_mlp": 0.01002143,
      "balance_loss_clip": 1.0144465,
      "balance_loss_mlp": 1.00087345,
      "epoch": 0.6637407563277821,
      "flos": 57840241086720.0,
      "grad_norm": 0.8172662890700659,
      "language_loss": 0.62488341,
      "learning_rate": 1.07326066310049e-06,
      "loss": 0.64564228,
      "num_input_tokens_seen": 118758365,
      "step": 5520,
      "time_per_iteration": 3.252656936645508
    },
    {
      "auxiliary_loss_clip": 0.01176958,
      "auxiliary_loss_mlp": 0.01028761,
      "balance_loss_clip": 1.05130434,
      "balance_loss_mlp": 1.02015388,
      "epoch": 0.6638609992184212,
      "flos": 27306007079040.0,
      "grad_norm": 1.9802701766989281,
      "language_loss": 0.79605734,
      "learning_rate": 1.0725704359380059e-06,
      "loss": 0.81811458,
      "num_input_tokens_seen": 118778220,
      "step": 5521,
      "time_per_iteration": 2.7025623321533203
    },
    {
      "auxiliary_loss_clip": 0.01173314,
      "auxiliary_loss_mlp": 0.0103218,
      "balance_loss_clip": 1.04923832,
      "balance_loss_mlp": 1.02452064,
      "epoch": 0.6639812421090603,
      "flos": 18624854419200.0,
      "grad_norm": 2.1602133326595014,
      "language_loss": 0.72184253,
      "learning_rate": 1.0718803494658497e-06,
      "loss": 0.74389744,
      "num_input_tokens_seen": 118797110,
      "step": 5522,
      "time_per_iteration": 2.795893430709839
    },
    {
      "auxiliary_loss_clip": 0.01175778,
      "auxiliary_loss_mlp": 0.01031396,
      "balance_loss_clip": 1.05146289,
      "balance_loss_mlp": 1.0235815,
      "epoch": 0.6641014849996993,
      "flos": 15924479806080.0,
      "grad_norm": 2.5224354827109257,
      "language_loss": 0.83705533,
      "learning_rate": 1.071190403788707e-06,
      "loss": 0.85912704,
      "num_input_tokens_seen": 118812415,
      "step": 5523,
      "time_per_iteration": 2.6135644912719727
    },
    {
      "auxiliary_loss_clip": 0.01176924,
      "auxiliary_loss_mlp": 0.01023467,
      "balance_loss_clip": 1.05088115,
      "balance_loss_mlp": 1.01529539,
      "epoch": 0.6642217278903385,
      "flos": 26505486622080.0,
      "grad_norm": 2.023612507717949,
      "language_loss": 0.75429225,
      "learning_rate": 1.0705005990112415e-06,
      "loss": 0.77629614,
      "num_input_tokens_seen": 118832195,
      "step": 5524,
      "time_per_iteration": 2.87309193611145
    },
    {
      "auxiliary_loss_clip": 0.01176398,
      "auxiliary_loss_mlp": 0.01022994,
      "balance_loss_clip": 1.05230439,
      "balance_loss_mlp": 1.01501334,
      "epoch": 0.6643419707809776,
      "flos": 15377308951680.0,
      "grad_norm": 2.4421914413147014,
      "language_loss": 0.74564546,
      "learning_rate": 1.0698109352380957e-06,
      "loss": 0.7676394,
      "num_input_tokens_seen": 118849795,
      "step": 5525,
      "time_per_iteration": 2.6971383094787598
    },
    {
      "auxiliary_loss_clip": 0.01173905,
      "auxiliary_loss_mlp": 0.01024631,
      "balance_loss_clip": 1.05098438,
      "balance_loss_mlp": 1.0173738,
      "epoch": 0.6644622136716166,
      "flos": 25117610970240.0,
      "grad_norm": 2.568929982366304,
      "language_loss": 0.78388143,
      "learning_rate": 1.0691214125738909e-06,
      "loss": 0.80586672,
      "num_input_tokens_seen": 118870000,
      "step": 5526,
      "time_per_iteration": 3.6323060989379883
    },
    {
      "auxiliary_loss_clip": 0.0107244,
      "auxiliary_loss_mlp": 0.0100125,
      "balance_loss_clip": 1.01353717,
      "balance_loss_mlp": 1.0000813,
      "epoch": 0.6645824565622558,
      "flos": 66201717680640.0,
      "grad_norm": 0.8037680674259784,
      "language_loss": 0.57497752,
      "learning_rate": 1.0684320311232287e-06,
      "loss": 0.59571445,
      "num_input_tokens_seen": 118932905,
      "step": 5527,
      "time_per_iteration": 3.2871713638305664
    },
    {
      "auxiliary_loss_clip": 0.01176656,
      "auxiliary_loss_mlp": 0.01028731,
      "balance_loss_clip": 1.0500083,
      "balance_loss_mlp": 1.02038038,
      "epoch": 0.6647026994528948,
      "flos": 25082131311360.0,
      "grad_norm": 1.7822020393786389,
      "language_loss": 0.81764859,
      "learning_rate": 1.0677427909906865e-06,
      "loss": 0.83970249,
      "num_input_tokens_seen": 118953355,
      "step": 5528,
      "time_per_iteration": 2.6563851833343506
    },
    {
      "auxiliary_loss_clip": 0.01178852,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.05201483,
      "balance_loss_mlp": 1.01709592,
      "epoch": 0.6648229423435339,
      "flos": 18222187979520.0,
      "grad_norm": 2.055995500763022,
      "language_loss": 0.71894091,
      "learning_rate": 1.0670536922808216e-06,
      "loss": 0.74097949,
      "num_input_tokens_seen": 118973480,
      "step": 5529,
      "time_per_iteration": 2.6624209880828857
    },
    {
      "auxiliary_loss_clip": 0.01173123,
      "auxiliary_loss_mlp": 0.01025322,
      "balance_loss_clip": 1.04967546,
      "balance_loss_mlp": 1.01748943,
      "epoch": 0.6649431852341731,
      "flos": 18296882311680.0,
      "grad_norm": 2.2220246324634,
      "language_loss": 0.72143555,
      "learning_rate": 1.06636473509817e-06,
      "loss": 0.74342006,
      "num_input_tokens_seen": 118989860,
      "step": 5530,
      "time_per_iteration": 2.6143085956573486
    },
    {
      "auxiliary_loss_clip": 0.01172338,
      "auxiliary_loss_mlp": 0.01024676,
      "balance_loss_clip": 1.04959249,
      "balance_loss_mlp": 1.01629567,
      "epoch": 0.6650634281248121,
      "flos": 17019575700480.0,
      "grad_norm": 2.047362300139619,
      "language_loss": 0.81050539,
      "learning_rate": 1.0656759195472447e-06,
      "loss": 0.83247554,
      "num_input_tokens_seen": 119007150,
      "step": 5531,
      "time_per_iteration": 2.6591460704803467
    },
    {
      "auxiliary_loss_clip": 0.01072029,
      "auxiliary_loss_mlp": 0.01001051,
      "balance_loss_clip": 1.01315045,
      "balance_loss_mlp": 0.99991858,
      "epoch": 0.6651836710154512,
      "flos": 69294810666240.0,
      "grad_norm": 0.7828289820638081,
      "language_loss": 0.59705478,
      "learning_rate": 1.0649872457325414e-06,
      "loss": 0.61778563,
      "num_input_tokens_seen": 119068435,
      "step": 5532,
      "time_per_iteration": 3.1827125549316406
    },
    {
      "auxiliary_loss_clip": 0.0107184,
      "auxiliary_loss_mlp": 0.01000687,
      "balance_loss_clip": 1.0129106,
      "balance_loss_mlp": 0.9995786,
      "epoch": 0.6653039139060903,
      "flos": 66883444882560.0,
      "grad_norm": 0.8491586470735354,
      "language_loss": 0.55103946,
      "learning_rate": 1.0642987137585278e-06,
      "loss": 0.57176471,
      "num_input_tokens_seen": 119127960,
      "step": 5533,
      "time_per_iteration": 4.110077857971191
    },
    {
      "auxiliary_loss_clip": 0.01174292,
      "auxiliary_loss_mlp": 0.01024664,
      "balance_loss_clip": 1.05049062,
      "balance_loss_mlp": 1.01700521,
      "epoch": 0.6654241567967294,
      "flos": 21470056669440.0,
      "grad_norm": 1.9639206646333305,
      "language_loss": 0.82512194,
      "learning_rate": 1.0636103237296561e-06,
      "loss": 0.84711152,
      "num_input_tokens_seen": 119146885,
      "step": 5534,
      "time_per_iteration": 3.626516580581665
    },
    {
      "auxiliary_loss_clip": 0.01172683,
      "auxiliary_loss_mlp": 0.01024923,
      "balance_loss_clip": 1.05193591,
      "balance_loss_mlp": 1.01717472,
      "epoch": 0.6655443996873684,
      "flos": 25119514391040.0,
      "grad_norm": 1.90106293578772,
      "language_loss": 0.84334141,
      "learning_rate": 1.062922075750353e-06,
      "loss": 0.86531746,
      "num_input_tokens_seen": 119166900,
      "step": 5535,
      "time_per_iteration": 3.612567186355591
    },
    {
      "auxiliary_loss_clip": 0.01171421,
      "auxiliary_loss_mlp": 0.01021121,
      "balance_loss_clip": 1.04905653,
      "balance_loss_mlp": 1.01382256,
      "epoch": 0.6656646425780076,
      "flos": 17457326749440.0,
      "grad_norm": 2.142544080718923,
      "language_loss": 0.71821404,
      "learning_rate": 1.0622339699250267e-06,
      "loss": 0.74013937,
      "num_input_tokens_seen": 119184820,
      "step": 5536,
      "time_per_iteration": 2.6391937732696533
    },
    {
      "auxiliary_loss_clip": 0.01174932,
      "auxiliary_loss_mlp": 0.01026496,
      "balance_loss_clip": 1.05101871,
      "balance_loss_mlp": 1.01848507,
      "epoch": 0.6657848854686467,
      "flos": 23434190213760.0,
      "grad_norm": 1.6334078904747833,
      "language_loss": 0.79228771,
      "learning_rate": 1.0615460063580624e-06,
      "loss": 0.81430203,
      "num_input_tokens_seen": 119203295,
      "step": 5537,
      "time_per_iteration": 2.7058327198028564
    },
    {
      "auxiliary_loss_clip": 0.0117547,
      "auxiliary_loss_mlp": 0.01022324,
      "balance_loss_clip": 1.05144191,
      "balance_loss_mlp": 1.01490343,
      "epoch": 0.6659051283592857,
      "flos": 11509909459200.0,
      "grad_norm": 1.7671712186306154,
      "language_loss": 0.73005033,
      "learning_rate": 1.060858185153821e-06,
      "loss": 0.75202823,
      "num_input_tokens_seen": 119221395,
      "step": 5538,
      "time_per_iteration": 2.8177402019500732
    },
    {
      "auxiliary_loss_clip": 0.01177931,
      "auxiliary_loss_mlp": 0.01031965,
      "balance_loss_clip": 1.05109143,
      "balance_loss_mlp": 1.02269626,
      "epoch": 0.6660253712499249,
      "flos": 20594554571520.0,
      "grad_norm": 3.299942381299755,
      "language_loss": 0.76412332,
      "learning_rate": 1.0601705064166474e-06,
      "loss": 0.78622228,
      "num_input_tokens_seen": 119239790,
      "step": 5539,
      "time_per_iteration": 2.843419313430786
    },
    {
      "auxiliary_loss_clip": 0.01172294,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04937267,
      "balance_loss_mlp": 1.01861882,
      "epoch": 0.666145614140564,
      "flos": 21251504367360.0,
      "grad_norm": 2.1057515730181726,
      "language_loss": 0.7329613,
      "learning_rate": 1.0594829702508596e-06,
      "loss": 0.75494277,
      "num_input_tokens_seen": 119257505,
      "step": 5540,
      "time_per_iteration": 2.7069509029388428
    },
    {
      "auxiliary_loss_clip": 0.01173495,
      "auxiliary_loss_mlp": 0.01024395,
      "balance_loss_clip": 1.05090988,
      "balance_loss_mlp": 1.0163517,
      "epoch": 0.666265857031203,
      "flos": 33726188200320.0,
      "grad_norm": 4.417298644906072,
      "language_loss": 0.54954231,
      "learning_rate": 1.0587955767607592e-06,
      "loss": 0.57152116,
      "num_input_tokens_seen": 119279365,
      "step": 5541,
      "time_per_iteration": 2.756624460220337
    },
    {
      "auxiliary_loss_clip": 0.01173061,
      "auxiliary_loss_mlp": 0.010267,
      "balance_loss_clip": 1.04840958,
      "balance_loss_mlp": 1.01809335,
      "epoch": 0.6663860999218422,
      "flos": 17456644391040.0,
      "grad_norm": 2.191998360530667,
      "language_loss": 0.77221048,
      "learning_rate": 1.0581083260506206e-06,
      "loss": 0.79420811,
      "num_input_tokens_seen": 119296150,
      "step": 5542,
      "time_per_iteration": 2.7549514770507812
    },
    {
      "auxiliary_loss_clip": 0.0117319,
      "auxiliary_loss_mlp": 0.01027436,
      "balance_loss_clip": 1.0494715,
      "balance_loss_mlp": 1.02000642,
      "epoch": 0.6665063428124812,
      "flos": 17676740977920.0,
      "grad_norm": 2.605310062299953,
      "language_loss": 0.76768619,
      "learning_rate": 1.0574212182246993e-06,
      "loss": 0.7896924,
      "num_input_tokens_seen": 119314845,
      "step": 5543,
      "time_per_iteration": 2.647678852081299
    },
    {
      "auxiliary_loss_clip": 0.01176323,
      "auxiliary_loss_mlp": 0.01025028,
      "balance_loss_clip": 1.04965007,
      "balance_loss_mlp": 1.01687384,
      "epoch": 0.6666265857031203,
      "flos": 27673265687040.0,
      "grad_norm": 2.6599810552469765,
      "language_loss": 0.75662255,
      "learning_rate": 1.0567342533872303e-06,
      "loss": 0.77863598,
      "num_input_tokens_seen": 119334875,
      "step": 5544,
      "time_per_iteration": 2.726754903793335
    },
    {
      "auxiliary_loss_clip": 0.01176449,
      "auxiliary_loss_mlp": 0.01030206,
      "balance_loss_clip": 1.0513587,
      "balance_loss_mlp": 1.02187371,
      "epoch": 0.6667468285937594,
      "flos": 25046831220480.0,
      "grad_norm": 1.8267173985509013,
      "language_loss": 0.81156921,
      "learning_rate": 1.0560474316424255e-06,
      "loss": 0.83363569,
      "num_input_tokens_seen": 119354635,
      "step": 5545,
      "time_per_iteration": 2.662764072418213
    },
    {
      "auxiliary_loss_clip": 0.01176018,
      "auxiliary_loss_mlp": 0.01028349,
      "balance_loss_clip": 1.04992139,
      "balance_loss_mlp": 1.01973605,
      "epoch": 0.6668670714843985,
      "flos": 22780472641920.0,
      "grad_norm": 2.2421848012150747,
      "language_loss": 0.7384451,
      "learning_rate": 1.0553607530944746e-06,
      "loss": 0.76048875,
      "num_input_tokens_seen": 119372690,
      "step": 5546,
      "time_per_iteration": 2.741546630859375
    },
    {
      "auxiliary_loss_clip": 0.01174819,
      "auxiliary_loss_mlp": 0.0102884,
      "balance_loss_clip": 1.0517633,
      "balance_loss_mlp": 1.02052546,
      "epoch": 0.6669873143750376,
      "flos": 22163886754560.0,
      "grad_norm": 2.671507740347378,
      "language_loss": 0.90233231,
      "learning_rate": 1.0546742178475463e-06,
      "loss": 0.92436892,
      "num_input_tokens_seen": 119391685,
      "step": 5547,
      "time_per_iteration": 2.68068790435791
    },
    {
      "auxiliary_loss_clip": 0.01171098,
      "auxiliary_loss_mlp": 0.01028321,
      "balance_loss_clip": 1.05036831,
      "balance_loss_mlp": 1.02084684,
      "epoch": 0.6671075572656767,
      "flos": 20514832335360.0,
      "grad_norm": 6.260675561189782,
      "language_loss": 0.86622512,
      "learning_rate": 1.0539878260057868e-06,
      "loss": 0.8882193,
      "num_input_tokens_seen": 119410725,
      "step": 5548,
      "time_per_iteration": 2.6891989707946777
    },
    {
      "auxiliary_loss_clip": 0.01180665,
      "auxiliary_loss_mlp": 0.01032009,
      "balance_loss_clip": 1.05237389,
      "balance_loss_mlp": 1.02354503,
      "epoch": 0.6672278001563158,
      "flos": 17931203902080.0,
      "grad_norm": 2.5670968313075218,
      "language_loss": 0.6831724,
      "learning_rate": 1.0533015776733226e-06,
      "loss": 0.70529914,
      "num_input_tokens_seen": 119426875,
      "step": 5549,
      "time_per_iteration": 2.58221435546875
    },
    {
      "auxiliary_loss_clip": 0.01173348,
      "auxiliary_loss_mlp": 0.01029949,
      "balance_loss_clip": 1.04975617,
      "balance_loss_mlp": 1.02196229,
      "epoch": 0.6673480430469548,
      "flos": 22342146975360.0,
      "grad_norm": 2.161932188218604,
      "language_loss": 0.78919327,
      "learning_rate": 1.0526154729542566e-06,
      "loss": 0.81122631,
      "num_input_tokens_seen": 119446935,
      "step": 5550,
      "time_per_iteration": 2.709115982055664
    },
    {
      "auxiliary_loss_clip": 0.01177055,
      "auxiliary_loss_mlp": 0.01025744,
      "balance_loss_clip": 1.05203462,
      "balance_loss_mlp": 1.01706517,
      "epoch": 0.6674682859375939,
      "flos": 20703830722560.0,
      "grad_norm": 2.4144869608665,
      "language_loss": 0.80017704,
      "learning_rate": 1.0519295119526699e-06,
      "loss": 0.82220507,
      "num_input_tokens_seen": 119463240,
      "step": 5551,
      "time_per_iteration": 2.627958297729492
    },
    {
      "auxiliary_loss_clip": 0.01175928,
      "auxiliary_loss_mlp": 0.01031232,
      "balance_loss_clip": 1.05055881,
      "balance_loss_mlp": 1.02360225,
      "epoch": 0.667588528828233,
      "flos": 26206673379840.0,
      "grad_norm": 1.7236943661171389,
      "language_loss": 0.83378077,
      "learning_rate": 1.0512436947726227e-06,
      "loss": 0.85585237,
      "num_input_tokens_seen": 119484655,
      "step": 5552,
      "time_per_iteration": 2.6828866004943848
    },
    {
      "auxiliary_loss_clip": 0.0117617,
      "auxiliary_loss_mlp": 0.01030403,
      "balance_loss_clip": 1.04982615,
      "balance_loss_mlp": 1.02205825,
      "epoch": 0.6677087717188721,
      "flos": 23071025756160.0,
      "grad_norm": 2.880076252030189,
      "language_loss": 0.65246332,
      "learning_rate": 1.0505580215181517e-06,
      "loss": 0.67452908,
      "num_input_tokens_seen": 119502895,
      "step": 5553,
      "time_per_iteration": 3.61603045463562
    },
    {
      "auxiliary_loss_clip": 0.01071483,
      "auxiliary_loss_mlp": 0.01001368,
      "balance_loss_clip": 1.01298332,
      "balance_loss_mlp": 1.00025296,
      "epoch": 0.6678290146095112,
      "flos": 70941315219840.0,
      "grad_norm": 0.7843495645987285,
      "language_loss": 0.56578571,
      "learning_rate": 1.0498724922932753e-06,
      "loss": 0.58651423,
      "num_input_tokens_seen": 119561010,
      "step": 5554,
      "time_per_iteration": 3.134091854095459
    },
    {
      "auxiliary_loss_clip": 0.01181642,
      "auxiliary_loss_mlp": 0.01026522,
      "balance_loss_clip": 1.0536468,
      "balance_loss_mlp": 1.01772749,
      "epoch": 0.6679492575001503,
      "flos": 18661088263680.0,
      "grad_norm": 2.704884048937454,
      "language_loss": 0.86587721,
      "learning_rate": 1.0491871072019851e-06,
      "loss": 0.88795888,
      "num_input_tokens_seen": 119578900,
      "step": 5555,
      "time_per_iteration": 2.686243772506714
    },
    {
      "auxiliary_loss_clip": 0.01175989,
      "auxiliary_loss_mlp": 0.01024863,
      "balance_loss_clip": 1.04981303,
      "balance_loss_mlp": 1.01715076,
      "epoch": 0.6680695003907894,
      "flos": 29711985822720.0,
      "grad_norm": 1.802161136891653,
      "language_loss": 0.64231932,
      "learning_rate": 1.0485018663482555e-06,
      "loss": 0.66432786,
      "num_input_tokens_seen": 119598920,
      "step": 5556,
      "time_per_iteration": 2.6777656078338623
    },
    {
      "auxiliary_loss_clip": 0.01177723,
      "auxiliary_loss_mlp": 0.01026516,
      "balance_loss_clip": 1.05355465,
      "balance_loss_mlp": 1.01784992,
      "epoch": 0.6681897432814284,
      "flos": 28218964083840.0,
      "grad_norm": 2.5682995588058275,
      "language_loss": 0.70866454,
      "learning_rate": 1.0478167698360354e-06,
      "loss": 0.73070693,
      "num_input_tokens_seen": 119618220,
      "step": 5557,
      "time_per_iteration": 2.6305391788482666
    },
    {
      "auxiliary_loss_clip": 0.01173149,
      "auxiliary_loss_mlp": 0.01024004,
      "balance_loss_clip": 1.04988623,
      "balance_loss_mlp": 1.01632071,
      "epoch": 0.6683099861720676,
      "flos": 25046543911680.0,
      "grad_norm": 1.9833972085417326,
      "language_loss": 0.70185256,
      "learning_rate": 1.0471318177692556e-06,
      "loss": 0.72382408,
      "num_input_tokens_seen": 119638520,
      "step": 5558,
      "time_per_iteration": 2.6667227745056152
    },
    {
      "auxiliary_loss_clip": 0.01173885,
      "auxiliary_loss_mlp": 0.01023482,
      "balance_loss_clip": 1.04920065,
      "balance_loss_mlp": 1.01573682,
      "epoch": 0.6684302290627067,
      "flos": 22996977868800.0,
      "grad_norm": 2.5895206283334686,
      "language_loss": 0.7593807,
      "learning_rate": 1.046447010251821e-06,
      "loss": 0.78135431,
      "num_input_tokens_seen": 119655850,
      "step": 5559,
      "time_per_iteration": 2.625943899154663
    },
    {
      "auxiliary_loss_clip": 0.01174734,
      "auxiliary_loss_mlp": 0.01024408,
      "balance_loss_clip": 1.05175567,
      "balance_loss_mlp": 1.01679993,
      "epoch": 0.6685504719533457,
      "flos": 26573824247040.0,
      "grad_norm": 4.33055840178003,
      "language_loss": 0.75498557,
      "learning_rate": 1.0457623473876157e-06,
      "loss": 0.77697706,
      "num_input_tokens_seen": 119675355,
      "step": 5560,
      "time_per_iteration": 2.6454930305480957
    },
    {
      "auxiliary_loss_clip": 0.01173102,
      "auxiliary_loss_mlp": 0.01029101,
      "balance_loss_clip": 1.05011868,
      "balance_loss_mlp": 1.02205563,
      "epoch": 0.6686707148439849,
      "flos": 28986087870720.0,
      "grad_norm": 1.9135272455019459,
      "language_loss": 0.70974451,
      "learning_rate": 1.0450778292805046e-06,
      "loss": 0.73176658,
      "num_input_tokens_seen": 119695340,
      "step": 5561,
      "time_per_iteration": 4.507492780685425
    },
    {
      "auxiliary_loss_clip": 0.01174299,
      "auxiliary_loss_mlp": 0.01024748,
      "balance_loss_clip": 1.04903626,
      "balance_loss_mlp": 1.01687407,
      "epoch": 0.6687909577346239,
      "flos": 23623152687360.0,
      "grad_norm": 2.298293126120275,
      "language_loss": 0.78913689,
      "learning_rate": 1.0443934560343267e-06,
      "loss": 0.81112731,
      "num_input_tokens_seen": 119716750,
      "step": 5562,
      "time_per_iteration": 3.5303709506988525
    },
    {
      "auxiliary_loss_clip": 0.01172976,
      "auxiliary_loss_mlp": 0.01022981,
      "balance_loss_clip": 1.04949045,
      "balance_loss_mlp": 1.01489246,
      "epoch": 0.668911200625263,
      "flos": 23148593176320.0,
      "grad_norm": 2.18184365301675,
      "language_loss": 0.7851156,
      "learning_rate": 1.0437092277529034e-06,
      "loss": 0.80707514,
      "num_input_tokens_seen": 119736005,
      "step": 5563,
      "time_per_iteration": 2.6693246364593506
    },
    {
      "auxiliary_loss_clip": 0.01172266,
      "auxiliary_loss_mlp": 0.0102689,
      "balance_loss_clip": 1.04905081,
      "balance_loss_mlp": 1.01911187,
      "epoch": 0.6690314435159022,
      "flos": 18551919853440.0,
      "grad_norm": 2.008657607150064,
      "language_loss": 0.74103379,
      "learning_rate": 1.0430251445400292e-06,
      "loss": 0.7630254,
      "num_input_tokens_seen": 119754050,
      "step": 5564,
      "time_per_iteration": 2.6523234844207764
    },
    {
      "auxiliary_loss_clip": 0.0117709,
      "auxiliary_loss_mlp": 0.01026273,
      "balance_loss_clip": 1.05243516,
      "balance_loss_mlp": 1.01804161,
      "epoch": 0.6691516864065412,
      "flos": 31759540704000.0,
      "grad_norm": 2.0950815333491004,
      "language_loss": 0.62761861,
      "learning_rate": 1.0423412064994787e-06,
      "loss": 0.64965224,
      "num_input_tokens_seen": 119774820,
      "step": 5565,
      "time_per_iteration": 2.8279402256011963
    },
    {
      "auxiliary_loss_clip": 0.01169034,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.04836142,
      "balance_loss_mlp": 1.01971698,
      "epoch": 0.6692719292971803,
      "flos": 34933864296960.0,
      "grad_norm": 2.5732476865749607,
      "language_loss": 0.73957145,
      "learning_rate": 1.0416574137350064e-06,
      "loss": 0.76153398,
      "num_input_tokens_seen": 119795525,
      "step": 5566,
      "time_per_iteration": 2.803365468978882
    },
    {
      "auxiliary_loss_clip": 0.01177493,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 1.05241764,
      "balance_loss_mlp": 1.0190469,
      "epoch": 0.6693921721878194,
      "flos": 20449188230400.0,
      "grad_norm": 2.9365258100855143,
      "language_loss": 0.81105685,
      "learning_rate": 1.0409737663503428e-06,
      "loss": 0.83310223,
      "num_input_tokens_seen": 119813905,
      "step": 5567,
      "time_per_iteration": 2.738090753555298
    },
    {
      "auxiliary_loss_clip": 0.0117215,
      "auxiliary_loss_mlp": 0.01029203,
      "balance_loss_clip": 1.04801893,
      "balance_loss_mlp": 1.02111495,
      "epoch": 0.6695124150784585,
      "flos": 16614538963200.0,
      "grad_norm": 2.458463335040119,
      "language_loss": 0.8298769,
      "learning_rate": 1.040290264449196e-06,
      "loss": 0.85189044,
      "num_input_tokens_seen": 119832010,
      "step": 5568,
      "time_per_iteration": 2.6993508338928223
    },
    {
      "auxiliary_loss_clip": 0.01169717,
      "auxiliary_loss_mlp": 0.01025765,
      "balance_loss_clip": 1.04812121,
      "balance_loss_mlp": 1.01788509,
      "epoch": 0.6696326579690975,
      "flos": 26652145852800.0,
      "grad_norm": 3.149433016304907,
      "language_loss": 0.64107466,
      "learning_rate": 1.0396069081352532e-06,
      "loss": 0.66302943,
      "num_input_tokens_seen": 119851165,
      "step": 5569,
      "time_per_iteration": 2.756073236465454
    },
    {
      "auxiliary_loss_clip": 0.01070699,
      "auxiliary_loss_mlp": 0.01000469,
      "balance_loss_clip": 1.01245403,
      "balance_loss_mlp": 0.99940825,
      "epoch": 0.6697529008597367,
      "flos": 66964603662720.0,
      "grad_norm": 0.7716208313540724,
      "language_loss": 0.55965346,
      "learning_rate": 1.0389236975121782e-06,
      "loss": 0.58036518,
      "num_input_tokens_seen": 119906015,
      "step": 5570,
      "time_per_iteration": 3.222053050994873
    },
    {
      "auxiliary_loss_clip": 0.01176361,
      "auxiliary_loss_mlp": 0.01030248,
      "balance_loss_clip": 1.04983342,
      "balance_loss_mlp": 1.02194476,
      "epoch": 0.6698731437503758,
      "flos": 20886939279360.0,
      "grad_norm": 2.892032880790181,
      "language_loss": 0.71464181,
      "learning_rate": 1.0382406326836147e-06,
      "loss": 0.73670793,
      "num_input_tokens_seen": 119925160,
      "step": 5571,
      "time_per_iteration": 2.5881614685058594
    },
    {
      "auxiliary_loss_clip": 0.01179402,
      "auxiliary_loss_mlp": 0.01028386,
      "balance_loss_clip": 1.0520879,
      "balance_loss_mlp": 1.01957059,
      "epoch": 0.6699933866410148,
      "flos": 20409470766720.0,
      "grad_norm": 2.037013467979169,
      "language_loss": 0.75634444,
      "learning_rate": 1.0375577137531828e-06,
      "loss": 0.77842236,
      "num_input_tokens_seen": 119943720,
      "step": 5572,
      "time_per_iteration": 2.68337345123291
    },
    {
      "auxiliary_loss_clip": 0.01173405,
      "auxiliary_loss_mlp": 0.01025688,
      "balance_loss_clip": 1.04893184,
      "balance_loss_mlp": 1.01774859,
      "epoch": 0.670113629531654,
      "flos": 29023075900800.0,
      "grad_norm": 2.118351065691845,
      "language_loss": 0.72075415,
      "learning_rate": 1.0368749408244802e-06,
      "loss": 0.7427451,
      "num_input_tokens_seen": 119966640,
      "step": 5573,
      "time_per_iteration": 2.6242990493774414
    },
    {
      "auxiliary_loss_clip": 0.01176064,
      "auxiliary_loss_mlp": 0.01023707,
      "balance_loss_clip": 1.05249369,
      "balance_loss_mlp": 1.0158149,
      "epoch": 0.670233872422293,
      "flos": 19791699730560.0,
      "grad_norm": 1.9461508365960576,
      "language_loss": 0.78905123,
      "learning_rate": 1.0361923140010836e-06,
      "loss": 0.81104898,
      "num_input_tokens_seen": 119985125,
      "step": 5574,
      "time_per_iteration": 2.625638484954834
    },
    {
      "auxiliary_loss_clip": 0.01177584,
      "auxiliary_loss_mlp": 0.01027019,
      "balance_loss_clip": 1.05077815,
      "balance_loss_mlp": 1.01860881,
      "epoch": 0.6703541153129321,
      "flos": 24243689070720.0,
      "grad_norm": 2.1198886902137177,
      "language_loss": 0.63414431,
      "learning_rate": 1.0355098333865455e-06,
      "loss": 0.65619034,
      "num_input_tokens_seen": 120004355,
      "step": 5575,
      "time_per_iteration": 2.6732919216156006
    },
    {
      "auxiliary_loss_clip": 0.01172582,
      "auxiliary_loss_mlp": 0.01027219,
      "balance_loss_clip": 1.05187142,
      "balance_loss_mlp": 1.0195601,
      "epoch": 0.6704743582035713,
      "flos": 26688523351680.0,
      "grad_norm": 10.43301676133284,
      "language_loss": 0.69545937,
      "learning_rate": 1.0348274990844006e-06,
      "loss": 0.71745735,
      "num_input_tokens_seen": 120027115,
      "step": 5576,
      "time_per_iteration": 2.699713945388794
    },
    {
      "auxiliary_loss_clip": 0.01171111,
      "auxiliary_loss_mlp": 0.01029027,
      "balance_loss_clip": 1.04822361,
      "balance_loss_mlp": 1.02114773,
      "epoch": 0.6705946010942103,
      "flos": 23514379326720.0,
      "grad_norm": 1.9112162810702824,
      "language_loss": 0.72612506,
      "learning_rate": 1.034145311198155e-06,
      "loss": 0.74812645,
      "num_input_tokens_seen": 120047130,
      "step": 5577,
      "time_per_iteration": 2.6675190925598145
    },
    {
      "auxiliary_loss_clip": 0.01171551,
      "auxiliary_loss_mlp": 0.01025685,
      "balance_loss_clip": 1.04875362,
      "balance_loss_mlp": 1.01841915,
      "epoch": 0.6707148439848494,
      "flos": 24061011477120.0,
      "grad_norm": 1.8588724376294392,
      "language_loss": 0.63856232,
      "learning_rate": 1.0334632698312989e-06,
      "loss": 0.66053474,
      "num_input_tokens_seen": 120067925,
      "step": 5578,
      "time_per_iteration": 2.6658449172973633
    },
    {
      "auxiliary_loss_clip": 0.0117376,
      "auxiliary_loss_mlp": 0.01023293,
      "balance_loss_clip": 1.05063629,
      "balance_loss_mlp": 1.01567006,
      "epoch": 0.6708350868754885,
      "flos": 22528667324160.0,
      "grad_norm": 3.07273262275632,
      "language_loss": 0.75372535,
      "learning_rate": 1.032781375087295e-06,
      "loss": 0.77569592,
      "num_input_tokens_seen": 120087825,
      "step": 5579,
      "time_per_iteration": 3.6263139247894287
    },
    {
      "auxiliary_loss_clip": 0.01174759,
      "auxiliary_loss_mlp": 0.01024658,
      "balance_loss_clip": 1.05208135,
      "balance_loss_mlp": 1.01720786,
      "epoch": 0.6709553297661276,
      "flos": 25227749047680.0,
      "grad_norm": 3.820282386310623,
      "language_loss": 0.67863297,
      "learning_rate": 1.0320996270695891e-06,
      "loss": 0.70062721,
      "num_input_tokens_seen": 120108895,
      "step": 5580,
      "time_per_iteration": 2.653106689453125
    },
    {
      "auxiliary_loss_clip": 0.01176437,
      "auxiliary_loss_mlp": 0.01025595,
      "balance_loss_clip": 1.05240512,
      "balance_loss_mlp": 1.01741767,
      "epoch": 0.6710755726567667,
      "flos": 20448757267200.0,
      "grad_norm": 1.9286650753759325,
      "language_loss": 0.73492777,
      "learning_rate": 1.0314180258815998e-06,
      "loss": 0.75694811,
      "num_input_tokens_seen": 120127535,
      "step": 5581,
      "time_per_iteration": 2.637791633605957
    },
    {
      "auxiliary_loss_clip": 0.01168811,
      "auxiliary_loss_mlp": 0.01023518,
      "balance_loss_clip": 1.04832232,
      "balance_loss_mlp": 1.01610935,
      "epoch": 0.6711958155474057,
      "flos": 25995411538560.0,
      "grad_norm": 1.585410710902255,
      "language_loss": 0.74458039,
      "learning_rate": 1.0307365716267247e-06,
      "loss": 0.76650369,
      "num_input_tokens_seen": 120147980,
      "step": 5582,
      "time_per_iteration": 2.60011625289917
    },
    {
      "auxiliary_loss_clip": 0.01175765,
      "auxiliary_loss_mlp": 0.01028859,
      "balance_loss_clip": 1.05043316,
      "balance_loss_mlp": 1.02088964,
      "epoch": 0.6713160584380449,
      "flos": 19937712516480.0,
      "grad_norm": 2.2326542293088885,
      "language_loss": 0.78550243,
      "learning_rate": 1.0300552644083423e-06,
      "loss": 0.8075487,
      "num_input_tokens_seen": 120166905,
      "step": 5583,
      "time_per_iteration": 2.641106128692627
    },
    {
      "auxiliary_loss_clip": 0.01176583,
      "auxiliary_loss_mlp": 0.01023891,
      "balance_loss_clip": 1.04931378,
      "balance_loss_mlp": 1.01547444,
      "epoch": 0.6714363013286839,
      "flos": 18223373128320.0,
      "grad_norm": 43.5666966424255,
      "language_loss": 0.71853822,
      "learning_rate": 1.0293741043298036e-06,
      "loss": 0.74054301,
      "num_input_tokens_seen": 120185255,
      "step": 5584,
      "time_per_iteration": 2.575324773788452
    },
    {
      "auxiliary_loss_clip": 0.01178814,
      "auxiliary_loss_mlp": 0.01023918,
      "balance_loss_clip": 1.05351472,
      "balance_loss_mlp": 1.01599097,
      "epoch": 0.671556544219323,
      "flos": 25812374808960.0,
      "grad_norm": 2.7428023296337725,
      "language_loss": 0.7121709,
      "learning_rate": 1.0286930914944436e-06,
      "loss": 0.73419821,
      "num_input_tokens_seen": 120205070,
      "step": 5585,
      "time_per_iteration": 2.726238250732422
    },
    {
      "auxiliary_loss_clip": 0.01173254,
      "auxiliary_loss_mlp": 0.0102703,
      "balance_loss_clip": 1.04700685,
      "balance_loss_mlp": 1.01866722,
      "epoch": 0.6716767871099621,
      "flos": 15850431918720.0,
      "grad_norm": 4.678513543366919,
      "language_loss": 0.7771759,
      "learning_rate": 1.0280122260055684e-06,
      "loss": 0.79917878,
      "num_input_tokens_seen": 120220780,
      "step": 5586,
      "time_per_iteration": 2.612778902053833
    },
    {
      "auxiliary_loss_clip": 0.01175676,
      "auxiliary_loss_mlp": 0.01024963,
      "balance_loss_clip": 1.05096555,
      "balance_loss_mlp": 1.01672006,
      "epoch": 0.6717970300006012,
      "flos": 19756112330880.0,
      "grad_norm": 2.4337115173404285,
      "language_loss": 0.82570493,
      "learning_rate": 1.0273315079664652e-06,
      "loss": 0.84771127,
      "num_input_tokens_seen": 120238735,
      "step": 5587,
      "time_per_iteration": 3.4365828037261963
    },
    {
      "auxiliary_loss_clip": 0.01173923,
      "auxiliary_loss_mlp": 0.01028099,
      "balance_loss_clip": 1.04994822,
      "balance_loss_mlp": 1.01980221,
      "epoch": 0.6719172728912403,
      "flos": 25485049146240.0,
      "grad_norm": 2.589495711347046,
      "language_loss": 0.74476033,
      "learning_rate": 1.0266509374803992e-06,
      "loss": 0.76678056,
      "num_input_tokens_seen": 120259895,
      "step": 5588,
      "time_per_iteration": 3.6231131553649902
    },
    {
      "auxiliary_loss_clip": 0.01174493,
      "auxiliary_loss_mlp": 0.01027314,
      "balance_loss_clip": 1.05007911,
      "balance_loss_mlp": 1.01947618,
      "epoch": 0.6720375157818794,
      "flos": 15880344969600.0,
      "grad_norm": 2.4746039532110147,
      "language_loss": 0.84318447,
      "learning_rate": 1.0259705146506123e-06,
      "loss": 0.86520249,
      "num_input_tokens_seen": 120274790,
      "step": 5589,
      "time_per_iteration": 3.4915919303894043
    },
    {
      "auxiliary_loss_clip": 0.01173897,
      "auxiliary_loss_mlp": 0.0102285,
      "balance_loss_clip": 1.04984081,
      "balance_loss_mlp": 1.01547694,
      "epoch": 0.6721577586725185,
      "flos": 32010843231360.0,
      "grad_norm": 2.2287337737746906,
      "language_loss": 0.77717698,
      "learning_rate": 1.025290239580324e-06,
      "loss": 0.79914445,
      "num_input_tokens_seen": 120295460,
      "step": 5590,
      "time_per_iteration": 2.675530433654785
    },
    {
      "auxiliary_loss_clip": 0.01173649,
      "auxiliary_loss_mlp": 0.01025467,
      "balance_loss_clip": 1.05000019,
      "balance_loss_mlp": 1.01731277,
      "epoch": 0.6722780015631575,
      "flos": 20737873837440.0,
      "grad_norm": 2.341759990786922,
      "language_loss": 0.75437039,
      "learning_rate": 1.0246101123727313e-06,
      "loss": 0.77636153,
      "num_input_tokens_seen": 120314440,
      "step": 5591,
      "time_per_iteration": 2.6325278282165527
    },
    {
      "auxiliary_loss_clip": 0.01172391,
      "auxiliary_loss_mlp": 0.01026367,
      "balance_loss_clip": 1.05067039,
      "balance_loss_mlp": 1.01887476,
      "epoch": 0.6723982444537967,
      "flos": 16909617191040.0,
      "grad_norm": 1.9673170585420756,
      "language_loss": 0.78697431,
      "learning_rate": 1.0239301331310085e-06,
      "loss": 0.80896193,
      "num_input_tokens_seen": 120332060,
      "step": 5592,
      "time_per_iteration": 2.604346513748169
    },
    {
      "auxiliary_loss_clip": 0.0117108,
      "auxiliary_loss_mlp": 0.01026452,
      "balance_loss_clip": 1.04994488,
      "balance_loss_mlp": 1.01920104,
      "epoch": 0.6725184873444358,
      "flos": 20667812359680.0,
      "grad_norm": 1.6721252250639094,
      "language_loss": 0.88556254,
      "learning_rate": 1.0232503019583088e-06,
      "loss": 0.90753794,
      "num_input_tokens_seen": 120351670,
      "step": 5593,
      "time_per_iteration": 2.656306266784668
    },
    {
      "auxiliary_loss_clip": 0.0117198,
      "auxiliary_loss_mlp": 0.01030394,
      "balance_loss_clip": 1.04882932,
      "balance_loss_mlp": 1.02232337,
      "epoch": 0.6726387302350748,
      "flos": 23727616416000.0,
      "grad_norm": 2.155679960520118,
      "language_loss": 0.69800901,
      "learning_rate": 1.0225706189577619e-06,
      "loss": 0.72003275,
      "num_input_tokens_seen": 120370195,
      "step": 5594,
      "time_per_iteration": 2.6629834175109863
    },
    {
      "auxiliary_loss_clip": 0.01174078,
      "auxiliary_loss_mlp": 0.01025339,
      "balance_loss_clip": 1.0491879,
      "balance_loss_mlp": 1.01696444,
      "epoch": 0.672758973125714,
      "flos": 15188274650880.0,
      "grad_norm": 2.794338044182494,
      "language_loss": 0.74848616,
      "learning_rate": 1.021891084232475e-06,
      "loss": 0.77048039,
      "num_input_tokens_seen": 120388130,
      "step": 5595,
      "time_per_iteration": 2.778447389602661
    },
    {
      "auxiliary_loss_clip": 0.01174937,
      "auxiliary_loss_mlp": 0.01026366,
      "balance_loss_clip": 1.05090117,
      "balance_loss_mlp": 1.01856399,
      "epoch": 0.672879216016353,
      "flos": 18077252601600.0,
      "grad_norm": 2.1347221051232435,
      "language_loss": 0.79924631,
      "learning_rate": 1.0212116978855325e-06,
      "loss": 0.82125926,
      "num_input_tokens_seen": 120406145,
      "step": 5596,
      "time_per_iteration": 2.6424896717071533
    },
    {
      "auxiliary_loss_clip": 0.01168724,
      "auxiliary_loss_mlp": 0.01027028,
      "balance_loss_clip": 1.04687977,
      "balance_loss_mlp": 1.01913607,
      "epoch": 0.6729994589069921,
      "flos": 23476349802240.0,
      "grad_norm": 2.2013698702590925,
      "language_loss": 0.78995037,
      "learning_rate": 1.020532460019997e-06,
      "loss": 0.81190795,
      "num_input_tokens_seen": 120425395,
      "step": 5597,
      "time_per_iteration": 2.6709401607513428
    },
    {
      "auxiliary_loss_clip": 0.01174508,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.04959667,
      "balance_loss_mlp": 1.02080107,
      "epoch": 0.6731197017976313,
      "flos": 26322018929280.0,
      "grad_norm": 1.9124270680224142,
      "language_loss": 0.71044016,
      "learning_rate": 1.0198533707389096e-06,
      "loss": 0.73247665,
      "num_input_tokens_seen": 120446270,
      "step": 5598,
      "time_per_iteration": 2.660219192504883
    },
    {
      "auxiliary_loss_clip": 0.01174182,
      "auxiliary_loss_mlp": 0.0102939,
      "balance_loss_clip": 1.0507741,
      "balance_loss_mlp": 1.02111638,
      "epoch": 0.6732399446882703,
      "flos": 21616428591360.0,
      "grad_norm": 2.0502704661254176,
      "language_loss": 0.73224473,
      "learning_rate": 1.0191744301452853e-06,
      "loss": 0.75428045,
      "num_input_tokens_seen": 120465570,
      "step": 5599,
      "time_per_iteration": 2.659208059310913
    },
    {
      "auxiliary_loss_clip": 0.01171256,
      "auxiliary_loss_mlp": 0.01024551,
      "balance_loss_clip": 1.04850435,
      "balance_loss_mlp": 1.01673114,
      "epoch": 0.6733601875789094,
      "flos": 25880173729920.0,
      "grad_norm": 2.2146965895233386,
      "language_loss": 0.7032181,
      "learning_rate": 1.0184956383421208e-06,
      "loss": 0.7251761,
      "num_input_tokens_seen": 120484220,
      "step": 5600,
      "time_per_iteration": 2.6635489463806152
    },
    {
      "auxiliary_loss_clip": 0.01175062,
      "auxiliary_loss_mlp": 0.01025729,
      "balance_loss_clip": 1.0525527,
      "balance_loss_mlp": 1.01764023,
      "epoch": 0.6734804304695485,
      "flos": 22929573997440.0,
      "grad_norm": 6.396218092113737,
      "language_loss": 0.65220118,
      "learning_rate": 1.017816995432387e-06,
      "loss": 0.67420912,
      "num_input_tokens_seen": 120503320,
      "step": 5601,
      "time_per_iteration": 2.692035675048828
    },
    {
      "auxiliary_loss_clip": 0.01170901,
      "auxiliary_loss_mlp": 0.01024816,
      "balance_loss_clip": 1.049155,
      "balance_loss_mlp": 1.01706195,
      "epoch": 0.6736006733601876,
      "flos": 18697968552960.0,
      "grad_norm": 2.1461315280222473,
      "language_loss": 0.74709773,
      "learning_rate": 1.0171385015190353e-06,
      "loss": 0.76905501,
      "num_input_tokens_seen": 120523180,
      "step": 5602,
      "time_per_iteration": 2.5673434734344482
    },
    {
      "auxiliary_loss_clip": 0.01168395,
      "auxiliary_loss_mlp": 0.01032739,
      "balance_loss_clip": 1.04756188,
      "balance_loss_mlp": 1.0252943,
      "epoch": 0.6737209162508266,
      "flos": 19427745173760.0,
      "grad_norm": 1.8961521234379712,
      "language_loss": 0.73270833,
      "learning_rate": 1.0164601567049908e-06,
      "loss": 0.75471967,
      "num_input_tokens_seen": 120541710,
      "step": 5603,
      "time_per_iteration": 2.6104555130004883
    },
    {
      "auxiliary_loss_clip": 0.01176903,
      "auxiliary_loss_mlp": 0.01028361,
      "balance_loss_clip": 1.05189514,
      "balance_loss_mlp": 1.01968586,
      "epoch": 0.6738411591414658,
      "flos": 20158060498560.0,
      "grad_norm": 1.7865038895471055,
      "language_loss": 0.80617452,
      "learning_rate": 1.015781961093158e-06,
      "loss": 0.82822728,
      "num_input_tokens_seen": 120561030,
      "step": 5604,
      "time_per_iteration": 2.618657350540161
    },
    {
      "auxiliary_loss_clip": 0.01171824,
      "auxiliary_loss_mlp": 0.01028235,
      "balance_loss_clip": 1.04889107,
      "balance_loss_mlp": 1.02045703,
      "epoch": 0.6739614020321049,
      "flos": 21653847584640.0,
      "grad_norm": 1.5352367323661915,
      "language_loss": 0.77471912,
      "learning_rate": 1.0151039147864197e-06,
      "loss": 0.79671967,
      "num_input_tokens_seen": 120581005,
      "step": 5605,
      "time_per_iteration": 2.6387856006622314
    },
    {
      "auxiliary_loss_clip": 0.01175766,
      "auxiliary_loss_mlp": 0.01031646,
      "balance_loss_clip": 1.05041265,
      "balance_loss_mlp": 1.02310205,
      "epoch": 0.6740816449227439,
      "flos": 19171702051200.0,
      "grad_norm": 2.1951675050710744,
      "language_loss": 0.66273499,
      "learning_rate": 1.0144260178876336e-06,
      "loss": 0.68480915,
      "num_input_tokens_seen": 120600350,
      "step": 5606,
      "time_per_iteration": 3.6394643783569336
    },
    {
      "auxiliary_loss_clip": 0.01173047,
      "auxiliary_loss_mlp": 0.01024148,
      "balance_loss_clip": 1.04920208,
      "balance_loss_mlp": 1.01636386,
      "epoch": 0.6742018878133831,
      "flos": 21097015971840.0,
      "grad_norm": 2.1936279371987912,
      "language_loss": 0.670241,
      "learning_rate": 1.0137482704996388e-06,
      "loss": 0.69221294,
      "num_input_tokens_seen": 120614700,
      "step": 5607,
      "time_per_iteration": 2.6009521484375
    },
    {
      "auxiliary_loss_clip": 0.01174088,
      "auxiliary_loss_mlp": 0.0102579,
      "balance_loss_clip": 1.04979455,
      "balance_loss_mlp": 1.01711774,
      "epoch": 0.6743221307040221,
      "flos": 23549966726400.0,
      "grad_norm": 2.119404741610667,
      "language_loss": 0.78994411,
      "learning_rate": 1.0130706727252461e-06,
      "loss": 0.81194293,
      "num_input_tokens_seen": 120631755,
      "step": 5608,
      "time_per_iteration": 2.708489418029785
    },
    {
      "auxiliary_loss_clip": 0.01171558,
      "auxiliary_loss_mlp": 0.01026329,
      "balance_loss_clip": 1.048208,
      "balance_loss_mlp": 1.01860976,
      "epoch": 0.6744423735946612,
      "flos": 16249542912000.0,
      "grad_norm": 4.078780120962047,
      "language_loss": 0.68134713,
      "learning_rate": 1.0123932246672468e-06,
      "loss": 0.70332599,
      "num_input_tokens_seen": 120645900,
      "step": 5609,
      "time_per_iteration": 2.541875123977661
    },
    {
      "auxiliary_loss_clip": 0.01072076,
      "auxiliary_loss_mlp": 0.01001415,
      "balance_loss_clip": 1.01267147,
      "balance_loss_mlp": 1.00015759,
      "epoch": 0.6745626164853004,
      "flos": 57843257829120.0,
      "grad_norm": 0.7459522886164833,
      "language_loss": 0.55813265,
      "learning_rate": 1.0117159264284114e-06,
      "loss": 0.57886755,
      "num_input_tokens_seen": 120709070,
      "step": 5610,
      "time_per_iteration": 3.176469564437866
    },
    {
      "auxiliary_loss_clip": 0.01171555,
      "auxiliary_loss_mlp": 0.01022888,
      "balance_loss_clip": 1.04927158,
      "balance_loss_mlp": 1.01564932,
      "epoch": 0.6746828593759394,
      "flos": 20485027025280.0,
      "grad_norm": 1.8970609505407692,
      "language_loss": 0.77228153,
      "learning_rate": 1.0110387781114837e-06,
      "loss": 0.79422605,
      "num_input_tokens_seen": 120727685,
      "step": 5611,
      "time_per_iteration": 2.6298940181732178
    },
    {
      "auxiliary_loss_clip": 0.01171611,
      "auxiliary_loss_mlp": 0.01024714,
      "balance_loss_clip": 1.04908156,
      "balance_loss_mlp": 1.01722765,
      "epoch": 0.6748031022665785,
      "flos": 19208223204480.0,
      "grad_norm": 2.1089253603161686,
      "language_loss": 0.77293253,
      "learning_rate": 1.0103617798191872e-06,
      "loss": 0.79489577,
      "num_input_tokens_seen": 120747160,
      "step": 5612,
      "time_per_iteration": 2.7130842208862305
    },
    {
      "auxiliary_loss_clip": 0.01173039,
      "auxiliary_loss_mlp": 0.01027068,
      "balance_loss_clip": 1.04847467,
      "balance_loss_mlp": 1.01843095,
      "epoch": 0.6749233451572175,
      "flos": 15195026407680.0,
      "grad_norm": 2.6817977324418116,
      "language_loss": 0.82704198,
      "learning_rate": 1.0096849316542217e-06,
      "loss": 0.84904313,
      "num_input_tokens_seen": 120763710,
      "step": 5613,
      "time_per_iteration": 3.5754618644714355
    },
    {
      "auxiliary_loss_clip": 0.01171071,
      "auxiliary_loss_mlp": 0.01026695,
      "balance_loss_clip": 1.04709125,
      "balance_loss_mlp": 1.0182488,
      "epoch": 0.6750435880478567,
      "flos": 26499489050880.0,
      "grad_norm": 2.17357792982467,
      "language_loss": 0.75141144,
      "learning_rate": 1.0090082337192643e-06,
      "loss": 0.7733891,
      "num_input_tokens_seen": 120783355,
      "step": 5614,
      "time_per_iteration": 3.5318777561187744
    },
    {
      "auxiliary_loss_clip": 0.0117009,
      "auxiliary_loss_mlp": 0.0102186,
      "balance_loss_clip": 1.04753697,
      "balance_loss_mlp": 1.01400423,
      "epoch": 0.6751638309384957,
      "flos": 23404313076480.0,
      "grad_norm": 2.144194847619179,
      "language_loss": 0.78571415,
      "learning_rate": 1.0083316861169705e-06,
      "loss": 0.80763364,
      "num_input_tokens_seen": 120802090,
      "step": 5615,
      "time_per_iteration": 3.4695210456848145
    },
    {
      "auxiliary_loss_clip": 0.0117677,
      "auxiliary_loss_mlp": 0.01026251,
      "balance_loss_clip": 1.04991996,
      "balance_loss_mlp": 1.01741755,
      "epoch": 0.6752840738291348,
      "flos": 23441408847360.0,
      "grad_norm": 2.1210116052035497,
      "language_loss": 0.71753567,
      "learning_rate": 1.0076552889499713e-06,
      "loss": 0.73956585,
      "num_input_tokens_seen": 120822855,
      "step": 5616,
      "time_per_iteration": 2.7180097103118896
    },
    {
      "auxiliary_loss_clip": 0.01174645,
      "auxiliary_loss_mlp": 0.01028838,
      "balance_loss_clip": 1.05128658,
      "balance_loss_mlp": 1.02098215,
      "epoch": 0.675404316719774,
      "flos": 30335826257280.0,
      "grad_norm": 2.4475751234443375,
      "language_loss": 0.73823935,
      "learning_rate": 1.006979042320876e-06,
      "loss": 0.76027417,
      "num_input_tokens_seen": 120843070,
      "step": 5617,
      "time_per_iteration": 2.6944780349731445
    },
    {
      "auxiliary_loss_clip": 0.01170056,
      "auxiliary_loss_mlp": 0.01024988,
      "balance_loss_clip": 1.04668021,
      "balance_loss_mlp": 1.01756108,
      "epoch": 0.675524559610413,
      "flos": 23622613983360.0,
      "grad_norm": 2.151661405122981,
      "language_loss": 0.62876564,
      "learning_rate": 1.0063029463322702e-06,
      "loss": 0.65071607,
      "num_input_tokens_seen": 120863345,
      "step": 5618,
      "time_per_iteration": 2.751242160797119
    },
    {
      "auxiliary_loss_clip": 0.0117329,
      "auxiliary_loss_mlp": 0.01025724,
      "balance_loss_clip": 1.04886806,
      "balance_loss_mlp": 1.01750135,
      "epoch": 0.6756448025010521,
      "flos": 21248631279360.0,
      "grad_norm": 2.6044690861241424,
      "language_loss": 0.75186104,
      "learning_rate": 1.0056270010867164e-06,
      "loss": 0.77385122,
      "num_input_tokens_seen": 120880915,
      "step": 5619,
      "time_per_iteration": 2.595681667327881
    },
    {
      "auxiliary_loss_clip": 0.01177205,
      "auxiliary_loss_mlp": 0.01026105,
      "balance_loss_clip": 1.05211353,
      "balance_loss_mlp": 1.01788878,
      "epoch": 0.6757650453916912,
      "flos": 21646521210240.0,
      "grad_norm": 2.9466959257233944,
      "language_loss": 0.7816152,
      "learning_rate": 1.004951206686758e-06,
      "loss": 0.80364835,
      "num_input_tokens_seen": 120899190,
      "step": 5620,
      "time_per_iteration": 2.6996169090270996
    },
    {
      "auxiliary_loss_clip": 0.01172557,
      "auxiliary_loss_mlp": 0.01029656,
      "balance_loss_clip": 1.04999387,
      "balance_loss_mlp": 1.0217284,
      "epoch": 0.6758852882823303,
      "flos": 21795658479360.0,
      "grad_norm": 2.039033225430522,
      "language_loss": 0.7192052,
      "learning_rate": 1.0042755632349087e-06,
      "loss": 0.74122733,
      "num_input_tokens_seen": 120916080,
      "step": 5621,
      "time_per_iteration": 2.6446845531463623
    },
    {
      "auxiliary_loss_clip": 0.01172384,
      "auxiliary_loss_mlp": 0.01024616,
      "balance_loss_clip": 1.04875493,
      "balance_loss_mlp": 1.0155797,
      "epoch": 0.6760055311729694,
      "flos": 27088783580160.0,
      "grad_norm": 2.710422977794558,
      "language_loss": 0.62785059,
      "learning_rate": 1.0036000708336653e-06,
      "loss": 0.64982057,
      "num_input_tokens_seen": 120935210,
      "step": 5622,
      "time_per_iteration": 2.7084288597106934
    },
    {
      "auxiliary_loss_clip": 0.01176227,
      "auxiliary_loss_mlp": 0.01020864,
      "balance_loss_clip": 1.05209219,
      "balance_loss_mlp": 1.01327002,
      "epoch": 0.6761257740636085,
      "flos": 17999792922240.0,
      "grad_norm": 2.542672992574404,
      "language_loss": 0.79336005,
      "learning_rate": 1.0029247295854984e-06,
      "loss": 0.81533098,
      "num_input_tokens_seen": 120951830,
      "step": 5623,
      "time_per_iteration": 2.667362689971924
    },
    {
      "auxiliary_loss_clip": 0.01174015,
      "auxiliary_loss_mlp": 0.01028883,
      "balance_loss_clip": 1.05292392,
      "balance_loss_mlp": 1.02108121,
      "epoch": 0.6762460169542476,
      "flos": 15121912273920.0,
      "grad_norm": 2.1484146031203197,
      "language_loss": 0.7196449,
      "learning_rate": 1.0022495395928588e-06,
      "loss": 0.74167395,
      "num_input_tokens_seen": 120970310,
      "step": 5624,
      "time_per_iteration": 2.7074928283691406
    },
    {
      "auxiliary_loss_clip": 0.0107321,
      "auxiliary_loss_mlp": 0.01001204,
      "balance_loss_clip": 1.01256859,
      "balance_loss_mlp": 0.99970818,
      "epoch": 0.6763662598448866,
      "flos": 67886970030720.0,
      "grad_norm": 0.7861425226558052,
      "language_loss": 0.62340021,
      "learning_rate": 1.0015745009581697e-06,
      "loss": 0.64414442,
      "num_input_tokens_seen": 121031915,
      "step": 5625,
      "time_per_iteration": 3.2432339191436768
    },
    {
      "auxiliary_loss_clip": 0.01175987,
      "auxiliary_loss_mlp": 0.01023,
      "balance_loss_clip": 1.05116153,
      "balance_loss_mlp": 1.01487005,
      "epoch": 0.6764865027355258,
      "flos": 20631829910400.0,
      "grad_norm": 2.1098643717467023,
      "language_loss": 0.66941524,
      "learning_rate": 1.0008996137838343e-06,
      "loss": 0.69140512,
      "num_input_tokens_seen": 121050890,
      "step": 5626,
      "time_per_iteration": 2.6747095584869385
    },
    {
      "auxiliary_loss_clip": 0.01179075,
      "auxiliary_loss_mlp": 0.01024475,
      "balance_loss_clip": 1.05142951,
      "balance_loss_mlp": 1.01572514,
      "epoch": 0.6766067456261649,
      "flos": 21215809226880.0,
      "grad_norm": 2.243448507775775,
      "language_loss": 0.80260092,
      "learning_rate": 1.000224878172234e-06,
      "loss": 0.8246364,
      "num_input_tokens_seen": 121070015,
      "step": 5627,
      "time_per_iteration": 2.6756410598754883
    },
    {
      "auxiliary_loss_clip": 0.011715,
      "auxiliary_loss_mlp": 0.01022207,
      "balance_loss_clip": 1.04904616,
      "balance_loss_mlp": 1.0144136,
      "epoch": 0.6767269885168039,
      "flos": 19938251220480.0,
      "grad_norm": 3.4278293724591085,
      "language_loss": 0.73319626,
      "learning_rate": 9.99550294225724e-07,
      "loss": 0.75513339,
      "num_input_tokens_seen": 121089170,
      "step": 5628,
      "time_per_iteration": 2.6725807189941406
    },
    {
      "auxiliary_loss_clip": 0.01174933,
      "auxiliary_loss_mlp": 0.01023524,
      "balance_loss_clip": 1.04876399,
      "balance_loss_mlp": 1.01583529,
      "epoch": 0.6768472314074431,
      "flos": 20814076540800.0,
      "grad_norm": 2.730462034136375,
      "language_loss": 0.72766691,
      "learning_rate": 9.988758620466402e-07,
      "loss": 0.74965149,
      "num_input_tokens_seen": 121108040,
      "step": 5629,
      "time_per_iteration": 2.6610348224639893
    },
    {
      "auxiliary_loss_clip": 0.01168045,
      "auxiliary_loss_mlp": 0.01022317,
      "balance_loss_clip": 1.04807675,
      "balance_loss_mlp": 1.01505136,
      "epoch": 0.6769674742980821,
      "flos": 23186012169600.0,
      "grad_norm": 2.108248107496151,
      "language_loss": 0.76580644,
      "learning_rate": 9.982015817372917e-07,
      "loss": 0.78771007,
      "num_input_tokens_seen": 121128480,
      "step": 5630,
      "time_per_iteration": 2.635354518890381
    },
    {
      "auxiliary_loss_clip": 0.01176174,
      "auxiliary_loss_mlp": 0.01025791,
      "balance_loss_clip": 1.05216503,
      "balance_loss_mlp": 1.01721406,
      "epoch": 0.6770877171887212,
      "flos": 24242934885120.0,
      "grad_norm": 4.173491642777342,
      "language_loss": 0.81768143,
      "learning_rate": 9.975274533999657e-07,
      "loss": 0.83970112,
      "num_input_tokens_seen": 121148010,
      "step": 5631,
      "time_per_iteration": 2.7294299602508545
    },
    {
      "auxiliary_loss_clip": 0.01175325,
      "auxiliary_loss_mlp": 0.01032093,
      "balance_loss_clip": 1.04946685,
      "balance_loss_mlp": 1.02345598,
      "epoch": 0.6772079600793603,
      "flos": 18141567903360.0,
      "grad_norm": 3.260698460002492,
      "language_loss": 0.84596133,
      "learning_rate": 9.96853477136929e-07,
      "loss": 0.8680355,
      "num_input_tokens_seen": 121162755,
      "step": 5632,
      "time_per_iteration": 3.455151081085205
    },
    {
      "auxiliary_loss_clip": 0.01172951,
      "auxiliary_loss_mlp": 0.01021343,
      "balance_loss_clip": 1.04898286,
      "balance_loss_mlp": 1.01365066,
      "epoch": 0.6773282029699994,
      "flos": 22452069571200.0,
      "grad_norm": 2.32738313277004,
      "language_loss": 0.75429541,
      "learning_rate": 9.96179653050422e-07,
      "loss": 0.77623832,
      "num_input_tokens_seen": 121182915,
      "step": 5633,
      "time_per_iteration": 2.9008028507232666
    },
    {
      "auxiliary_loss_clip": 0.01174706,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.04989564,
      "balance_loss_mlp": 1.01606619,
      "epoch": 0.6774484458606385,
      "flos": 18693730748160.0,
      "grad_norm": 3.362971496385384,
      "language_loss": 0.73902237,
      "learning_rate": 9.955059812426635e-07,
      "loss": 0.76101053,
      "num_input_tokens_seen": 121200445,
      "step": 5634,
      "time_per_iteration": 2.6708548069000244
    },
    {
      "auxiliary_loss_clip": 0.01175827,
      "auxiliary_loss_mlp": 0.01027314,
      "balance_loss_clip": 1.05253696,
      "balance_loss_mlp": 1.01945519,
      "epoch": 0.6775686887512776,
      "flos": 25994046821760.0,
      "grad_norm": 2.925523030031409,
      "language_loss": 0.82927877,
      "learning_rate": 9.948324618158493e-07,
      "loss": 0.85131019,
      "num_input_tokens_seen": 121220785,
      "step": 5635,
      "time_per_iteration": 2.707113265991211
    },
    {
      "auxiliary_loss_clip": 0.01175575,
      "auxiliary_loss_mlp": 0.01028231,
      "balance_loss_clip": 1.04942942,
      "balance_loss_mlp": 1.02064943,
      "epoch": 0.6776889316419167,
      "flos": 13587987922560.0,
      "grad_norm": 3.166573914929639,
      "language_loss": 0.77608323,
      "learning_rate": 9.941590948721502e-07,
      "loss": 0.79812127,
      "num_input_tokens_seen": 121237985,
      "step": 5636,
      "time_per_iteration": 2.605391025543213
    },
    {
      "auxiliary_loss_clip": 0.01174,
      "auxiliary_loss_mlp": 0.01024525,
      "balance_loss_clip": 1.05006051,
      "balance_loss_mlp": 1.01614451,
      "epoch": 0.6778091745325557,
      "flos": 27601121220480.0,
      "grad_norm": 1.6221588591809148,
      "language_loss": 0.7669872,
      "learning_rate": 9.934858805137188e-07,
      "loss": 0.7889725,
      "num_input_tokens_seen": 121258635,
      "step": 5637,
      "time_per_iteration": 2.6569643020629883
    },
    {
      "auxiliary_loss_clip": 0.01172976,
      "auxiliary_loss_mlp": 0.01020428,
      "balance_loss_clip": 1.04999936,
      "balance_loss_mlp": 1.01229167,
      "epoch": 0.6779294174231949,
      "flos": 18734058743040.0,
      "grad_norm": 2.792793434358393,
      "language_loss": 0.80731213,
      "learning_rate": 9.92812818842677e-07,
      "loss": 0.82924616,
      "num_input_tokens_seen": 121277810,
      "step": 5638,
      "time_per_iteration": 2.701890707015991
    },
    {
      "auxiliary_loss_clip": 0.01171941,
      "auxiliary_loss_mlp": 0.01021186,
      "balance_loss_clip": 1.04716897,
      "balance_loss_mlp": 1.01297307,
      "epoch": 0.678049660313834,
      "flos": 45873797765760.0,
      "grad_norm": 1.8932044465576554,
      "language_loss": 0.64445889,
      "learning_rate": 9.921399099611306e-07,
      "loss": 0.66639018,
      "num_input_tokens_seen": 121298975,
      "step": 5639,
      "time_per_iteration": 2.936962604522705
    },
    {
      "auxiliary_loss_clip": 0.01173316,
      "auxiliary_loss_mlp": 0.01026918,
      "balance_loss_clip": 1.05039096,
      "balance_loss_mlp": 1.01945567,
      "epoch": 0.678169903204473,
      "flos": 19974556892160.0,
      "grad_norm": 2.145876264112026,
      "language_loss": 0.68490231,
      "learning_rate": 9.914671539711588e-07,
      "loss": 0.70690465,
      "num_input_tokens_seen": 121318495,
      "step": 5640,
      "time_per_iteration": 4.544239282608032
    },
    {
      "auxiliary_loss_clip": 0.01172503,
      "auxiliary_loss_mlp": 0.01030718,
      "balance_loss_clip": 1.04800081,
      "balance_loss_mlp": 1.02238572,
      "epoch": 0.6782901460951122,
      "flos": 21395613732480.0,
      "grad_norm": 3.4034430938535465,
      "language_loss": 0.778669,
      "learning_rate": 9.90794550974817e-07,
      "loss": 0.80070126,
      "num_input_tokens_seen": 121338890,
      "step": 5641,
      "time_per_iteration": 3.516292095184326
    },
    {
      "auxiliary_loss_clip": 0.01178916,
      "auxiliary_loss_mlp": 0.0102557,
      "balance_loss_clip": 1.05211747,
      "balance_loss_mlp": 1.01776505,
      "epoch": 0.6784103889857512,
      "flos": 21434002392960.0,
      "grad_norm": 2.8590254291008077,
      "language_loss": 0.81321871,
      "learning_rate": 9.901221010741407e-07,
      "loss": 0.83526361,
      "num_input_tokens_seen": 121358210,
      "step": 5642,
      "time_per_iteration": 2.580087184906006
    },
    {
      "auxiliary_loss_clip": 0.01177202,
      "auxiliary_loss_mlp": 0.01023696,
      "balance_loss_clip": 1.05197954,
      "balance_loss_mlp": 1.01532114,
      "epoch": 0.6785306318763903,
      "flos": 32671923091200.0,
      "grad_norm": 1.989869203503109,
      "language_loss": 0.74782705,
      "learning_rate": 9.894498043711375e-07,
      "loss": 0.76983595,
      "num_input_tokens_seen": 121379955,
      "step": 5643,
      "time_per_iteration": 2.6835103034973145
    },
    {
      "auxiliary_loss_clip": 0.01173497,
      "auxiliary_loss_mlp": 0.01024975,
      "balance_loss_clip": 1.05000865,
      "balance_loss_mlp": 1.01759553,
      "epoch": 0.6786508747670293,
      "flos": 25632139340160.0,
      "grad_norm": 2.139263190338392,
      "language_loss": 0.69674301,
      "learning_rate": 9.887776609677962e-07,
      "loss": 0.71872771,
      "num_input_tokens_seen": 121401325,
      "step": 5644,
      "time_per_iteration": 2.7088475227355957
    },
    {
      "auxiliary_loss_clip": 0.01171494,
      "auxiliary_loss_mlp": 0.01022584,
      "balance_loss_clip": 1.04854941,
      "balance_loss_mlp": 1.01466227,
      "epoch": 0.6787711176576685,
      "flos": 19171881619200.0,
      "grad_norm": 1.8107930491650799,
      "language_loss": 0.72662419,
      "learning_rate": 9.88105670966079e-07,
      "loss": 0.74856496,
      "num_input_tokens_seen": 121419785,
      "step": 5645,
      "time_per_iteration": 2.620922088623047
    },
    {
      "auxiliary_loss_clip": 0.01170263,
      "auxiliary_loss_mlp": 0.01032647,
      "balance_loss_clip": 1.04841018,
      "balance_loss_mlp": 1.02458262,
      "epoch": 0.6788913605483076,
      "flos": 13985159581440.0,
      "grad_norm": 2.2660781971964417,
      "language_loss": 0.78720939,
      "learning_rate": 9.874338344679283e-07,
      "loss": 0.80923855,
      "num_input_tokens_seen": 121435630,
      "step": 5646,
      "time_per_iteration": 2.597813606262207
    },
    {
      "auxiliary_loss_clip": 0.01172802,
      "auxiliary_loss_mlp": 0.01024482,
      "balance_loss_clip": 1.05071354,
      "balance_loss_mlp": 1.01722217,
      "epoch": 0.6790116034389466,
      "flos": 22017586659840.0,
      "grad_norm": 1.80310421999605,
      "language_loss": 0.73692191,
      "learning_rate": 9.86762151575259e-07,
      "loss": 0.7588948,
      "num_input_tokens_seen": 121455625,
      "step": 5647,
      "time_per_iteration": 2.644386053085327
    },
    {
      "auxiliary_loss_clip": 0.01168778,
      "auxiliary_loss_mlp": 0.01022279,
      "balance_loss_clip": 1.04875302,
      "balance_loss_mlp": 1.01504326,
      "epoch": 0.6791318463295858,
      "flos": 20922454851840.0,
      "grad_norm": 1.6273047279032327,
      "language_loss": 0.80405819,
      "learning_rate": 9.860906223899651e-07,
      "loss": 0.82596874,
      "num_input_tokens_seen": 121475020,
      "step": 5648,
      "time_per_iteration": 2.63097882270813
    },
    {
      "auxiliary_loss_clip": 0.01175166,
      "auxiliary_loss_mlp": 0.01027458,
      "balance_loss_clip": 1.05076134,
      "balance_loss_mlp": 1.01958144,
      "epoch": 0.6792520892202248,
      "flos": 28512749422080.0,
      "grad_norm": 2.2435415422882663,
      "language_loss": 0.75178182,
      "learning_rate": 9.854192470139184e-07,
      "loss": 0.77380812,
      "num_input_tokens_seen": 121496500,
      "step": 5649,
      "time_per_iteration": 2.7116825580596924
    },
    {
      "auxiliary_loss_clip": 0.0117531,
      "auxiliary_loss_mlp": 0.01022366,
      "balance_loss_clip": 1.05199718,
      "balance_loss_mlp": 1.01469469,
      "epoch": 0.6793723321108639,
      "flos": 20011904058240.0,
      "grad_norm": 2.377186282511466,
      "language_loss": 0.7193774,
      "learning_rate": 9.847480255489645e-07,
      "loss": 0.74135411,
      "num_input_tokens_seen": 121515525,
      "step": 5650,
      "time_per_iteration": 2.649829149246216
    },
    {
      "auxiliary_loss_clip": 0.01170412,
      "auxiliary_loss_mlp": 0.01024423,
      "balance_loss_clip": 1.04812026,
      "balance_loss_mlp": 1.01643634,
      "epoch": 0.6794925750015031,
      "flos": 26649488246400.0,
      "grad_norm": 6.306826941711874,
      "language_loss": 0.69114089,
      "learning_rate": 9.840769580969295e-07,
      "loss": 0.71308929,
      "num_input_tokens_seen": 121535965,
      "step": 5651,
      "time_per_iteration": 2.6395323276519775
    },
    {
      "auxiliary_loss_clip": 0.01174548,
      "auxiliary_loss_mlp": 0.01027807,
      "balance_loss_clip": 1.05147862,
      "balance_loss_mlp": 1.01989746,
      "epoch": 0.6796128178921421,
      "flos": 21580374314880.0,
      "grad_norm": 2.7148035492084115,
      "language_loss": 0.80347252,
      "learning_rate": 9.834060447596114e-07,
      "loss": 0.82549608,
      "num_input_tokens_seen": 121555235,
      "step": 5652,
      "time_per_iteration": 2.6432204246520996
    },
    {
      "auxiliary_loss_clip": 0.01173833,
      "auxiliary_loss_mlp": 0.01023137,
      "balance_loss_clip": 1.0499053,
      "balance_loss_mlp": 1.01565087,
      "epoch": 0.6797330607827812,
      "flos": 22492002516480.0,
      "grad_norm": 2.0699141268203136,
      "language_loss": 0.78373921,
      "learning_rate": 9.827352856387868e-07,
      "loss": 0.80570894,
      "num_input_tokens_seen": 121574945,
      "step": 5653,
      "time_per_iteration": 2.627704381942749
    },
    {
      "auxiliary_loss_clip": 0.01073593,
      "auxiliary_loss_mlp": 0.01000583,
      "balance_loss_clip": 1.01251566,
      "balance_loss_mlp": 0.99914026,
      "epoch": 0.6798533036734203,
      "flos": 66306648286080.0,
      "grad_norm": 0.7905072065347505,
      "language_loss": 0.64203304,
      "learning_rate": 9.820646808362118e-07,
      "loss": 0.6627748,
      "num_input_tokens_seen": 121641200,
      "step": 5654,
      "time_per_iteration": 3.3309898376464844
    },
    {
      "auxiliary_loss_clip": 0.01171857,
      "auxiliary_loss_mlp": 0.01023274,
      "balance_loss_clip": 1.04900157,
      "balance_loss_mlp": 1.01518607,
      "epoch": 0.6799735465640594,
      "flos": 16180163792640.0,
      "grad_norm": 3.4704563497617835,
      "language_loss": 0.73448938,
      "learning_rate": 9.813942304536154e-07,
      "loss": 0.75644064,
      "num_input_tokens_seen": 121659170,
      "step": 5655,
      "time_per_iteration": 2.6366164684295654
    },
    {
      "auxiliary_loss_clip": 0.01172905,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.05062497,
      "balance_loss_mlp": 1.01675332,
      "epoch": 0.6800937894546984,
      "flos": 22125749489280.0,
      "grad_norm": 1.9179527093702273,
      "language_loss": 0.63956177,
      "learning_rate": 9.807239345927043e-07,
      "loss": 0.66153628,
      "num_input_tokens_seen": 121679180,
      "step": 5656,
      "time_per_iteration": 2.6553964614868164
    },
    {
      "auxiliary_loss_clip": 0.01170769,
      "auxiliary_loss_mlp": 0.01027365,
      "balance_loss_clip": 1.04808307,
      "balance_loss_mlp": 1.01975322,
      "epoch": 0.6802140323453376,
      "flos": 31612953300480.0,
      "grad_norm": 5.271454868497028,
      "language_loss": 0.71689737,
      "learning_rate": 9.80053793355162e-07,
      "loss": 0.73887873,
      "num_input_tokens_seen": 121697875,
      "step": 5657,
      "time_per_iteration": 2.7078135013580322
    },
    {
      "auxiliary_loss_clip": 0.01173902,
      "auxiliary_loss_mlp": 0.01029232,
      "balance_loss_clip": 1.049456,
      "balance_loss_mlp": 1.02082181,
      "epoch": 0.6803342752359767,
      "flos": 17712938908800.0,
      "grad_norm": 2.063446443385272,
      "language_loss": 0.75049353,
      "learning_rate": 9.793838068426472e-07,
      "loss": 0.77252483,
      "num_input_tokens_seen": 121715570,
      "step": 5658,
      "time_per_iteration": 3.546396255493164
    },
    {
      "auxiliary_loss_clip": 0.01172633,
      "auxiliary_loss_mlp": 0.01030659,
      "balance_loss_clip": 1.04973555,
      "balance_loss_mlp": 1.02264237,
      "epoch": 0.6804545181266157,
      "flos": 11326800902400.0,
      "grad_norm": 2.893899714332549,
      "language_loss": 0.60958219,
      "learning_rate": 9.78713975156799e-07,
      "loss": 0.63161516,
      "num_input_tokens_seen": 121731435,
      "step": 5659,
      "time_per_iteration": 2.6000540256500244
    },
    {
      "auxiliary_loss_clip": 0.01176251,
      "auxiliary_loss_mlp": 0.01025625,
      "balance_loss_clip": 1.05144548,
      "balance_loss_mlp": 1.01733422,
      "epoch": 0.6805747610172549,
      "flos": 29350976181120.0,
      "grad_norm": 3.3319968325372247,
      "language_loss": 0.72045422,
      "learning_rate": 9.780442983992273e-07,
      "loss": 0.74247301,
      "num_input_tokens_seen": 121749950,
      "step": 5660,
      "time_per_iteration": 2.6706953048706055
    },
    {
      "auxiliary_loss_clip": 0.01174961,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.05047679,
      "balance_loss_mlp": 1.02164543,
      "epoch": 0.680695003907894,
      "flos": 37631868612480.0,
      "grad_norm": 1.8599403358369138,
      "language_loss": 0.71577358,
      "learning_rate": 9.773747766715238e-07,
      "loss": 0.73782331,
      "num_input_tokens_seen": 121770770,
      "step": 5661,
      "time_per_iteration": 2.7886154651641846
    },
    {
      "auxiliary_loss_clip": 0.01173555,
      "auxiliary_loss_mlp": 0.01026003,
      "balance_loss_clip": 1.04956627,
      "balance_loss_mlp": 1.01805162,
      "epoch": 0.680815246798533,
      "flos": 22127365601280.0,
      "grad_norm": 1.761275624650862,
      "language_loss": 0.80001891,
      "learning_rate": 9.767054100752536e-07,
      "loss": 0.82201445,
      "num_input_tokens_seen": 121790720,
      "step": 5662,
      "time_per_iteration": 2.622438430786133
    },
    {
      "auxiliary_loss_clip": 0.01173505,
      "auxiliary_loss_mlp": 0.01025901,
      "balance_loss_clip": 1.0495894,
      "balance_loss_mlp": 1.01802671,
      "epoch": 0.6809354896891722,
      "flos": 17201822330880.0,
      "grad_norm": 2.0892346227306473,
      "language_loss": 0.81692755,
      "learning_rate": 9.760361987119584e-07,
      "loss": 0.83892155,
      "num_input_tokens_seen": 121808455,
      "step": 5663,
      "time_per_iteration": 2.653848886489868
    },
    {
      "auxiliary_loss_clip": 0.01175074,
      "auxiliary_loss_mlp": 0.01024346,
      "balance_loss_clip": 1.05021596,
      "balance_loss_mlp": 1.01606655,
      "epoch": 0.6810557325798112,
      "flos": 12458166554880.0,
      "grad_norm": 2.9207412581055374,
      "language_loss": 0.67816269,
      "learning_rate": 9.753671426831592e-07,
      "loss": 0.70015693,
      "num_input_tokens_seen": 121824470,
      "step": 5664,
      "time_per_iteration": 2.628817319869995
    },
    {
      "auxiliary_loss_clip": 0.01170853,
      "auxiliary_loss_mlp": 0.01024445,
      "balance_loss_clip": 1.04813099,
      "balance_loss_mlp": 1.0168339,
      "epoch": 0.6811759754704503,
      "flos": 22156165330560.0,
      "grad_norm": 2.122738445607996,
      "language_loss": 0.79997647,
      "learning_rate": 9.746982420903483e-07,
      "loss": 0.82192951,
      "num_input_tokens_seen": 121842665,
      "step": 5665,
      "time_per_iteration": 2.651496171951294
    },
    {
      "auxiliary_loss_clip": 0.01174451,
      "auxiliary_loss_mlp": 0.01024743,
      "balance_loss_clip": 1.05078983,
      "balance_loss_mlp": 1.01738191,
      "epoch": 0.6812962183610894,
      "flos": 17525377065600.0,
      "grad_norm": 2.8222201599027747,
      "language_loss": 0.74764919,
      "learning_rate": 9.740294970349993e-07,
      "loss": 0.76964116,
      "num_input_tokens_seen": 121859080,
      "step": 5666,
      "time_per_iteration": 3.4458322525024414
    },
    {
      "auxiliary_loss_clip": 0.01077017,
      "auxiliary_loss_mlp": 0.0100191,
      "balance_loss_clip": 1.01514721,
      "balance_loss_mlp": 1.0002768,
      "epoch": 0.6814164612517285,
      "flos": 60274480855680.0,
      "grad_norm": 0.882324554670893,
      "language_loss": 0.60941625,
      "learning_rate": 9.733609076185594e-07,
      "loss": 0.63020551,
      "num_input_tokens_seen": 121915485,
      "step": 5667,
      "time_per_iteration": 4.0919225215911865
    },
    {
      "auxiliary_loss_clip": 0.01175676,
      "auxiliary_loss_mlp": 0.01041425,
      "balance_loss_clip": 1.05178881,
      "balance_loss_mlp": 1.03287148,
      "epoch": 0.6815367041423676,
      "flos": 19317750750720.0,
      "grad_norm": 1.9220659649438183,
      "language_loss": 0.83917028,
      "learning_rate": 9.72692473942455e-07,
      "loss": 0.86134124,
      "num_input_tokens_seen": 121932710,
      "step": 5668,
      "time_per_iteration": 3.634934425354004
    },
    {
      "auxiliary_loss_clip": 0.01177069,
      "auxiliary_loss_mlp": 0.01029409,
      "balance_loss_clip": 1.05041409,
      "balance_loss_mlp": 1.02115333,
      "epoch": 0.6816569470330067,
      "flos": 22161696024960.0,
      "grad_norm": 1.7286312279624838,
      "language_loss": 0.77571195,
      "learning_rate": 9.720241961080849e-07,
      "loss": 0.79777676,
      "num_input_tokens_seen": 121952025,
      "step": 5669,
      "time_per_iteration": 2.6352241039276123
    },
    {
      "auxiliary_loss_clip": 0.01173976,
      "auxiliary_loss_mlp": 0.01022702,
      "balance_loss_clip": 1.0491612,
      "balance_loss_mlp": 1.01552248,
      "epoch": 0.6817771899236458,
      "flos": 41463501137280.0,
      "grad_norm": 2.264745727935034,
      "language_loss": 0.7347517,
      "learning_rate": 9.713560742168259e-07,
      "loss": 0.75671852,
      "num_input_tokens_seen": 121974650,
      "step": 5670,
      "time_per_iteration": 2.756688117980957
    },
    {
      "auxiliary_loss_clip": 0.01172696,
      "auxiliary_loss_mlp": 0.01023513,
      "balance_loss_clip": 1.05103326,
      "balance_loss_mlp": 1.01648235,
      "epoch": 0.6818974328142848,
      "flos": 21106138026240.0,
      "grad_norm": 2.0354649428803495,
      "language_loss": 0.7172904,
      "learning_rate": 9.706881083700333e-07,
      "loss": 0.73925245,
      "num_input_tokens_seen": 121994335,
      "step": 5671,
      "time_per_iteration": 2.658008098602295
    },
    {
      "auxiliary_loss_clip": 0.01177182,
      "auxiliary_loss_mlp": 0.0102956,
      "balance_loss_clip": 1.05069065,
      "balance_loss_mlp": 1.02078605,
      "epoch": 0.682017675704924,
      "flos": 20441897769600.0,
      "grad_norm": 2.13919510951038,
      "language_loss": 0.82646298,
      "learning_rate": 9.700202986690357e-07,
      "loss": 0.84853041,
      "num_input_tokens_seen": 122012635,
      "step": 5672,
      "time_per_iteration": 2.6391453742980957
    },
    {
      "auxiliary_loss_clip": 0.0117355,
      "auxiliary_loss_mlp": 0.01028697,
      "balance_loss_clip": 1.05012417,
      "balance_loss_mlp": 1.02093673,
      "epoch": 0.682137918595563,
      "flos": 20044438801920.0,
      "grad_norm": 3.122128663736516,
      "language_loss": 0.66561151,
      "learning_rate": 9.693526452151413e-07,
      "loss": 0.68763399,
      "num_input_tokens_seen": 122031685,
      "step": 5673,
      "time_per_iteration": 2.6443352699279785
    },
    {
      "auxiliary_loss_clip": 0.01176211,
      "auxiliary_loss_mlp": 0.01027425,
      "balance_loss_clip": 1.05093431,
      "balance_loss_mlp": 1.01903296,
      "epoch": 0.6822581614862021,
      "flos": 31684559063040.0,
      "grad_norm": 2.041220240403182,
      "language_loss": 0.7547133,
      "learning_rate": 9.686851481096305e-07,
      "loss": 0.77674961,
      "num_input_tokens_seen": 122052995,
      "step": 5674,
      "time_per_iteration": 2.6403756141662598
    },
    {
      "auxiliary_loss_clip": 0.01170083,
      "auxiliary_loss_mlp": 0.01020527,
      "balance_loss_clip": 1.04645693,
      "balance_loss_mlp": 1.01286519,
      "epoch": 0.6823784043768413,
      "flos": 23477570864640.0,
      "grad_norm": 2.0006530517227623,
      "language_loss": 0.71480554,
      "learning_rate": 9.68017807453762e-07,
      "loss": 0.73671162,
      "num_input_tokens_seen": 122071740,
      "step": 5675,
      "time_per_iteration": 2.6344168186187744
    },
    {
      "auxiliary_loss_clip": 0.01173174,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.05002058,
      "balance_loss_mlp": 1.01977396,
      "epoch": 0.6824986472674803,
      "flos": 14137134024960.0,
      "grad_norm": 1.9998298102767995,
      "language_loss": 0.73310471,
      "learning_rate": 9.673506233487721e-07,
      "loss": 0.75511396,
      "num_input_tokens_seen": 122089705,
      "step": 5676,
      "time_per_iteration": 2.5822951793670654
    },
    {
      "auxiliary_loss_clip": 0.01168417,
      "auxiliary_loss_mlp": 0.01027229,
      "balance_loss_clip": 1.04746461,
      "balance_loss_mlp": 1.01975429,
      "epoch": 0.6826188901581194,
      "flos": 21504997624320.0,
      "grad_norm": 2.123438861166877,
      "language_loss": 0.86175555,
      "learning_rate": 9.666835958958717e-07,
      "loss": 0.88371205,
      "num_input_tokens_seen": 122109025,
      "step": 5677,
      "time_per_iteration": 2.6329407691955566
    },
    {
      "auxiliary_loss_clip": 0.01173244,
      "auxiliary_loss_mlp": 0.01026193,
      "balance_loss_clip": 1.05177784,
      "balance_loss_mlp": 1.01845598,
      "epoch": 0.6827391330487584,
      "flos": 20810126044800.0,
      "grad_norm": 3.1839641894720048,
      "language_loss": 0.80749655,
      "learning_rate": 9.660167251962484e-07,
      "loss": 0.82949096,
      "num_input_tokens_seen": 122127385,
      "step": 5678,
      "time_per_iteration": 2.6492788791656494
    },
    {
      "auxiliary_loss_clip": 0.01171574,
      "auxiliary_loss_mlp": 0.01025655,
      "balance_loss_clip": 1.04990292,
      "balance_loss_mlp": 1.01772761,
      "epoch": 0.6828593759393976,
      "flos": 21688788539520.0,
      "grad_norm": 1.6964490511319101,
      "language_loss": 0.78069395,
      "learning_rate": 9.653500113510654e-07,
      "loss": 0.80266625,
      "num_input_tokens_seen": 122146500,
      "step": 5679,
      "time_per_iteration": 2.672457695007324
    },
    {
      "auxiliary_loss_clip": 0.01173506,
      "auxiliary_loss_mlp": 0.01032532,
      "balance_loss_clip": 1.0506115,
      "balance_loss_mlp": 1.02495074,
      "epoch": 0.6829796188300367,
      "flos": 25337707557120.0,
      "grad_norm": 5.496116593339279,
      "language_loss": 0.67515898,
      "learning_rate": 9.646834544614627e-07,
      "loss": 0.69721931,
      "num_input_tokens_seen": 122167000,
      "step": 5680,
      "time_per_iteration": 2.7087347507476807
    },
    {
      "auxiliary_loss_clip": 0.01174068,
      "auxiliary_loss_mlp": 0.01024905,
      "balance_loss_clip": 1.05188656,
      "balance_loss_mlp": 1.01700759,
      "epoch": 0.6830998617206757,
      "flos": 20704800389760.0,
      "grad_norm": 4.315264556422335,
      "language_loss": 0.76075238,
      "learning_rate": 9.64017054628558e-07,
      "loss": 0.78274214,
      "num_input_tokens_seen": 122185825,
      "step": 5681,
      "time_per_iteration": 2.6024224758148193
    },
    {
      "auxiliary_loss_clip": 0.0117041,
      "auxiliary_loss_mlp": 0.01026531,
      "balance_loss_clip": 1.04609394,
      "balance_loss_mlp": 1.01840687,
      "epoch": 0.6832201046113149,
      "flos": 21726638496000.0,
      "grad_norm": 5.352985570952934,
      "language_loss": 0.79412925,
      "learning_rate": 9.63350811953441e-07,
      "loss": 0.81609869,
      "num_input_tokens_seen": 122206200,
      "step": 5682,
      "time_per_iteration": 2.8360660076141357
    },
    {
      "auxiliary_loss_clip": 0.01173017,
      "auxiliary_loss_mlp": 0.01026348,
      "balance_loss_clip": 1.04912853,
      "balance_loss_mlp": 1.01832843,
      "epoch": 0.6833403475019539,
      "flos": 19536554448000.0,
      "grad_norm": 88.79365828986664,
      "language_loss": 0.70607114,
      "learning_rate": 9.626847265371826e-07,
      "loss": 0.72806478,
      "num_input_tokens_seen": 122225520,
      "step": 5683,
      "time_per_iteration": 2.642153024673462
    },
    {
      "auxiliary_loss_clip": 0.01170355,
      "auxiliary_loss_mlp": 0.01026565,
      "balance_loss_clip": 1.04744315,
      "balance_loss_mlp": 1.01914406,
      "epoch": 0.683460590392593,
      "flos": 19352153001600.0,
      "grad_norm": 2.41239719712045,
      "language_loss": 0.78920317,
      "learning_rate": 9.620187984808262e-07,
      "loss": 0.81117237,
      "num_input_tokens_seen": 122244320,
      "step": 5684,
      "time_per_iteration": 3.603663206100464
    },
    {
      "auxiliary_loss_clip": 0.01172182,
      "auxiliary_loss_mlp": 0.0103125,
      "balance_loss_clip": 1.04884839,
      "balance_loss_mlp": 1.02298892,
      "epoch": 0.6835808332832322,
      "flos": 23288500650240.0,
      "grad_norm": 3.9535722405698257,
      "language_loss": 0.85866392,
      "learning_rate": 9.613530278853919e-07,
      "loss": 0.88069826,
      "num_input_tokens_seen": 122264295,
      "step": 5685,
      "time_per_iteration": 2.725029230117798
    },
    {
      "auxiliary_loss_clip": 0.01175019,
      "auxiliary_loss_mlp": 0.01027086,
      "balance_loss_clip": 1.0495944,
      "balance_loss_mlp": 1.01926005,
      "epoch": 0.6837010761738712,
      "flos": 21653416621440.0,
      "grad_norm": 2.4176475124622963,
      "language_loss": 0.7424857,
      "learning_rate": 9.60687414851879e-07,
      "loss": 0.76450676,
      "num_input_tokens_seen": 122285300,
      "step": 5686,
      "time_per_iteration": 2.833343029022217
    },
    {
      "auxiliary_loss_clip": 0.0117342,
      "auxiliary_loss_mlp": 0.01024817,
      "balance_loss_clip": 1.04855192,
      "balance_loss_mlp": 1.0167768,
      "epoch": 0.6838213190645103,
      "flos": 17566387418880.0,
      "grad_norm": 2.5461657222017275,
      "language_loss": 0.77720284,
      "learning_rate": 9.600219594812575e-07,
      "loss": 0.79918522,
      "num_input_tokens_seen": 122303240,
      "step": 5687,
      "time_per_iteration": 2.7061562538146973
    },
    {
      "auxiliary_loss_clip": 0.01171681,
      "auxiliary_loss_mlp": 0.01021698,
      "balance_loss_clip": 1.04939079,
      "balance_loss_mlp": 1.01429176,
      "epoch": 0.6839415619551494,
      "flos": 23112538899840.0,
      "grad_norm": 1.904924184041145,
      "language_loss": 0.72952235,
      "learning_rate": 9.593566618744786e-07,
      "loss": 0.75145614,
      "num_input_tokens_seen": 122323390,
      "step": 5688,
      "time_per_iteration": 2.662832736968994
    },
    {
      "auxiliary_loss_clip": 0.01173482,
      "auxiliary_loss_mlp": 0.01026518,
      "balance_loss_clip": 1.04986835,
      "balance_loss_mlp": 1.01897168,
      "epoch": 0.6840618048457885,
      "flos": 22127868391680.0,
      "grad_norm": 3.8599822588324715,
      "language_loss": 0.73877329,
      "learning_rate": 9.58691522132466e-07,
      "loss": 0.7607733,
      "num_input_tokens_seen": 122342200,
      "step": 5689,
      "time_per_iteration": 2.68371844291687
    },
    {
      "auxiliary_loss_clip": 0.011776,
      "auxiliary_loss_mlp": 0.01030454,
      "balance_loss_clip": 1.05095625,
      "balance_loss_mlp": 1.02232981,
      "epoch": 0.6841820477364275,
      "flos": 22015898720640.0,
      "grad_norm": 3.1014228434090745,
      "language_loss": 0.84915787,
      "learning_rate": 9.58026540356123e-07,
      "loss": 0.87123835,
      "num_input_tokens_seen": 122360465,
      "step": 5690,
      "time_per_iteration": 2.612520933151245
    },
    {
      "auxiliary_loss_clip": 0.01173999,
      "auxiliary_loss_mlp": 0.01031026,
      "balance_loss_clip": 1.05043364,
      "balance_loss_mlp": 1.02280402,
      "epoch": 0.6843022906270667,
      "flos": 24900531125760.0,
      "grad_norm": 2.292811834698199,
      "language_loss": 0.86810237,
      "learning_rate": 9.573617166463246e-07,
      "loss": 0.89015263,
      "num_input_tokens_seen": 122381680,
      "step": 5691,
      "time_per_iteration": 2.7285614013671875
    },
    {
      "auxiliary_loss_clip": 0.01172179,
      "auxiliary_loss_mlp": 0.01030932,
      "balance_loss_clip": 1.04832172,
      "balance_loss_mlp": 1.02277863,
      "epoch": 0.6844225335177058,
      "flos": 19969924037760.0,
      "grad_norm": 2.120434409506327,
      "language_loss": 0.60617173,
      "learning_rate": 9.56697051103924e-07,
      "loss": 0.6282028,
      "num_input_tokens_seen": 122399120,
      "step": 5692,
      "time_per_iteration": 2.625521659851074
    },
    {
      "auxiliary_loss_clip": 0.01169217,
      "auxiliary_loss_mlp": 0.01023472,
      "balance_loss_clip": 1.04807043,
      "balance_loss_mlp": 1.01628387,
      "epoch": 0.6845427764083448,
      "flos": 25883334126720.0,
      "grad_norm": 2.129542552995289,
      "language_loss": 0.81083953,
      "learning_rate": 9.560325438297522e-07,
      "loss": 0.83276635,
      "num_input_tokens_seen": 122417430,
      "step": 5693,
      "time_per_iteration": 3.514620304107666
    },
    {
      "auxiliary_loss_clip": 0.01174201,
      "auxiliary_loss_mlp": 0.01024158,
      "balance_loss_clip": 1.05074763,
      "balance_loss_mlp": 1.01712477,
      "epoch": 0.684663019298984,
      "flos": 18880143356160.0,
      "grad_norm": 2.4913554027895533,
      "language_loss": 0.86788523,
      "learning_rate": 9.553681949246127e-07,
      "loss": 0.8898688,
      "num_input_tokens_seen": 122435055,
      "step": 5694,
      "time_per_iteration": 3.6335837841033936
    },
    {
      "auxiliary_loss_clip": 0.01177803,
      "auxiliary_loss_mlp": 0.01027096,
      "balance_loss_clip": 1.0521754,
      "balance_loss_mlp": 1.01792896,
      "epoch": 0.684783262189623,
      "flos": 54193725302400.0,
      "grad_norm": 2.575392953479529,
      "language_loss": 0.75414968,
      "learning_rate": 9.547040044892886e-07,
      "loss": 0.77619869,
      "num_input_tokens_seen": 122462570,
      "step": 5695,
      "time_per_iteration": 2.9170053005218506
    },
    {
      "auxiliary_loss_clip": 0.01074665,
      "auxiliary_loss_mlp": 0.01002564,
      "balance_loss_clip": 1.01419032,
      "balance_loss_mlp": 1.00126445,
      "epoch": 0.6849035050802621,
      "flos": 63970264143360.0,
      "grad_norm": 0.8787982477963039,
      "language_loss": 0.6006816,
      "learning_rate": 9.540399726245354e-07,
      "loss": 0.62145388,
      "num_input_tokens_seen": 122519275,
      "step": 5696,
      "time_per_iteration": 3.0685393810272217
    },
    {
      "auxiliary_loss_clip": 0.01176729,
      "auxiliary_loss_mlp": 0.01027335,
      "balance_loss_clip": 1.05042148,
      "balance_loss_mlp": 1.01962805,
      "epoch": 0.6850237479709013,
      "flos": 25224121774080.0,
      "grad_norm": 2.7724984863053126,
      "language_loss": 0.68968689,
      "learning_rate": 9.533760994310859e-07,
      "loss": 0.7117275,
      "num_input_tokens_seen": 122539675,
      "step": 5697,
      "time_per_iteration": 2.6600818634033203
    },
    {
      "auxiliary_loss_clip": 0.01175043,
      "auxiliary_loss_mlp": 0.01028822,
      "balance_loss_clip": 1.05029535,
      "balance_loss_mlp": 1.02052546,
      "epoch": 0.6851439908615403,
      "flos": 19354128249600.0,
      "grad_norm": 2.0304777737975206,
      "language_loss": 0.74834335,
      "learning_rate": 9.527123850096508e-07,
      "loss": 0.77038199,
      "num_input_tokens_seen": 122558035,
      "step": 5698,
      "time_per_iteration": 2.6193294525146484
    },
    {
      "auxiliary_loss_clip": 0.01171965,
      "auxiliary_loss_mlp": 0.01023051,
      "balance_loss_clip": 1.04874301,
      "balance_loss_mlp": 1.01538312,
      "epoch": 0.6852642337521794,
      "flos": 23182133500800.0,
      "grad_norm": 2.0393788282099887,
      "language_loss": 0.72022462,
      "learning_rate": 9.520488294609142e-07,
      "loss": 0.74217474,
      "num_input_tokens_seen": 122576815,
      "step": 5699,
      "time_per_iteration": 2.6144614219665527
    },
    {
      "auxiliary_loss_clip": 0.01074375,
      "auxiliary_loss_mlp": 0.01002267,
      "balance_loss_clip": 1.01432443,
      "balance_loss_mlp": 1.0009799,
      "epoch": 0.6853844766428185,
      "flos": 62647206583680.0,
      "grad_norm": 0.7881944995346938,
      "language_loss": 0.53815246,
      "learning_rate": 9.513854328855368e-07,
      "loss": 0.55891883,
      "num_input_tokens_seen": 122634690,
      "step": 5700,
      "time_per_iteration": 3.131068468093872
    },
    {
      "auxiliary_loss_clip": 0.0116854,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.0473597,
      "balance_loss_mlp": 1.01687551,
      "epoch": 0.6855047195334576,
      "flos": 23437242869760.0,
      "grad_norm": 2.4930130997375404,
      "language_loss": 0.81193572,
      "learning_rate": 9.507221953841558e-07,
      "loss": 0.83386666,
      "num_input_tokens_seen": 122652320,
      "step": 5701,
      "time_per_iteration": 2.6392555236816406
    },
    {
      "auxiliary_loss_clip": 0.01176707,
      "auxiliary_loss_mlp": 0.01028137,
      "balance_loss_clip": 1.05208993,
      "balance_loss_mlp": 1.0201143,
      "epoch": 0.6856249624240967,
      "flos": 20664831530880.0,
      "grad_norm": 1.6414350288753605,
      "language_loss": 0.77967983,
      "learning_rate": 9.500591170573824e-07,
      "loss": 0.80172825,
      "num_input_tokens_seen": 122672340,
      "step": 5702,
      "time_per_iteration": 2.643770456314087
    },
    {
      "auxiliary_loss_clip": 0.0117137,
      "auxiliary_loss_mlp": 0.01027847,
      "balance_loss_clip": 1.04867697,
      "balance_loss_mlp": 1.01989317,
      "epoch": 0.6857452053147358,
      "flos": 17087302794240.0,
      "grad_norm": 2.579247136807774,
      "language_loss": 0.74346811,
      "learning_rate": 9.493961980058078e-07,
      "loss": 0.76546025,
      "num_input_tokens_seen": 122689935,
      "step": 5703,
      "time_per_iteration": 2.5881075859069824
    },
    {
      "auxiliary_loss_clip": 0.01170585,
      "auxiliary_loss_mlp": 0.01026511,
      "balance_loss_clip": 1.04778028,
      "balance_loss_mlp": 1.01928723,
      "epoch": 0.6858654482053749,
      "flos": 30847266057600.0,
      "grad_norm": 2.3141395641929994,
      "language_loss": 0.67580372,
      "learning_rate": 9.48733438329993e-07,
      "loss": 0.69777471,
      "num_input_tokens_seen": 122710200,
      "step": 5704,
      "time_per_iteration": 2.671708822250366
    },
    {
      "auxiliary_loss_clip": 0.01170424,
      "auxiliary_loss_mlp": 0.01026134,
      "balance_loss_clip": 1.04958522,
      "balance_loss_mlp": 1.01855826,
      "epoch": 0.6859856910960139,
      "flos": 28877314510080.0,
      "grad_norm": 3.338475414974976,
      "language_loss": 0.74425888,
      "learning_rate": 9.480708381304807e-07,
      "loss": 0.7662245,
      "num_input_tokens_seen": 122731495,
      "step": 5705,
      "time_per_iteration": 2.6893630027770996
    },
    {
      "auxiliary_loss_clip": 0.01174316,
      "auxiliary_loss_mlp": 0.01024886,
      "balance_loss_clip": 1.05089164,
      "balance_loss_mlp": 1.01694083,
      "epoch": 0.6861059339866531,
      "flos": 19354523299200.0,
      "grad_norm": 2.3235979530466264,
      "language_loss": 0.83670318,
      "learning_rate": 9.474083975077858e-07,
      "loss": 0.85869521,
      "num_input_tokens_seen": 122748620,
      "step": 5706,
      "time_per_iteration": 2.551306962966919
    },
    {
      "auxiliary_loss_clip": 0.01174256,
      "auxiliary_loss_mlp": 0.01028037,
      "balance_loss_clip": 1.05095816,
      "balance_loss_mlp": 1.02035975,
      "epoch": 0.6862261768772921,
      "flos": 22199976944640.0,
      "grad_norm": 2.2243302489374694,
      "language_loss": 0.80292034,
      "learning_rate": 9.467461165623994e-07,
      "loss": 0.82494324,
      "num_input_tokens_seen": 122767670,
      "step": 5707,
      "time_per_iteration": 2.668081760406494
    },
    {
      "auxiliary_loss_clip": 0.01172956,
      "auxiliary_loss_mlp": 0.01022605,
      "balance_loss_clip": 1.04874921,
      "balance_loss_mlp": 1.01480305,
      "epoch": 0.6863464197679312,
      "flos": 26285677344000.0,
      "grad_norm": 2.0765553989778036,
      "language_loss": 0.79172105,
      "learning_rate": 9.46083995394791e-07,
      "loss": 0.81367666,
      "num_input_tokens_seen": 122785480,
      "step": 5708,
      "time_per_iteration": 2.6628427505493164
    },
    {
      "auxiliary_loss_clip": 0.01170271,
      "auxiliary_loss_mlp": 0.01022112,
      "balance_loss_clip": 1.0484091,
      "balance_loss_mlp": 1.01466191,
      "epoch": 0.6864666626585703,
      "flos": 37815228564480.0,
      "grad_norm": 4.820353470567534,
      "language_loss": 0.63428319,
      "learning_rate": 9.454220341054012e-07,
      "loss": 0.65620703,
      "num_input_tokens_seen": 122810265,
      "step": 5709,
      "time_per_iteration": 2.8057050704956055
    },
    {
      "auxiliary_loss_clip": 0.01173983,
      "auxiliary_loss_mlp": 0.01024929,
      "balance_loss_clip": 1.0508678,
      "balance_loss_mlp": 1.01711452,
      "epoch": 0.6865869055492094,
      "flos": 19391152193280.0,
      "grad_norm": 2.397048432149257,
      "language_loss": 0.80426705,
      "learning_rate": 9.447602327946512e-07,
      "loss": 0.82625616,
      "num_input_tokens_seen": 122828905,
      "step": 5710,
      "time_per_iteration": 2.6520333290100098
    },
    {
      "auxiliary_loss_clip": 0.01175671,
      "auxiliary_loss_mlp": 0.01027094,
      "balance_loss_clip": 1.0507251,
      "balance_loss_mlp": 1.01872253,
      "epoch": 0.6867071484398485,
      "flos": 20375966355840.0,
      "grad_norm": 1.9407419639124328,
      "language_loss": 0.76572239,
      "learning_rate": 9.440985915629338e-07,
      "loss": 0.78775001,
      "num_input_tokens_seen": 122846235,
      "step": 5711,
      "time_per_iteration": 3.528663396835327
    },
    {
      "auxiliary_loss_clip": 0.01172155,
      "auxiliary_loss_mlp": 0.01027806,
      "balance_loss_clip": 1.0506494,
      "balance_loss_mlp": 1.02003312,
      "epoch": 0.6868273913304875,
      "flos": 15889143801600.0,
      "grad_norm": 2.0747634825946943,
      "language_loss": 0.73097837,
      "learning_rate": 9.434371105106223e-07,
      "loss": 0.75297797,
      "num_input_tokens_seen": 122863835,
      "step": 5712,
      "time_per_iteration": 2.6636276245117188
    },
    {
      "auxiliary_loss_clip": 0.01174619,
      "auxiliary_loss_mlp": 0.01025701,
      "balance_loss_clip": 1.05034161,
      "balance_loss_mlp": 1.01761246,
      "epoch": 0.6869476342211267,
      "flos": 24462492768000.0,
      "grad_norm": 2.1944049482507886,
      "language_loss": 0.7048561,
      "learning_rate": 9.427757897380602e-07,
      "loss": 0.72685927,
      "num_input_tokens_seen": 122883235,
      "step": 5713,
      "time_per_iteration": 2.6801095008850098
    },
    {
      "auxiliary_loss_clip": 0.01173582,
      "auxiliary_loss_mlp": 0.01029017,
      "balance_loss_clip": 1.04874718,
      "balance_loss_mlp": 1.02106249,
      "epoch": 0.6870678771117658,
      "flos": 18442571875200.0,
      "grad_norm": 2.4251044795829637,
      "language_loss": 0.85024178,
      "learning_rate": 9.421146293455695e-07,
      "loss": 0.87226772,
      "num_input_tokens_seen": 122898975,
      "step": 5714,
      "time_per_iteration": 2.735743522644043
    },
    {
      "auxiliary_loss_clip": 0.01175616,
      "auxiliary_loss_mlp": 0.01026074,
      "balance_loss_clip": 1.05240679,
      "balance_loss_mlp": 1.01782489,
      "epoch": 0.6871881200024048,
      "flos": 22200371994240.0,
      "grad_norm": 3.3494542329973376,
      "language_loss": 0.6864087,
      "learning_rate": 9.414536294334489e-07,
      "loss": 0.70842558,
      "num_input_tokens_seen": 122918995,
      "step": 5715,
      "time_per_iteration": 2.6342873573303223
    },
    {
      "auxiliary_loss_clip": 0.01172406,
      "auxiliary_loss_mlp": 0.01024238,
      "balance_loss_clip": 1.0479126,
      "balance_loss_mlp": 1.0164957,
      "epoch": 0.687308362893044,
      "flos": 22127724737280.0,
      "grad_norm": 1.82834833309212,
      "language_loss": 0.69524336,
      "learning_rate": 9.407927901019708e-07,
      "loss": 0.71720982,
      "num_input_tokens_seen": 122938125,
      "step": 5716,
      "time_per_iteration": 2.6565558910369873
    },
    {
      "auxiliary_loss_clip": 0.01170673,
      "auxiliary_loss_mlp": 0.01023261,
      "balance_loss_clip": 1.04822683,
      "balance_loss_mlp": 1.01577222,
      "epoch": 0.687428605783683,
      "flos": 25040546340480.0,
      "grad_norm": 2.624911999286854,
      "language_loss": 0.77056628,
      "learning_rate": 9.401321114513854e-07,
      "loss": 0.79250562,
      "num_input_tokens_seen": 122957020,
      "step": 5717,
      "time_per_iteration": 2.6320064067840576
    },
    {
      "auxiliary_loss_clip": 0.01174773,
      "auxiliary_loss_mlp": 0.01030846,
      "balance_loss_clip": 1.05075717,
      "balance_loss_mlp": 1.02204847,
      "epoch": 0.6875488486743221,
      "flos": 23770063313280.0,
      "grad_norm": 1.876192448153004,
      "language_loss": 0.75481987,
      "learning_rate": 9.394715935819155e-07,
      "loss": 0.77687609,
      "num_input_tokens_seen": 122977410,
      "step": 5718,
      "time_per_iteration": 2.6583096981048584
    },
    {
      "auxiliary_loss_clip": 0.01173338,
      "auxiliary_loss_mlp": 0.01026305,
      "balance_loss_clip": 1.04999697,
      "balance_loss_mlp": 1.01781154,
      "epoch": 0.6876690915649613,
      "flos": 25516937445120.0,
      "grad_norm": 3.445497187463424,
      "language_loss": 0.62515378,
      "learning_rate": 9.388112365937608e-07,
      "loss": 0.64715016,
      "num_input_tokens_seen": 122996875,
      "step": 5719,
      "time_per_iteration": 3.580037832260132
    },
    {
      "auxiliary_loss_clip": 0.01176503,
      "auxiliary_loss_mlp": 0.01026845,
      "balance_loss_clip": 1.05119491,
      "balance_loss_mlp": 1.01897693,
      "epoch": 0.6877893344556003,
      "flos": 19427996568960.0,
      "grad_norm": 2.3434469482707443,
      "language_loss": 0.834481,
      "learning_rate": 9.381510405870985e-07,
      "loss": 0.85651451,
      "num_input_tokens_seen": 123015890,
      "step": 5720,
      "time_per_iteration": 2.6210360527038574
    },
    {
      "auxiliary_loss_clip": 0.01174427,
      "auxiliary_loss_mlp": 0.010274,
      "balance_loss_clip": 1.05036569,
      "balance_loss_mlp": 1.01913261,
      "epoch": 0.6879095773462394,
      "flos": 18661303745280.0,
      "grad_norm": 2.629788899338599,
      "language_loss": 0.77275372,
      "learning_rate": 9.374910056620791e-07,
      "loss": 0.79477197,
      "num_input_tokens_seen": 123034955,
      "step": 5721,
      "time_per_iteration": 3.650524854660034
    },
    {
      "auxiliary_loss_clip": 0.01175417,
      "auxiliary_loss_mlp": 0.01029325,
      "balance_loss_clip": 1.05037034,
      "balance_loss_mlp": 1.02151668,
      "epoch": 0.6880298202368785,
      "flos": 20883132437760.0,
      "grad_norm": 2.389398890085589,
      "language_loss": 0.80948699,
      "learning_rate": 9.368311319188293e-07,
      "loss": 0.83153439,
      "num_input_tokens_seen": 123052770,
      "step": 5722,
      "time_per_iteration": 2.598024845123291
    },
    {
      "auxiliary_loss_clip": 0.0117172,
      "auxiliary_loss_mlp": 0.01026941,
      "balance_loss_clip": 1.04848325,
      "balance_loss_mlp": 1.01921034,
      "epoch": 0.6881500631275176,
      "flos": 30153292318080.0,
      "grad_norm": 3.256820375568688,
      "language_loss": 0.79447544,
      "learning_rate": 9.361714194574515e-07,
      "loss": 0.81646204,
      "num_input_tokens_seen": 123075105,
      "step": 5723,
      "time_per_iteration": 2.698824644088745
    },
    {
      "auxiliary_loss_clip": 0.01073307,
      "auxiliary_loss_mlp": 0.01001276,
      "balance_loss_clip": 1.0135951,
      "balance_loss_mlp": 1.00007188,
      "epoch": 0.6882703060181566,
      "flos": 66181537215360.0,
      "grad_norm": 0.7375393469563665,
      "language_loss": 0.58325034,
      "learning_rate": 9.355118683780228e-07,
      "loss": 0.60399616,
      "num_input_tokens_seen": 123145175,
      "step": 5724,
      "time_per_iteration": 3.3263986110687256
    },
    {
      "auxiliary_loss_clip": 0.0116992,
      "auxiliary_loss_mlp": 0.01027599,
      "balance_loss_clip": 1.0467422,
      "balance_loss_mlp": 1.01983845,
      "epoch": 0.6883905489087958,
      "flos": 18214646123520.0,
      "grad_norm": 3.4132545318501646,
      "language_loss": 0.79352367,
      "learning_rate": 9.348524787805987e-07,
      "loss": 0.81549883,
      "num_input_tokens_seen": 123160365,
      "step": 5725,
      "time_per_iteration": 2.598545789718628
    },
    {
      "auxiliary_loss_clip": 0.01172337,
      "auxiliary_loss_mlp": 0.01025496,
      "balance_loss_clip": 1.04860735,
      "balance_loss_mlp": 1.01739597,
      "epoch": 0.6885107917994349,
      "flos": 14056262553600.0,
      "grad_norm": 5.913710676749295,
      "language_loss": 0.85298705,
      "learning_rate": 9.341932507652053e-07,
      "loss": 0.87496537,
      "num_input_tokens_seen": 123174855,
      "step": 5726,
      "time_per_iteration": 2.5658810138702393
    },
    {
      "auxiliary_loss_clip": 0.01173666,
      "auxiliary_loss_mlp": 0.01023808,
      "balance_loss_clip": 1.0489316,
      "balance_loss_mlp": 1.01616061,
      "epoch": 0.6886310346900739,
      "flos": 28690722334080.0,
      "grad_norm": 1.9050042201947113,
      "language_loss": 0.78986919,
      "learning_rate": 9.335341844318489e-07,
      "loss": 0.81184387,
      "num_input_tokens_seen": 123194995,
      "step": 5727,
      "time_per_iteration": 2.696760654449463
    },
    {
      "auxiliary_loss_clip": 0.01172245,
      "auxiliary_loss_mlp": 0.01023708,
      "balance_loss_clip": 1.04900599,
      "balance_loss_mlp": 1.01603961,
      "epoch": 0.6887512775807131,
      "flos": 24535319592960.0,
      "grad_norm": 9.39493471519886,
      "language_loss": 0.73218024,
      "learning_rate": 9.328752798805091e-07,
      "loss": 0.75413978,
      "num_input_tokens_seen": 123213465,
      "step": 5728,
      "time_per_iteration": 2.671658515930176
    },
    {
      "auxiliary_loss_clip": 0.01174475,
      "auxiliary_loss_mlp": 0.010279,
      "balance_loss_clip": 1.05055428,
      "balance_loss_mlp": 1.01977634,
      "epoch": 0.6888715204713521,
      "flos": 22414363269120.0,
      "grad_norm": 2.1448446758325197,
      "language_loss": 0.7656498,
      "learning_rate": 9.322165372111399e-07,
      "loss": 0.78767347,
      "num_input_tokens_seen": 123231610,
      "step": 5729,
      "time_per_iteration": 2.651212453842163
    },
    {
      "auxiliary_loss_clip": 0.01170304,
      "auxiliary_loss_mlp": 0.01025849,
      "balance_loss_clip": 1.0479759,
      "balance_loss_mlp": 1.01784956,
      "epoch": 0.6889917633619912,
      "flos": 22054323294720.0,
      "grad_norm": 2.115507317834385,
      "language_loss": 0.75323397,
      "learning_rate": 9.315579565236747e-07,
      "loss": 0.77519548,
      "num_input_tokens_seen": 123250715,
      "step": 5730,
      "time_per_iteration": 2.6429648399353027
    },
    {
      "auxiliary_loss_clip": 0.01172747,
      "auxiliary_loss_mlp": 0.01027984,
      "balance_loss_clip": 1.04944611,
      "balance_loss_mlp": 1.01988935,
      "epoch": 0.6891120062526304,
      "flos": 23949724164480.0,
      "grad_norm": 1.9505139382159313,
      "language_loss": 0.74383748,
      "learning_rate": 9.308995379180162e-07,
      "loss": 0.7658447,
      "num_input_tokens_seen": 123270270,
      "step": 5731,
      "time_per_iteration": 2.671938896179199
    },
    {
      "auxiliary_loss_clip": 0.01072142,
      "auxiliary_loss_mlp": 0.0100195,
      "balance_loss_clip": 1.01271462,
      "balance_loss_mlp": 1.00072765,
      "epoch": 0.6892322491432694,
      "flos": 64117354337280.0,
      "grad_norm": 0.8632800751182471,
      "language_loss": 0.59514642,
      "learning_rate": 9.302412814940488e-07,
      "loss": 0.61588728,
      "num_input_tokens_seen": 123333045,
      "step": 5732,
      "time_per_iteration": 3.315345048904419
    },
    {
      "auxiliary_loss_clip": 0.011748,
      "auxiliary_loss_mlp": 0.01032156,
      "balance_loss_clip": 1.05053687,
      "balance_loss_mlp": 1.02360225,
      "epoch": 0.6893524920339085,
      "flos": 23002436736000.0,
      "grad_norm": 10.709634451533471,
      "language_loss": 0.70875221,
      "learning_rate": 9.295831873516276e-07,
      "loss": 0.73082179,
      "num_input_tokens_seen": 123352320,
      "step": 5733,
      "time_per_iteration": 2.604952573776245
    },
    {
      "auxiliary_loss_clip": 0.01173781,
      "auxiliary_loss_mlp": 0.01022009,
      "balance_loss_clip": 1.05136347,
      "balance_loss_mlp": 1.0146482,
      "epoch": 0.6894727349245476,
      "flos": 21396260177280.0,
      "grad_norm": 1.9264648822172976,
      "language_loss": 0.76190686,
      "learning_rate": 9.289252555905873e-07,
      "loss": 0.78386474,
      "num_input_tokens_seen": 123372400,
      "step": 5734,
      "time_per_iteration": 2.592771291732788
    },
    {
      "auxiliary_loss_clip": 0.01176144,
      "auxiliary_loss_mlp": 0.01025481,
      "balance_loss_clip": 1.05314159,
      "balance_loss_mlp": 1.0176549,
      "epoch": 0.6895929778151867,
      "flos": 19865316654720.0,
      "grad_norm": 2.1884219405123266,
      "language_loss": 0.76241493,
      "learning_rate": 9.282674863107334e-07,
      "loss": 0.78443122,
      "num_input_tokens_seen": 123390215,
      "step": 5735,
      "time_per_iteration": 2.549870014190674
    },
    {
      "auxiliary_loss_clip": 0.01172547,
      "auxiliary_loss_mlp": 0.01026789,
      "balance_loss_clip": 1.05015039,
      "balance_loss_mlp": 1.01902211,
      "epoch": 0.6897132207058257,
      "flos": 18179166464640.0,
      "grad_norm": 2.735099388752326,
      "language_loss": 0.76044399,
      "learning_rate": 9.276098796118488e-07,
      "loss": 0.78243738,
      "num_input_tokens_seen": 123406700,
      "step": 5736,
      "time_per_iteration": 2.538939952850342
    },
    {
      "auxiliary_loss_clip": 0.01173127,
      "auxiliary_loss_mlp": 0.0102496,
      "balance_loss_clip": 1.04989707,
      "balance_loss_mlp": 1.01799214,
      "epoch": 0.6898334635964649,
      "flos": 32561641359360.0,
      "grad_norm": 1.8536469088574943,
      "language_loss": 0.66136855,
      "learning_rate": 9.269524355936938e-07,
      "loss": 0.68334949,
      "num_input_tokens_seen": 123429880,
      "step": 5737,
      "time_per_iteration": 3.4749598503112793
    },
    {
      "auxiliary_loss_clip": 0.01168333,
      "auxiliary_loss_mlp": 0.0102421,
      "balance_loss_clip": 1.04771733,
      "balance_loss_mlp": 1.01733172,
      "epoch": 0.689953706487104,
      "flos": 22819004956800.0,
      "grad_norm": 4.014545524244438,
      "language_loss": 0.85014063,
      "learning_rate": 9.262951543560002e-07,
      "loss": 0.87206602,
      "num_input_tokens_seen": 123449105,
      "step": 5738,
      "time_per_iteration": 2.46004581451416
    },
    {
      "auxiliary_loss_clip": 0.01176299,
      "auxiliary_loss_mlp": 0.01026262,
      "balance_loss_clip": 1.05049634,
      "balance_loss_mlp": 1.01872253,
      "epoch": 0.690073949377743,
      "flos": 18515362786560.0,
      "grad_norm": 2.180570891304872,
      "language_loss": 0.86640549,
      "learning_rate": 9.256380359984795e-07,
      "loss": 0.88843113,
      "num_input_tokens_seen": 123466215,
      "step": 5739,
      "time_per_iteration": 2.6102147102355957
    },
    {
      "auxiliary_loss_clip": 0.01172386,
      "auxiliary_loss_mlp": 0.01027569,
      "balance_loss_clip": 1.04905045,
      "balance_loss_mlp": 1.01989222,
      "epoch": 0.6901941922683821,
      "flos": 34857194716800.0,
      "grad_norm": 1.8673059111278603,
      "language_loss": 0.7457366,
      "learning_rate": 9.249810806208139e-07,
      "loss": 0.76773614,
      "num_input_tokens_seen": 123485480,
      "step": 5740,
      "time_per_iteration": 2.6562182903289795
    },
    {
      "auxiliary_loss_clip": 0.01170835,
      "auxiliary_loss_mlp": 0.01025791,
      "balance_loss_clip": 1.04853439,
      "balance_loss_mlp": 1.01856136,
      "epoch": 0.6903144351590212,
      "flos": 16253672976000.0,
      "grad_norm": 2.4416169615486996,
      "language_loss": 0.8030802,
      "learning_rate": 9.243242883226627e-07,
      "loss": 0.82504642,
      "num_input_tokens_seen": 123504575,
      "step": 5741,
      "time_per_iteration": 2.557037830352783
    },
    {
      "auxiliary_loss_clip": 0.01173271,
      "auxiliary_loss_mlp": 0.01027113,
      "balance_loss_clip": 1.04811144,
      "balance_loss_mlp": 1.01875019,
      "epoch": 0.6904346780496603,
      "flos": 28035137255040.0,
      "grad_norm": 1.8697765897246257,
      "language_loss": 0.70146203,
      "learning_rate": 9.236676592036628e-07,
      "loss": 0.7234658,
      "num_input_tokens_seen": 123524250,
      "step": 5742,
      "time_per_iteration": 2.665703773498535
    },
    {
      "auxiliary_loss_clip": 0.01175287,
      "auxiliary_loss_mlp": 0.01026909,
      "balance_loss_clip": 1.05268574,
      "balance_loss_mlp": 1.01875508,
      "epoch": 0.6905549209402994,
      "flos": 23624266008960.0,
      "grad_norm": 2.6323747268635427,
      "language_loss": 0.73661149,
      "learning_rate": 9.230111933634228e-07,
      "loss": 0.75863349,
      "num_input_tokens_seen": 123545845,
      "step": 5743,
      "time_per_iteration": 2.6665234565734863
    },
    {
      "auxiliary_loss_clip": 0.01174231,
      "auxiliary_loss_mlp": 0.01028849,
      "balance_loss_clip": 1.05182076,
      "balance_loss_mlp": 1.02125585,
      "epoch": 0.6906751638309385,
      "flos": 23114945111040.0,
      "grad_norm": 1.5909936116513266,
      "language_loss": 0.8084445,
      "learning_rate": 9.223548909015288e-07,
      "loss": 0.83047533,
      "num_input_tokens_seen": 123567535,
      "step": 5744,
      "time_per_iteration": 2.6186842918395996
    },
    {
      "auxiliary_loss_clip": 0.01172567,
      "auxiliary_loss_mlp": 0.01023886,
      "balance_loss_clip": 1.04924345,
      "balance_loss_mlp": 1.01611936,
      "epoch": 0.6907954067215776,
      "flos": 27305468375040.0,
      "grad_norm": 4.354093252065987,
      "language_loss": 0.72315705,
      "learning_rate": 9.216987519175407e-07,
      "loss": 0.74512154,
      "num_input_tokens_seen": 123587710,
      "step": 5745,
      "time_per_iteration": 2.6784896850585938
    },
    {
      "auxiliary_loss_clip": 0.01172529,
      "auxiliary_loss_mlp": 0.01029465,
      "balance_loss_clip": 1.05005455,
      "balance_loss_mlp": 1.02211857,
      "epoch": 0.6909156496122166,
      "flos": 21689399070720.0,
      "grad_norm": 1.9391958213189244,
      "language_loss": 0.68627548,
      "learning_rate": 9.210427765109942e-07,
      "loss": 0.70829546,
      "num_input_tokens_seen": 123607385,
      "step": 5746,
      "time_per_iteration": 2.570429563522339
    },
    {
      "auxiliary_loss_clip": 0.01178613,
      "auxiliary_loss_mlp": 0.01027301,
      "balance_loss_clip": 1.05096889,
      "balance_loss_mlp": 1.01919758,
      "epoch": 0.6910358925028558,
      "flos": 22561453463040.0,
      "grad_norm": 2.16650861726077,
      "language_loss": 0.81370199,
      "learning_rate": 9.20386964781402e-07,
      "loss": 0.83576107,
      "num_input_tokens_seen": 123625405,
      "step": 5747,
      "time_per_iteration": 5.366653919219971
    },
    {
      "auxiliary_loss_clip": 0.01173646,
      "auxiliary_loss_mlp": 0.01022424,
      "balance_loss_clip": 1.05042648,
      "balance_loss_mlp": 1.01463413,
      "epoch": 0.6911561353934949,
      "flos": 22054107813120.0,
      "grad_norm": 3.062066513689261,
      "language_loss": 0.84061629,
      "learning_rate": 9.197313168282472e-07,
      "loss": 0.86257696,
      "num_input_tokens_seen": 123642850,
      "step": 5748,
      "time_per_iteration": 2.6127777099609375
    },
    {
      "auxiliary_loss_clip": 0.01172517,
      "auxiliary_loss_mlp": 0.01026655,
      "balance_loss_clip": 1.04820967,
      "balance_loss_mlp": 1.01834059,
      "epoch": 0.6912763782841339,
      "flos": 24206557386240.0,
      "grad_norm": 2.2250087670966825,
      "language_loss": 0.7246033,
      "learning_rate": 9.190758327509935e-07,
      "loss": 0.74659503,
      "num_input_tokens_seen": 123661595,
      "step": 5749,
      "time_per_iteration": 2.616994619369507
    },
    {
      "auxiliary_loss_clip": 0.01072256,
      "auxiliary_loss_mlp": 0.01001147,
      "balance_loss_clip": 1.01350188,
      "balance_loss_mlp": 1.0000149,
      "epoch": 0.6913966211747731,
      "flos": 52329641091840.0,
      "grad_norm": 0.9307515742216789,
      "language_loss": 0.64346164,
      "learning_rate": 9.184205126490767e-07,
      "loss": 0.66419572,
      "num_input_tokens_seen": 123710490,
      "step": 5750,
      "time_per_iteration": 3.022404432296753
    },
    {
      "auxiliary_loss_clip": 0.0107224,
      "auxiliary_loss_mlp": 0.0100177,
      "balance_loss_clip": 1.01347315,
      "balance_loss_mlp": 1.00060749,
      "epoch": 0.6915168640654121,
      "flos": 66741274851840.0,
      "grad_norm": 1.1153048397180134,
      "language_loss": 0.59620166,
      "learning_rate": 9.177653566219075e-07,
      "loss": 0.61694169,
      "num_input_tokens_seen": 123765215,
      "step": 5751,
      "time_per_iteration": 3.0476486682891846
    },
    {
      "auxiliary_loss_clip": 0.01174118,
      "auxiliary_loss_mlp": 0.01023146,
      "balance_loss_clip": 1.05039597,
      "balance_loss_mlp": 1.01520658,
      "epoch": 0.6916371069560512,
      "flos": 18296523175680.0,
      "grad_norm": 2.2644249946076167,
      "language_loss": 0.76693761,
      "learning_rate": 9.171103647688744e-07,
      "loss": 0.78891027,
      "num_input_tokens_seen": 123783955,
      "step": 5752,
      "time_per_iteration": 2.570174217224121
    },
    {
      "auxiliary_loss_clip": 0.01167151,
      "auxiliary_loss_mlp": 0.01022166,
      "balance_loss_clip": 1.04785132,
      "balance_loss_mlp": 1.01442993,
      "epoch": 0.6917573498466904,
      "flos": 19645794685440.0,
      "grad_norm": 1.7344456457953454,
      "language_loss": 0.68925881,
      "learning_rate": 9.164555371893367e-07,
      "loss": 0.71115196,
      "num_input_tokens_seen": 123803885,
      "step": 5753,
      "time_per_iteration": 2.6013801097869873
    },
    {
      "auxiliary_loss_clip": 0.01171193,
      "auxiliary_loss_mlp": 0.0102763,
      "balance_loss_clip": 1.04988039,
      "balance_loss_mlp": 1.02053726,
      "epoch": 0.6918775927373294,
      "flos": 14210319985920.0,
      "grad_norm": 2.291383325897772,
      "language_loss": 0.75566196,
      "learning_rate": 9.158008739826333e-07,
      "loss": 0.77765018,
      "num_input_tokens_seen": 123821485,
      "step": 5754,
      "time_per_iteration": 2.568338632583618
    },
    {
      "auxiliary_loss_clip": 0.01174121,
      "auxiliary_loss_mlp": 0.01026166,
      "balance_loss_clip": 1.04982877,
      "balance_loss_mlp": 1.01801252,
      "epoch": 0.6919978356279685,
      "flos": 23985455218560.0,
      "grad_norm": 1.7197388144564458,
      "language_loss": 0.86797661,
      "learning_rate": 9.151463752480744e-07,
      "loss": 0.88997948,
      "num_input_tokens_seen": 123840215,
      "step": 5755,
      "time_per_iteration": 2.6842188835144043
    },
    {
      "auxiliary_loss_clip": 0.01169135,
      "auxiliary_loss_mlp": 0.01021653,
      "balance_loss_clip": 1.04774892,
      "balance_loss_mlp": 1.01437545,
      "epoch": 0.6921180785186076,
      "flos": 23622937205760.0,
      "grad_norm": 5.787111916645679,
      "language_loss": 0.80424964,
      "learning_rate": 9.144920410849493e-07,
      "loss": 0.82615757,
      "num_input_tokens_seen": 123861450,
      "step": 5756,
      "time_per_iteration": 2.6573400497436523
    },
    {
      "auxiliary_loss_clip": 0.01170873,
      "auxiliary_loss_mlp": 0.01026514,
      "balance_loss_clip": 1.04907179,
      "balance_loss_mlp": 1.01871133,
      "epoch": 0.6922383214092467,
      "flos": 21142623265920.0,
      "grad_norm": 2.4360110819569027,
      "language_loss": 0.8014375,
      "learning_rate": 9.138378715925176e-07,
      "loss": 0.82341135,
      "num_input_tokens_seen": 123880545,
      "step": 5757,
      "time_per_iteration": 2.6184909343719482
    },
    {
      "auxiliary_loss_clip": 0.01169922,
      "auxiliary_loss_mlp": 0.01022698,
      "balance_loss_clip": 1.04794836,
      "balance_loss_mlp": 1.01477015,
      "epoch": 0.6923585642998857,
      "flos": 21470667200640.0,
      "grad_norm": 1.9680579593995609,
      "language_loss": 0.8138535,
      "learning_rate": 9.131838668700167e-07,
      "loss": 0.83577967,
      "num_input_tokens_seen": 123900615,
      "step": 5758,
      "time_per_iteration": 2.5919864177703857
    },
    {
      "auxiliary_loss_clip": 0.01173299,
      "auxiliary_loss_mlp": 0.01029436,
      "balance_loss_clip": 1.05057096,
      "balance_loss_mlp": 1.02185404,
      "epoch": 0.6924788071905249,
      "flos": 21105204272640.0,
      "grad_norm": 2.872480328090041,
      "language_loss": 0.8671903,
      "learning_rate": 9.125300270166598e-07,
      "loss": 0.88921762,
      "num_input_tokens_seen": 123921220,
      "step": 5759,
      "time_per_iteration": 2.663618326187134
    },
    {
      "auxiliary_loss_clip": 0.01171077,
      "auxiliary_loss_mlp": 0.01025069,
      "balance_loss_clip": 1.04835224,
      "balance_loss_mlp": 1.01773787,
      "epoch": 0.692599050081164,
      "flos": 26250018117120.0,
      "grad_norm": 1.7221246807064376,
      "language_loss": 0.86028445,
      "learning_rate": 9.118763521316324e-07,
      "loss": 0.88224596,
      "num_input_tokens_seen": 123941795,
      "step": 5760,
      "time_per_iteration": 2.634003162384033
    },
    {
      "auxiliary_loss_clip": 0.01170408,
      "auxiliary_loss_mlp": 0.01022417,
      "balance_loss_clip": 1.04809642,
      "balance_loss_mlp": 1.01518142,
      "epoch": 0.692719292971803,
      "flos": 20885215426560.0,
      "grad_norm": 1.8406174292542563,
      "language_loss": 0.76177514,
      "learning_rate": 9.112228423140987e-07,
      "loss": 0.78370339,
      "num_input_tokens_seen": 123960715,
      "step": 5761,
      "time_per_iteration": 2.6395843029022217
    },
    {
      "auxiliary_loss_clip": 0.01172913,
      "auxiliary_loss_mlp": 0.01029921,
      "balance_loss_clip": 1.0494709,
      "balance_loss_mlp": 1.0218271,
      "epoch": 0.6928395358624422,
      "flos": 25921938268800.0,
      "grad_norm": 2.5929995266653494,
      "language_loss": 0.86622441,
      "learning_rate": 9.105694976631932e-07,
      "loss": 0.88825274,
      "num_input_tokens_seen": 123978625,
      "step": 5762,
      "time_per_iteration": 2.59049654006958
    },
    {
      "auxiliary_loss_clip": 0.01174981,
      "auxiliary_loss_mlp": 0.01024788,
      "balance_loss_clip": 1.05180633,
      "balance_loss_mlp": 1.01708102,
      "epoch": 0.6929597787530812,
      "flos": 23586559706880.0,
      "grad_norm": 2.1578070474934736,
      "language_loss": 0.72906268,
      "learning_rate": 9.099163182780283e-07,
      "loss": 0.75106037,
      "num_input_tokens_seen": 123996780,
      "step": 5763,
      "time_per_iteration": 2.6187868118286133
    },
    {
      "auxiliary_loss_clip": 0.01171708,
      "auxiliary_loss_mlp": 0.01025943,
      "balance_loss_clip": 1.04860759,
      "balance_loss_mlp": 1.01780748,
      "epoch": 0.6930800216437203,
      "flos": 18255656476800.0,
      "grad_norm": 2.611217396903836,
      "language_loss": 0.49306896,
      "learning_rate": 9.092633042576916e-07,
      "loss": 0.51504546,
      "num_input_tokens_seen": 124014045,
      "step": 5764,
      "time_per_iteration": 3.552968978881836
    },
    {
      "auxiliary_loss_clip": 0.01172419,
      "auxiliary_loss_mlp": 0.01024579,
      "balance_loss_clip": 1.0506382,
      "balance_loss_mlp": 1.01670527,
      "epoch": 0.6932002645343595,
      "flos": 29168621809920.0,
      "grad_norm": 4.438272308613782,
      "language_loss": 0.56566948,
      "learning_rate": 9.086104557012446e-07,
      "loss": 0.58763945,
      "num_input_tokens_seen": 124034615,
      "step": 5765,
      "time_per_iteration": 2.626354694366455
    },
    {
      "auxiliary_loss_clip": 0.01169609,
      "auxiliary_loss_mlp": 0.01025445,
      "balance_loss_clip": 1.04930973,
      "balance_loss_mlp": 1.01785707,
      "epoch": 0.6933205074249985,
      "flos": 23842746483840.0,
      "grad_norm": 4.651428442164186,
      "language_loss": 0.65878344,
      "learning_rate": 9.079577727077239e-07,
      "loss": 0.68073392,
      "num_input_tokens_seen": 124053445,
      "step": 5766,
      "time_per_iteration": 2.6552436351776123
    },
    {
      "auxiliary_loss_clip": 0.01175067,
      "auxiliary_loss_mlp": 0.01025489,
      "balance_loss_clip": 1.05160069,
      "balance_loss_mlp": 1.0167625,
      "epoch": 0.6934407503156376,
      "flos": 24166696268160.0,
      "grad_norm": 2.405113837766944,
      "language_loss": 0.72386867,
      "learning_rate": 9.073052553761404e-07,
      "loss": 0.74587423,
      "num_input_tokens_seen": 124072810,
      "step": 5767,
      "time_per_iteration": 2.5764124393463135
    },
    {
      "auxiliary_loss_clip": 0.01179748,
      "auxiliary_loss_mlp": 0.01025398,
      "balance_loss_clip": 1.05195987,
      "balance_loss_mlp": 1.01668417,
      "epoch": 0.6935609932062767,
      "flos": 20631327120000.0,
      "grad_norm": 1.707975261466464,
      "language_loss": 0.77995759,
      "learning_rate": 9.066529038054805e-07,
      "loss": 0.80200911,
      "num_input_tokens_seen": 124092875,
      "step": 5768,
      "time_per_iteration": 2.69997239112854
    },
    {
      "auxiliary_loss_clip": 0.01173898,
      "auxiliary_loss_mlp": 0.0102736,
      "balance_loss_clip": 1.05106187,
      "balance_loss_mlp": 1.02035046,
      "epoch": 0.6936812360969158,
      "flos": 18254184019200.0,
      "grad_norm": 1.9337738497295036,
      "language_loss": 0.74108076,
      "learning_rate": 9.060007180947071e-07,
      "loss": 0.76309329,
      "num_input_tokens_seen": 124110930,
      "step": 5769,
      "time_per_iteration": 2.565394401550293
    },
    {
      "auxiliary_loss_clip": 0.0117301,
      "auxiliary_loss_mlp": 0.01024226,
      "balance_loss_clip": 1.04956889,
      "balance_loss_mlp": 1.0159409,
      "epoch": 0.6938014789875548,
      "flos": 31317336368640.0,
      "grad_norm": 1.9420093181182785,
      "language_loss": 0.73365957,
      "learning_rate": 9.053486983427534e-07,
      "loss": 0.75563192,
      "num_input_tokens_seen": 124132180,
      "step": 5770,
      "time_per_iteration": 2.7321598529815674
    },
    {
      "auxiliary_loss_clip": 0.01171173,
      "auxiliary_loss_mlp": 0.01020841,
      "balance_loss_clip": 1.04786229,
      "balance_loss_mlp": 1.01370382,
      "epoch": 0.6939217218781939,
      "flos": 17528429721600.0,
      "grad_norm": 1.8385604070161632,
      "language_loss": 0.70569074,
      "learning_rate": 9.046968446485326e-07,
      "loss": 0.72761095,
      "num_input_tokens_seen": 124150585,
      "step": 5771,
      "time_per_iteration": 2.636018991470337
    },
    {
      "auxiliary_loss_clip": 0.01175575,
      "auxiliary_loss_mlp": 0.01027965,
      "balance_loss_clip": 1.05060339,
      "balance_loss_mlp": 1.01975787,
      "epoch": 0.6940419647688331,
      "flos": 18551776199040.0,
      "grad_norm": 3.4380268836918764,
      "language_loss": 0.70292413,
      "learning_rate": 9.040451571109295e-07,
      "loss": 0.72495955,
      "num_input_tokens_seen": 124166205,
      "step": 5772,
      "time_per_iteration": 2.6433908939361572
    },
    {
      "auxiliary_loss_clip": 0.0107156,
      "auxiliary_loss_mlp": 0.01001472,
      "balance_loss_clip": 1.01278973,
      "balance_loss_mlp": 1.00041103,
      "epoch": 0.6941622076594721,
      "flos": 66926286829440.0,
      "grad_norm": 1.0237705033762934,
      "language_loss": 0.60424197,
      "learning_rate": 9.033936358288042e-07,
      "loss": 0.62497222,
      "num_input_tokens_seen": 124219940,
      "step": 5773,
      "time_per_iteration": 3.8834173679351807
    },
    {
      "auxiliary_loss_clip": 0.01173498,
      "auxiliary_loss_mlp": 0.0102484,
      "balance_loss_clip": 1.04959679,
      "balance_loss_mlp": 1.0176456,
      "epoch": 0.6942824505501112,
      "flos": 26578062051840.0,
      "grad_norm": 2.365886007290814,
      "language_loss": 0.82479805,
      "learning_rate": 9.027422809009937e-07,
      "loss": 0.84678149,
      "num_input_tokens_seen": 124239885,
      "step": 5774,
      "time_per_iteration": 3.600712299346924
    },
    {
      "auxiliary_loss_clip": 0.01172274,
      "auxiliary_loss_mlp": 0.01024745,
      "balance_loss_clip": 1.04929399,
      "balance_loss_mlp": 1.01728523,
      "epoch": 0.6944026934407503,
      "flos": 21248308056960.0,
      "grad_norm": 1.7680480807040102,
      "language_loss": 0.83248878,
      "learning_rate": 9.020910924263054e-07,
      "loss": 0.85445893,
      "num_input_tokens_seen": 124258410,
      "step": 5775,
      "time_per_iteration": 3.457395553588867
    },
    {
      "auxiliary_loss_clip": 0.01071762,
      "auxiliary_loss_mlp": 0.01000739,
      "balance_loss_clip": 1.01326764,
      "balance_loss_mlp": 0.99966592,
      "epoch": 0.6945229363313894,
      "flos": 70677191537280.0,
      "grad_norm": 0.8189197476775089,
      "language_loss": 0.58140671,
      "learning_rate": 9.014400705035261e-07,
      "loss": 0.60213172,
      "num_input_tokens_seen": 124315315,
      "step": 5776,
      "time_per_iteration": 3.215857982635498
    },
    {
      "auxiliary_loss_clip": 0.01173345,
      "auxiliary_loss_mlp": 0.01034245,
      "balance_loss_clip": 1.05248117,
      "balance_loss_mlp": 1.02631116,
      "epoch": 0.6946431792220285,
      "flos": 18952934267520.0,
      "grad_norm": 2.006298812283388,
      "language_loss": 0.76610047,
      "learning_rate": 9.00789215231414e-07,
      "loss": 0.78817636,
      "num_input_tokens_seen": 124333710,
      "step": 5777,
      "time_per_iteration": 2.6221752166748047
    },
    {
      "auxiliary_loss_clip": 0.01174128,
      "auxiliary_loss_mlp": 0.01027058,
      "balance_loss_clip": 1.04976511,
      "balance_loss_mlp": 1.01922655,
      "epoch": 0.6947634221126676,
      "flos": 20338834671360.0,
      "grad_norm": 2.289878584795425,
      "language_loss": 0.82018119,
      "learning_rate": 9.001385267087056e-07,
      "loss": 0.84219301,
      "num_input_tokens_seen": 124352855,
      "step": 5778,
      "time_per_iteration": 2.637181520462036
    },
    {
      "auxiliary_loss_clip": 0.01173156,
      "auxiliary_loss_mlp": 0.01024969,
      "balance_loss_clip": 1.05013061,
      "balance_loss_mlp": 1.01744664,
      "epoch": 0.6948836650033067,
      "flos": 21833723917440.0,
      "grad_norm": 1.5900927582112545,
      "language_loss": 0.70738167,
      "learning_rate": 8.994880050341072e-07,
      "loss": 0.72936296,
      "num_input_tokens_seen": 124372960,
      "step": 5779,
      "time_per_iteration": 2.6539978981018066
    },
    {
      "auxiliary_loss_clip": 0.01178188,
      "auxiliary_loss_mlp": 0.01028441,
      "balance_loss_clip": 1.05573654,
      "balance_loss_mlp": 1.02023315,
      "epoch": 0.6950039078939457,
      "flos": 23657519024640.0,
      "grad_norm": 2.6158775875694174,
      "language_loss": 0.77614427,
      "learning_rate": 8.988376503063026e-07,
      "loss": 0.7982105,
      "num_input_tokens_seen": 124394220,
      "step": 5780,
      "time_per_iteration": 2.6911001205444336
    },
    {
      "auxiliary_loss_clip": 0.01175639,
      "auxiliary_loss_mlp": 0.01024065,
      "balance_loss_clip": 1.05065727,
      "balance_loss_mlp": 1.01554751,
      "epoch": 0.6951241507845849,
      "flos": 21792462168960.0,
      "grad_norm": 2.0410873754490613,
      "language_loss": 0.81382811,
      "learning_rate": 8.981874626239521e-07,
      "loss": 0.83582509,
      "num_input_tokens_seen": 124412795,
      "step": 5781,
      "time_per_iteration": 2.5998964309692383
    },
    {
      "auxiliary_loss_clip": 0.01177591,
      "auxiliary_loss_mlp": 0.01021789,
      "balance_loss_clip": 1.05296171,
      "balance_loss_mlp": 1.01413548,
      "epoch": 0.695244393675224,
      "flos": 14647568244480.0,
      "grad_norm": 2.121900510409872,
      "language_loss": 0.8870182,
      "learning_rate": 8.975374420856872e-07,
      "loss": 0.90901196,
      "num_input_tokens_seen": 124429690,
      "step": 5782,
      "time_per_iteration": 2.6206421852111816
    },
    {
      "auxiliary_loss_clip": 0.01169828,
      "auxiliary_loss_mlp": 0.01025032,
      "balance_loss_clip": 1.04801631,
      "balance_loss_mlp": 1.01772428,
      "epoch": 0.695364636565863,
      "flos": 16873203778560.0,
      "grad_norm": 2.1263167797003,
      "language_loss": 0.72556049,
      "learning_rate": 8.968875887901157e-07,
      "loss": 0.74750912,
      "num_input_tokens_seen": 124447070,
      "step": 5783,
      "time_per_iteration": 2.594132900238037
    },
    {
      "auxiliary_loss_clip": 0.01173763,
      "auxiliary_loss_mlp": 0.01030014,
      "balance_loss_clip": 1.05096126,
      "balance_loss_mlp": 1.02161026,
      "epoch": 0.6954848794565022,
      "flos": 19354523299200.0,
      "grad_norm": 6.400250161289617,
      "language_loss": 0.62998641,
      "learning_rate": 8.9623790283582e-07,
      "loss": 0.65202421,
      "num_input_tokens_seen": 124464950,
      "step": 5784,
      "time_per_iteration": 2.585742473602295
    },
    {
      "auxiliary_loss_clip": 0.01173509,
      "auxiliary_loss_mlp": 0.01029741,
      "balance_loss_clip": 1.05038309,
      "balance_loss_mlp": 1.0217545,
      "epoch": 0.6956051223471412,
      "flos": 18990209606400.0,
      "grad_norm": 2.9812838203322194,
      "language_loss": 0.76928127,
      "learning_rate": 8.955883843213561e-07,
      "loss": 0.79131377,
      "num_input_tokens_seen": 124483965,
      "step": 5785,
      "time_per_iteration": 2.678738832473755
    },
    {
      "auxiliary_loss_clip": 0.01174843,
      "auxiliary_loss_mlp": 0.01027539,
      "balance_loss_clip": 1.04867494,
      "balance_loss_mlp": 1.0194211,
      "epoch": 0.6957253652377803,
      "flos": 16107229226880.0,
      "grad_norm": 1.9865763746186007,
      "language_loss": 0.87510312,
      "learning_rate": 8.949390333452569e-07,
      "loss": 0.89712691,
      "num_input_tokens_seen": 124501910,
      "step": 5786,
      "time_per_iteration": 2.641763925552368
    },
    {
      "auxiliary_loss_clip": 0.01173072,
      "auxiliary_loss_mlp": 0.01023754,
      "balance_loss_clip": 1.05094182,
      "balance_loss_mlp": 1.01613379,
      "epoch": 0.6958456081284194,
      "flos": 29388646569600.0,
      "grad_norm": 2.083244059550654,
      "language_loss": 0.68161678,
      "learning_rate": 8.942898500060279e-07,
      "loss": 0.70358503,
      "num_input_tokens_seen": 124521625,
      "step": 5787,
      "time_per_iteration": 2.6395633220672607
    },
    {
      "auxiliary_loss_clip": 0.01179181,
      "auxiliary_loss_mlp": 0.01028965,
      "balance_loss_clip": 1.05383086,
      "balance_loss_mlp": 1.02041745,
      "epoch": 0.6959658510190585,
      "flos": 25154850395520.0,
      "grad_norm": 2.3830494928224835,
      "language_loss": 0.72396064,
      "learning_rate": 8.936408344021493e-07,
      "loss": 0.74604207,
      "num_input_tokens_seen": 124538540,
      "step": 5788,
      "time_per_iteration": 2.697237253189087
    },
    {
      "auxiliary_loss_clip": 0.01179381,
      "auxiliary_loss_mlp": 0.01027624,
      "balance_loss_clip": 1.05307913,
      "balance_loss_mlp": 1.01953542,
      "epoch": 0.6960860939096976,
      "flos": 42814388759040.0,
      "grad_norm": 2.0657476786949727,
      "language_loss": 0.71409345,
      "learning_rate": 8.929919866320765e-07,
      "loss": 0.7361635,
      "num_input_tokens_seen": 124559355,
      "step": 5789,
      "time_per_iteration": 2.797950029373169
    },
    {
      "auxiliary_loss_clip": 0.01174937,
      "auxiliary_loss_mlp": 0.01028427,
      "balance_loss_clip": 1.05074847,
      "balance_loss_mlp": 1.02026677,
      "epoch": 0.6962063368003367,
      "flos": 17566566986880.0,
      "grad_norm": 1.9629568874737366,
      "language_loss": 0.81323743,
      "learning_rate": 8.923433067942385e-07,
      "loss": 0.835271,
      "num_input_tokens_seen": 124577920,
      "step": 5790,
      "time_per_iteration": 3.5486867427825928
    },
    {
      "auxiliary_loss_clip": 0.01171011,
      "auxiliary_loss_mlp": 0.01030028,
      "balance_loss_clip": 1.05007005,
      "balance_loss_mlp": 1.02236271,
      "epoch": 0.6963265796909758,
      "flos": 21251648021760.0,
      "grad_norm": 1.9681664167729511,
      "language_loss": 0.68806756,
      "learning_rate": 8.916947949870417e-07,
      "loss": 0.710078,
      "num_input_tokens_seen": 124597585,
      "step": 5791,
      "time_per_iteration": 2.631840705871582
    },
    {
      "auxiliary_loss_clip": 0.01071476,
      "auxiliary_loss_mlp": 0.01001896,
      "balance_loss_clip": 1.01256597,
      "balance_loss_mlp": 1.00078773,
      "epoch": 0.6964468225816148,
      "flos": 68828295801600.0,
      "grad_norm": 0.7545465504647046,
      "language_loss": 0.58135086,
      "learning_rate": 8.910464513088615e-07,
      "loss": 0.60208458,
      "num_input_tokens_seen": 124661625,
      "step": 5792,
      "time_per_iteration": 3.2799603939056396
    },
    {
      "auxiliary_loss_clip": 0.01172467,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.04886425,
      "balance_loss_mlp": 1.01663733,
      "epoch": 0.696567065472254,
      "flos": 18950887192320.0,
      "grad_norm": 1.9516529130781834,
      "language_loss": 0.78341323,
      "learning_rate": 8.903982758580542e-07,
      "loss": 0.80538213,
      "num_input_tokens_seen": 124680565,
      "step": 5793,
      "time_per_iteration": 2.6256792545318604
    },
    {
      "auxiliary_loss_clip": 0.01175643,
      "auxiliary_loss_mlp": 0.01030812,
      "balance_loss_clip": 1.05041599,
      "balance_loss_mlp": 1.02218747,
      "epoch": 0.696687308362893,
      "flos": 22856675345280.0,
      "grad_norm": 2.171469189408998,
      "language_loss": 0.80092448,
      "learning_rate": 8.897502687329457e-07,
      "loss": 0.82298905,
      "num_input_tokens_seen": 124700365,
      "step": 5794,
      "time_per_iteration": 2.620657205581665
    },
    {
      "auxiliary_loss_clip": 0.01172864,
      "auxiliary_loss_mlp": 0.01029051,
      "balance_loss_clip": 1.05036438,
      "balance_loss_mlp": 1.02145708,
      "epoch": 0.6968075512535321,
      "flos": 24972926987520.0,
      "grad_norm": 12.75106015760131,
      "language_loss": 0.80183578,
      "learning_rate": 8.891024300318382e-07,
      "loss": 0.82385492,
      "num_input_tokens_seen": 124718935,
      "step": 5795,
      "time_per_iteration": 2.648496150970459
    },
    {
      "auxiliary_loss_clip": 0.01169868,
      "auxiliary_loss_mlp": 0.01021213,
      "balance_loss_clip": 1.0495069,
      "balance_loss_mlp": 1.01373816,
      "epoch": 0.6969277941441713,
      "flos": 21030438113280.0,
      "grad_norm": 1.4923594433361076,
      "language_loss": 0.75571352,
      "learning_rate": 8.884547598530103e-07,
      "loss": 0.77762431,
      "num_input_tokens_seen": 124739505,
      "step": 5796,
      "time_per_iteration": 2.650059223175049
    },
    {
      "auxiliary_loss_clip": 0.01174464,
      "auxiliary_loss_mlp": 0.01026401,
      "balance_loss_clip": 1.05018938,
      "balance_loss_mlp": 1.01862884,
      "epoch": 0.6970480370348103,
      "flos": 21579404647680.0,
      "grad_norm": 1.774333235354109,
      "language_loss": 0.75395471,
      "learning_rate": 8.8780725829471e-07,
      "loss": 0.77596331,
      "num_input_tokens_seen": 124757410,
      "step": 5797,
      "time_per_iteration": 2.6342782974243164
    },
    {
      "auxiliary_loss_clip": 0.01172525,
      "auxiliary_loss_mlp": 0.01025959,
      "balance_loss_clip": 1.04922223,
      "balance_loss_mlp": 1.01806998,
      "epoch": 0.6971682799254494,
      "flos": 22419175691520.0,
      "grad_norm": 2.430757271155554,
      "language_loss": 0.7797749,
      "learning_rate": 8.87159925455165e-07,
      "loss": 0.80175972,
      "num_input_tokens_seen": 124777240,
      "step": 5798,
      "time_per_iteration": 2.6946935653686523
    },
    {
      "auxiliary_loss_clip": 0.0117004,
      "auxiliary_loss_mlp": 0.01025493,
      "balance_loss_clip": 1.04852331,
      "balance_loss_mlp": 1.01814091,
      "epoch": 0.6972885228160886,
      "flos": 20005834659840.0,
      "grad_norm": 3.1598086831531136,
      "language_loss": 0.73161179,
      "learning_rate": 8.865127614325738e-07,
      "loss": 0.75356716,
      "num_input_tokens_seen": 124795670,
      "step": 5799,
      "time_per_iteration": 2.568962335586548
    },
    {
      "auxiliary_loss_clip": 0.01177254,
      "auxiliary_loss_mlp": 0.01026159,
      "balance_loss_clip": 1.05345881,
      "balance_loss_mlp": 1.01782012,
      "epoch": 0.6974087657067276,
      "flos": 37853437656960.0,
      "grad_norm": 2.0472657263735123,
      "language_loss": 0.6656369,
      "learning_rate": 8.85865766325113e-07,
      "loss": 0.68767095,
      "num_input_tokens_seen": 124819600,
      "step": 5800,
      "time_per_iteration": 3.5883281230926514
    },
    {
      "auxiliary_loss_clip": 0.01178267,
      "auxiliary_loss_mlp": 0.0102255,
      "balance_loss_clip": 1.05234742,
      "balance_loss_mlp": 1.01485479,
      "epoch": 0.6975290085973667,
      "flos": 29489267543040.0,
      "grad_norm": 2.3406036507189687,
      "language_loss": 0.72249722,
      "learning_rate": 8.852189402309287e-07,
      "loss": 0.74450541,
      "num_input_tokens_seen": 124838785,
      "step": 5801,
      "time_per_iteration": 3.6026828289031982
    },
    {
      "auxiliary_loss_clip": 0.01177367,
      "auxiliary_loss_mlp": 0.0102474,
      "balance_loss_clip": 1.05159259,
      "balance_loss_mlp": 1.01650858,
      "epoch": 0.6976492514880057,
      "flos": 12895630295040.0,
      "grad_norm": 5.675120620609496,
      "language_loss": 0.74666673,
      "learning_rate": 8.845722832481441e-07,
      "loss": 0.76868784,
      "num_input_tokens_seen": 124854215,
      "step": 5802,
      "time_per_iteration": 2.5674707889556885
    },
    {
      "auxiliary_loss_clip": 0.01170419,
      "auxiliary_loss_mlp": 0.01026029,
      "balance_loss_clip": 1.04713249,
      "balance_loss_mlp": 1.01757121,
      "epoch": 0.6977694943786449,
      "flos": 24352929308160.0,
      "grad_norm": 5.063323599282646,
      "language_loss": 0.77555943,
      "learning_rate": 8.83925795474858e-07,
      "loss": 0.79752398,
      "num_input_tokens_seen": 124874340,
      "step": 5803,
      "time_per_iteration": 2.7980587482452393
    },
    {
      "auxiliary_loss_clip": 0.0117142,
      "auxiliary_loss_mlp": 0.01021167,
      "balance_loss_clip": 1.04802871,
      "balance_loss_mlp": 1.01337421,
      "epoch": 0.6978897372692839,
      "flos": 29898470257920.0,
      "grad_norm": 4.236564131486457,
      "language_loss": 0.59358019,
      "learning_rate": 8.832794770091414e-07,
      "loss": 0.61550611,
      "num_input_tokens_seen": 124895175,
      "step": 5804,
      "time_per_iteration": 2.672621488571167
    },
    {
      "auxiliary_loss_clip": 0.01170149,
      "auxiliary_loss_mlp": 0.01021369,
      "balance_loss_clip": 1.04781079,
      "balance_loss_mlp": 1.01358461,
      "epoch": 0.698009980159923,
      "flos": 21761579450880.0,
      "grad_norm": 1.9504303195656858,
      "language_loss": 0.82703602,
      "learning_rate": 8.826333279490401e-07,
      "loss": 0.84895116,
      "num_input_tokens_seen": 124915810,
      "step": 5805,
      "time_per_iteration": 2.622953176498413
    },
    {
      "auxiliary_loss_clip": 0.01174812,
      "auxiliary_loss_mlp": 0.01029821,
      "balance_loss_clip": 1.05318868,
      "balance_loss_mlp": 1.02227163,
      "epoch": 0.6981302230505622,
      "flos": 19857164267520.0,
      "grad_norm": 3.679271273717413,
      "language_loss": 0.67942768,
      "learning_rate": 8.819873483925748e-07,
      "loss": 0.70147401,
      "num_input_tokens_seen": 124932930,
      "step": 5806,
      "time_per_iteration": 2.6050987243652344
    },
    {
      "auxiliary_loss_clip": 0.01173091,
      "auxiliary_loss_mlp": 0.01027261,
      "balance_loss_clip": 1.05048227,
      "balance_loss_mlp": 1.01877379,
      "epoch": 0.6982504659412012,
      "flos": 22198648141440.0,
      "grad_norm": 2.223380077415655,
      "language_loss": 0.74663901,
      "learning_rate": 8.81341538437739e-07,
      "loss": 0.76864254,
      "num_input_tokens_seen": 124951220,
      "step": 5807,
      "time_per_iteration": 2.5928585529327393
    },
    {
      "auxiliary_loss_clip": 0.01172191,
      "auxiliary_loss_mlp": 0.01025854,
      "balance_loss_clip": 1.04757392,
      "balance_loss_mlp": 1.0184269,
      "epoch": 0.6983707088318403,
      "flos": 35588479708800.0,
      "grad_norm": 1.6965532660783547,
      "language_loss": 0.68399644,
      "learning_rate": 8.80695898182503e-07,
      "loss": 0.70597684,
      "num_input_tokens_seen": 124972200,
      "step": 5808,
      "time_per_iteration": 2.6839675903320312
    },
    {
      "auxiliary_loss_clip": 0.0107259,
      "auxiliary_loss_mlp": 0.01002289,
      "balance_loss_clip": 1.01319361,
      "balance_loss_mlp": 1.00113821,
      "epoch": 0.6984909517224794,
      "flos": 65440052760960.0,
      "grad_norm": 0.8210371085980239,
      "language_loss": 0.65015835,
      "learning_rate": 8.800504277248093e-07,
      "loss": 0.67090714,
      "num_input_tokens_seen": 125036950,
      "step": 5809,
      "time_per_iteration": 3.2099432945251465
    },
    {
      "auxiliary_loss_clip": 0.01175434,
      "auxiliary_loss_mlp": 0.01027104,
      "balance_loss_clip": 1.05214691,
      "balance_loss_mlp": 1.01975155,
      "epoch": 0.6986111946131185,
      "flos": 18546927863040.0,
      "grad_norm": 1.8142351138303299,
      "language_loss": 0.74965274,
      "learning_rate": 8.794051271625753e-07,
      "loss": 0.77167809,
      "num_input_tokens_seen": 125054585,
      "step": 5810,
      "time_per_iteration": 2.5663459300994873
    },
    {
      "auxiliary_loss_clip": 0.01170997,
      "auxiliary_loss_mlp": 0.01021307,
      "balance_loss_clip": 1.04814506,
      "balance_loss_mlp": 1.01343274,
      "epoch": 0.6987314375037575,
      "flos": 23039173370880.0,
      "grad_norm": 1.7255129935853772,
      "language_loss": 0.83150041,
      "learning_rate": 8.787599965936925e-07,
      "loss": 0.85342348,
      "num_input_tokens_seen": 125075515,
      "step": 5811,
      "time_per_iteration": 2.6469168663024902
    },
    {
      "auxiliary_loss_clip": 0.01171726,
      "auxiliary_loss_mlp": 0.01022883,
      "balance_loss_clip": 1.05066264,
      "balance_loss_mlp": 1.01580143,
      "epoch": 0.6988516803943967,
      "flos": 38400393029760.0,
      "grad_norm": 1.5922480911456454,
      "language_loss": 0.71677476,
      "learning_rate": 8.781150361160261e-07,
      "loss": 0.73872083,
      "num_input_tokens_seen": 125097425,
      "step": 5812,
      "time_per_iteration": 2.747957706451416
    },
    {
      "auxiliary_loss_clip": 0.01172255,
      "auxiliary_loss_mlp": 0.0102921,
      "balance_loss_clip": 1.05079913,
      "balance_loss_mlp": 1.02136016,
      "epoch": 0.6989719232850358,
      "flos": 24096993926400.0,
      "grad_norm": 1.663899941586857,
      "language_loss": 0.73526508,
      "learning_rate": 8.774702458274181e-07,
      "loss": 0.75727975,
      "num_input_tokens_seen": 125117830,
      "step": 5813,
      "time_per_iteration": 2.6249005794525146
    },
    {
      "auxiliary_loss_clip": 0.01171999,
      "auxiliary_loss_mlp": 0.01024469,
      "balance_loss_clip": 1.04727888,
      "balance_loss_mlp": 1.01617503,
      "epoch": 0.6990921661756748,
      "flos": 14866838818560.0,
      "grad_norm": 2.4982878289857444,
      "language_loss": 0.7058363,
      "learning_rate": 8.768256258256799e-07,
      "loss": 0.72780097,
      "num_input_tokens_seen": 125134455,
      "step": 5814,
      "time_per_iteration": 2.5844337940216064
    },
    {
      "auxiliary_loss_clip": 0.01173082,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 1.05096173,
      "balance_loss_mlp": 1.01882553,
      "epoch": 0.699212409066314,
      "flos": 20193719725440.0,
      "grad_norm": 2.0339558689811787,
      "language_loss": 0.73994315,
      "learning_rate": 8.76181176208602e-07,
      "loss": 0.76194012,
      "num_input_tokens_seen": 125152555,
      "step": 5815,
      "time_per_iteration": 2.6686155796051025
    },
    {
      "auxiliary_loss_clip": 0.01171813,
      "auxiliary_loss_mlp": 0.0102607,
      "balance_loss_clip": 1.0489881,
      "balance_loss_mlp": 1.01837504,
      "epoch": 0.699332651956953,
      "flos": 19427888828160.0,
      "grad_norm": 1.7557810110303975,
      "language_loss": 0.7343114,
      "learning_rate": 8.755368970739461e-07,
      "loss": 0.75629032,
      "num_input_tokens_seen": 125171915,
      "step": 5816,
      "time_per_iteration": 3.5687732696533203
    },
    {
      "auxiliary_loss_clip": 0.01175062,
      "auxiliary_loss_mlp": 0.01031018,
      "balance_loss_clip": 1.0495559,
      "balance_loss_mlp": 1.02289987,
      "epoch": 0.6994528948475921,
      "flos": 16143714466560.0,
      "grad_norm": 7.352234814037665,
      "language_loss": 0.61483842,
      "learning_rate": 8.748927885194479e-07,
      "loss": 0.63689923,
      "num_input_tokens_seen": 125190220,
      "step": 5817,
      "time_per_iteration": 2.5151829719543457
    },
    {
      "auxiliary_loss_clip": 0.01071084,
      "auxiliary_loss_mlp": 0.01001802,
      "balance_loss_clip": 1.01260328,
      "balance_loss_mlp": 1.00069964,
      "epoch": 0.6995731377382313,
      "flos": 64952420699520.0,
      "grad_norm": 0.9708992495394452,
      "language_loss": 0.57387209,
      "learning_rate": 8.742488506428209e-07,
      "loss": 0.59460104,
      "num_input_tokens_seen": 125249310,
      "step": 5818,
      "time_per_iteration": 3.1119589805603027
    },
    {
      "auxiliary_loss_clip": 0.01167718,
      "auxiliary_loss_mlp": 0.01027902,
      "balance_loss_clip": 1.0468713,
      "balance_loss_mlp": 1.02096391,
      "epoch": 0.6996933806288703,
      "flos": 24900136076160.0,
      "grad_norm": 2.2784589436347895,
      "language_loss": 0.78483295,
      "learning_rate": 8.736050835417466e-07,
      "loss": 0.80678916,
      "num_input_tokens_seen": 125269350,
      "step": 5819,
      "time_per_iteration": 2.6306686401367188
    },
    {
      "auxiliary_loss_clip": 0.01172281,
      "auxiliary_loss_mlp": 0.01026924,
      "balance_loss_clip": 1.04973698,
      "balance_loss_mlp": 1.01904464,
      "epoch": 0.6998136235195094,
      "flos": 20777806782720.0,
      "grad_norm": 1.9490573120227423,
      "language_loss": 0.61595088,
      "learning_rate": 8.729614873138862e-07,
      "loss": 0.63794291,
      "num_input_tokens_seen": 125286985,
      "step": 5820,
      "time_per_iteration": 2.5858113765716553
    },
    {
      "auxiliary_loss_clip": 0.01176761,
      "auxiliary_loss_mlp": 0.01034789,
      "balance_loss_clip": 1.0523591,
      "balance_loss_mlp": 1.02653933,
      "epoch": 0.6999338664101485,
      "flos": 23733470332800.0,
      "grad_norm": 2.940413037468152,
      "language_loss": 0.77658927,
      "learning_rate": 8.723180620568716e-07,
      "loss": 0.7987048,
      "num_input_tokens_seen": 125306240,
      "step": 5821,
      "time_per_iteration": 2.653369665145874
    },
    {
      "auxiliary_loss_clip": 0.01169829,
      "auxiliary_loss_mlp": 0.01025999,
      "balance_loss_clip": 1.04976869,
      "balance_loss_mlp": 1.01878715,
      "epoch": 0.7000541093007876,
      "flos": 19864598382720.0,
      "grad_norm": 1.8267287463770572,
      "language_loss": 0.8500101,
      "learning_rate": 8.716748078683116e-07,
      "loss": 0.87196833,
      "num_input_tokens_seen": 125323015,
      "step": 5822,
      "time_per_iteration": 2.5899527072906494
    },
    {
      "auxiliary_loss_clip": 0.01174478,
      "auxiliary_loss_mlp": 0.01025906,
      "balance_loss_clip": 1.04988325,
      "balance_loss_mlp": 1.01756787,
      "epoch": 0.7001743521914267,
      "flos": 29679056029440.0,
      "grad_norm": 2.5602874148194026,
      "language_loss": 0.68775773,
      "learning_rate": 8.710317248457855e-07,
      "loss": 0.70976156,
      "num_input_tokens_seen": 125342630,
      "step": 5823,
      "time_per_iteration": 2.684260129928589
    },
    {
      "auxiliary_loss_clip": 0.01171549,
      "auxiliary_loss_mlp": 0.01021704,
      "balance_loss_clip": 1.05038142,
      "balance_loss_mlp": 1.01364875,
      "epoch": 0.7002945950820658,
      "flos": 27489762080640.0,
      "grad_norm": 1.9490382525162187,
      "language_loss": 0.72448558,
      "learning_rate": 8.703888130868482e-07,
      "loss": 0.74641806,
      "num_input_tokens_seen": 125364480,
      "step": 5824,
      "time_per_iteration": 2.636254072189331
    },
    {
      "auxiliary_loss_clip": 0.01170132,
      "auxiliary_loss_mlp": 0.01024839,
      "balance_loss_clip": 1.0485431,
      "balance_loss_mlp": 1.01732934,
      "epoch": 0.7004148379727049,
      "flos": 22158463800960.0,
      "grad_norm": 1.9468919798919868,
      "language_loss": 0.81883574,
      "learning_rate": 8.697460726890307e-07,
      "loss": 0.84078538,
      "num_input_tokens_seen": 125381625,
      "step": 5825,
      "time_per_iteration": 2.615417718887329
    },
    {
      "auxiliary_loss_clip": 0.01172164,
      "auxiliary_loss_mlp": 0.0102845,
      "balance_loss_clip": 1.04873896,
      "balance_loss_mlp": 1.02043939,
      "epoch": 0.7005350808633439,
      "flos": 19423758764160.0,
      "grad_norm": 2.3464245052665227,
      "language_loss": 0.90439093,
      "learning_rate": 8.691035037498354e-07,
      "loss": 0.92639709,
      "num_input_tokens_seen": 125397615,
      "step": 5826,
      "time_per_iteration": 3.5217339992523193
    },
    {
      "auxiliary_loss_clip": 0.0117064,
      "auxiliary_loss_mlp": 0.0102372,
      "balance_loss_clip": 1.04888225,
      "balance_loss_mlp": 1.01642752,
      "epoch": 0.7006553237539831,
      "flos": 23476708938240.0,
      "grad_norm": 1.6310254174741239,
      "language_loss": 0.72222793,
      "learning_rate": 8.684611063667391e-07,
      "loss": 0.7441715,
      "num_input_tokens_seen": 125418080,
      "step": 5827,
      "time_per_iteration": 2.9573872089385986
    },
    {
      "auxiliary_loss_clip": 0.01172811,
      "auxiliary_loss_mlp": 0.010216,
      "balance_loss_clip": 1.05001724,
      "balance_loss_mlp": 1.01383924,
      "epoch": 0.7007755666446221,
      "flos": 31212872640000.0,
      "grad_norm": 2.1422127916608873,
      "language_loss": 0.77052855,
      "learning_rate": 8.678188806371935e-07,
      "loss": 0.79247266,
      "num_input_tokens_seen": 125440115,
      "step": 5828,
      "time_per_iteration": 3.7735486030578613
    },
    {
      "auxiliary_loss_clip": 0.01167977,
      "auxiliary_loss_mlp": 0.01025323,
      "balance_loss_clip": 1.04857898,
      "balance_loss_mlp": 1.01861119,
      "epoch": 0.7008958095352612,
      "flos": 18149899858560.0,
      "grad_norm": 1.8541826185247763,
      "language_loss": 0.85397112,
      "learning_rate": 8.671768266586228e-07,
      "loss": 0.87590408,
      "num_input_tokens_seen": 125458240,
      "step": 5829,
      "time_per_iteration": 2.594162702560425
    },
    {
      "auxiliary_loss_clip": 0.01175103,
      "auxiliary_loss_mlp": 0.0102541,
      "balance_loss_clip": 1.05163705,
      "balance_loss_mlp": 1.01777434,
      "epoch": 0.7010160524259004,
      "flos": 27452307173760.0,
      "grad_norm": 1.6978831901069111,
      "language_loss": 0.78327996,
      "learning_rate": 8.665349445284275e-07,
      "loss": 0.8052851,
      "num_input_tokens_seen": 125477980,
      "step": 5830,
      "time_per_iteration": 2.6938273906707764
    },
    {
      "auxiliary_loss_clip": 0.011701,
      "auxiliary_loss_mlp": 0.01022669,
      "balance_loss_clip": 1.04856133,
      "balance_loss_mlp": 1.01471746,
      "epoch": 0.7011362953165394,
      "flos": 23842064125440.0,
      "grad_norm": 1.6503596247277255,
      "language_loss": 0.80863541,
      "learning_rate": 8.658932343439799e-07,
      "loss": 0.83056307,
      "num_input_tokens_seen": 125497765,
      "step": 5831,
      "time_per_iteration": 2.6103198528289795
    },
    {
      "auxiliary_loss_clip": 0.01171301,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.04906392,
      "balance_loss_mlp": 1.01814699,
      "epoch": 0.7012565382071785,
      "flos": 24823430582400.0,
      "grad_norm": 1.9433845133063166,
      "language_loss": 0.77421117,
      "learning_rate": 8.65251696202627e-07,
      "loss": 0.79618895,
      "num_input_tokens_seen": 125514145,
      "step": 5832,
      "time_per_iteration": 2.6503546237945557
    },
    {
      "auxiliary_loss_clip": 0.01174238,
      "auxiliary_loss_mlp": 0.0102382,
      "balance_loss_clip": 1.04994249,
      "balance_loss_mlp": 1.01620281,
      "epoch": 0.7013767810978175,
      "flos": 21397445326080.0,
      "grad_norm": 2.315106635763324,
      "language_loss": 0.87520301,
      "learning_rate": 8.646103302016896e-07,
      "loss": 0.8971836,
      "num_input_tokens_seen": 125533115,
      "step": 5833,
      "time_per_iteration": 2.6104681491851807
    },
    {
      "auxiliary_loss_clip": 0.01175074,
      "auxiliary_loss_mlp": 0.01027891,
      "balance_loss_clip": 1.05024445,
      "balance_loss_mlp": 1.01931405,
      "epoch": 0.7014970239884567,
      "flos": 16687150306560.0,
      "grad_norm": 2.2261985215042346,
      "language_loss": 0.88568562,
      "learning_rate": 8.639691364384614e-07,
      "loss": 0.90771532,
      "num_input_tokens_seen": 125550740,
      "step": 5834,
      "time_per_iteration": 2.616144895553589
    },
    {
      "auxiliary_loss_clip": 0.0117338,
      "auxiliary_loss_mlp": 0.01027948,
      "balance_loss_clip": 1.04850841,
      "balance_loss_mlp": 1.02023554,
      "epoch": 0.7016172668790958,
      "flos": 12568268718720.0,
      "grad_norm": 2.07225541168193,
      "language_loss": 0.72759646,
      "learning_rate": 8.633281150102136e-07,
      "loss": 0.74960971,
      "num_input_tokens_seen": 125567590,
      "step": 5835,
      "time_per_iteration": 2.6272265911102295
    },
    {
      "auxiliary_loss_clip": 0.0116896,
      "auxiliary_loss_mlp": 0.01023146,
      "balance_loss_clip": 1.04724896,
      "balance_loss_mlp": 1.01539111,
      "epoch": 0.7017375097697348,
      "flos": 17452729808640.0,
      "grad_norm": 2.6247031872984006,
      "language_loss": 0.67916453,
      "learning_rate": 8.626872660141855e-07,
      "loss": 0.70108557,
      "num_input_tokens_seen": 125585500,
      "step": 5836,
      "time_per_iteration": 2.628304958343506
    },
    {
      "auxiliary_loss_clip": 0.0117024,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.04818225,
      "balance_loss_mlp": 1.01781225,
      "epoch": 0.701857752660374,
      "flos": 18513028402560.0,
      "grad_norm": 1.680099366311922,
      "language_loss": 0.75039393,
      "learning_rate": 8.620465895475957e-07,
      "loss": 0.77235293,
      "num_input_tokens_seen": 125603720,
      "step": 5837,
      "time_per_iteration": 2.5874547958374023
    },
    {
      "auxiliary_loss_clip": 0.01170483,
      "auxiliary_loss_mlp": 0.01025953,
      "balance_loss_clip": 1.05054569,
      "balance_loss_mlp": 1.01882982,
      "epoch": 0.701977995551013,
      "flos": 24425971614720.0,
      "grad_norm": 1.5050676299698522,
      "language_loss": 0.75247401,
      "learning_rate": 8.614060857076333e-07,
      "loss": 0.77443838,
      "num_input_tokens_seen": 125624390,
      "step": 5838,
      "time_per_iteration": 2.7250797748565674
    },
    {
      "auxiliary_loss_clip": 0.01172803,
      "auxiliary_loss_mlp": 0.01021628,
      "balance_loss_clip": 1.05013204,
      "balance_loss_mlp": 1.01412725,
      "epoch": 0.7020982384416521,
      "flos": 23002759958400.0,
      "grad_norm": 2.3575086564109533,
      "language_loss": 0.75105673,
      "learning_rate": 8.60765754591462e-07,
      "loss": 0.77300107,
      "num_input_tokens_seen": 125644085,
      "step": 5839,
      "time_per_iteration": 2.7125322818756104
    },
    {
      "auxiliary_loss_clip": 0.0116981,
      "auxiliary_loss_mlp": 0.01022096,
      "balance_loss_clip": 1.04746866,
      "balance_loss_mlp": 1.01433539,
      "epoch": 0.7022184813322913,
      "flos": 20449080489600.0,
      "grad_norm": 2.041742398790034,
      "language_loss": 0.72939038,
      "learning_rate": 8.601255962962211e-07,
      "loss": 0.75130945,
      "num_input_tokens_seen": 125663095,
      "step": 5840,
      "time_per_iteration": 2.6311194896698
    },
    {
      "auxiliary_loss_clip": 0.01177999,
      "auxiliary_loss_mlp": 0.01031579,
      "balance_loss_clip": 1.05104136,
      "balance_loss_mlp": 1.02245402,
      "epoch": 0.7023387242229303,
      "flos": 19790514581760.0,
      "grad_norm": 3.4323807177479297,
      "language_loss": 0.72555101,
      "learning_rate": 8.594856109190194e-07,
      "loss": 0.74764681,
      "num_input_tokens_seen": 125680125,
      "step": 5841,
      "time_per_iteration": 3.5278751850128174
    },
    {
      "auxiliary_loss_clip": 0.01170987,
      "auxiliary_loss_mlp": 0.0102665,
      "balance_loss_clip": 1.04730749,
      "balance_loss_mlp": 1.01902699,
      "epoch": 0.7024589671135694,
      "flos": 33259278286080.0,
      "grad_norm": 1.835893604645983,
      "language_loss": 0.69295549,
      "learning_rate": 8.588457985569446e-07,
      "loss": 0.71493191,
      "num_input_tokens_seen": 125703035,
      "step": 5842,
      "time_per_iteration": 2.6989285945892334
    },
    {
      "auxiliary_loss_clip": 0.01172214,
      "auxiliary_loss_mlp": 0.01025905,
      "balance_loss_clip": 1.04850864,
      "balance_loss_mlp": 1.01828718,
      "epoch": 0.7025792100042085,
      "flos": 19098982967040.0,
      "grad_norm": 2.1692209192844705,
      "language_loss": 0.72018087,
      "learning_rate": 8.582061593070542e-07,
      "loss": 0.74216205,
      "num_input_tokens_seen": 125723765,
      "step": 5843,
      "time_per_iteration": 2.6467669010162354
    },
    {
      "auxiliary_loss_clip": 0.01172487,
      "auxiliary_loss_mlp": 0.01025643,
      "balance_loss_clip": 1.04894972,
      "balance_loss_mlp": 1.0180372,
      "epoch": 0.7026994528948476,
      "flos": 18952611045120.0,
      "grad_norm": 3.0732994023121845,
      "language_loss": 0.76791596,
      "learning_rate": 8.57566693266383e-07,
      "loss": 0.7898972,
      "num_input_tokens_seen": 125741455,
      "step": 5844,
      "time_per_iteration": 2.5780129432678223
    },
    {
      "auxiliary_loss_clip": 0.0117027,
      "auxiliary_loss_mlp": 0.01028174,
      "balance_loss_clip": 1.04688978,
      "balance_loss_mlp": 1.02010322,
      "epoch": 0.7028196957854866,
      "flos": 19536662188800.0,
      "grad_norm": 2.6998439227823834,
      "language_loss": 0.69214106,
      "learning_rate": 8.569274005319354e-07,
      "loss": 0.71412551,
      "num_input_tokens_seen": 125759855,
      "step": 5845,
      "time_per_iteration": 2.577169179916382
    },
    {
      "auxiliary_loss_clip": 0.01172646,
      "auxiliary_loss_mlp": 0.01032393,
      "balance_loss_clip": 1.05001533,
      "balance_loss_mlp": 1.02436125,
      "epoch": 0.7029399386761258,
      "flos": 20845318394880.0,
      "grad_norm": 2.499753003962324,
      "language_loss": 0.79685032,
      "learning_rate": 8.562882812006913e-07,
      "loss": 0.8189007,
      "num_input_tokens_seen": 125777345,
      "step": 5846,
      "time_per_iteration": 2.5651094913482666
    },
    {
      "auxiliary_loss_clip": 0.01170474,
      "auxiliary_loss_mlp": 0.01026679,
      "balance_loss_clip": 1.04825437,
      "balance_loss_mlp": 1.01914227,
      "epoch": 0.7030601815667649,
      "flos": 22055005653120.0,
      "grad_norm": 2.149797200874436,
      "language_loss": 0.77706486,
      "learning_rate": 8.556493353696066e-07,
      "loss": 0.79903638,
      "num_input_tokens_seen": 125796345,
      "step": 5847,
      "time_per_iteration": 2.6058919429779053
    },
    {
      "auxiliary_loss_clip": 0.01174535,
      "auxiliary_loss_mlp": 0.01030665,
      "balance_loss_clip": 1.05060112,
      "balance_loss_mlp": 1.0221417,
      "epoch": 0.7031804244574039,
      "flos": 27198742089600.0,
      "grad_norm": 2.308350938089843,
      "language_loss": 0.68237615,
      "learning_rate": 8.550105631356077e-07,
      "loss": 0.7044282,
      "num_input_tokens_seen": 125816070,
      "step": 5848,
      "time_per_iteration": 2.6042230129241943
    },
    {
      "auxiliary_loss_clip": 0.01171205,
      "auxiliary_loss_mlp": 0.01023812,
      "balance_loss_clip": 1.04880393,
      "balance_loss_mlp": 1.01562285,
      "epoch": 0.7033006673480431,
      "flos": 22379853277440.0,
      "grad_norm": 2.6493088647173164,
      "language_loss": 0.77194685,
      "learning_rate": 8.543719645955961e-07,
      "loss": 0.79389703,
      "num_input_tokens_seen": 125834400,
      "step": 5849,
      "time_per_iteration": 2.653477191925049
    },
    {
      "auxiliary_loss_clip": 0.01171302,
      "auxiliary_loss_mlp": 0.0102436,
      "balance_loss_clip": 1.04706013,
      "balance_loss_mlp": 1.01627123,
      "epoch": 0.7034209102386821,
      "flos": 24715986024960.0,
      "grad_norm": 1.6826630680162782,
      "language_loss": 0.74803507,
      "learning_rate": 8.537335398464467e-07,
      "loss": 0.76999176,
      "num_input_tokens_seen": 125854720,
      "step": 5850,
      "time_per_iteration": 2.669689178466797
    },
    {
      "auxiliary_loss_clip": 0.01172934,
      "auxiliary_loss_mlp": 0.0102787,
      "balance_loss_clip": 1.04814076,
      "balance_loss_mlp": 1.01938796,
      "epoch": 0.7035411531293212,
      "flos": 22556174163840.0,
      "grad_norm": 2.468329535682835,
      "language_loss": 0.85202891,
      "learning_rate": 8.53095288985007e-07,
      "loss": 0.87403697,
      "num_input_tokens_seen": 125868455,
      "step": 5851,
      "time_per_iteration": 2.624257802963257
    },
    {
      "auxiliary_loss_clip": 0.01167482,
      "auxiliary_loss_mlp": 0.01021298,
      "balance_loss_clip": 1.04741859,
      "balance_loss_mlp": 1.01448202,
      "epoch": 0.7036613960199604,
      "flos": 22674967418880.0,
      "grad_norm": 1.7051336211967763,
      "language_loss": 0.82568544,
      "learning_rate": 8.524572121081009e-07,
      "loss": 0.84757328,
      "num_input_tokens_seen": 125888555,
      "step": 5852,
      "time_per_iteration": 3.5747077465057373
    },
    {
      "auxiliary_loss_clip": 0.01174199,
      "auxiliary_loss_mlp": 0.01029485,
      "balance_loss_clip": 1.0492866,
      "balance_loss_mlp": 1.02147722,
      "epoch": 0.7037816389105994,
      "flos": 22492146170880.0,
      "grad_norm": 1.9332550790628904,
      "language_loss": 0.62566924,
      "learning_rate": 8.518193093125232e-07,
      "loss": 0.64770609,
      "num_input_tokens_seen": 125907610,
      "step": 5853,
      "time_per_iteration": 2.617488384246826
    },
    {
      "auxiliary_loss_clip": 0.01171002,
      "auxiliary_loss_mlp": 0.01023126,
      "balance_loss_clip": 1.04862607,
      "balance_loss_mlp": 1.0158844,
      "epoch": 0.7039018818012385,
      "flos": 27087490690560.0,
      "grad_norm": 1.7890250876225884,
      "language_loss": 0.81076849,
      "learning_rate": 8.511815806950436e-07,
      "loss": 0.83270979,
      "num_input_tokens_seen": 125928640,
      "step": 5854,
      "time_per_iteration": 3.6122331619262695
    },
    {
      "auxiliary_loss_clip": 0.0116741,
      "auxiliary_loss_mlp": 0.01021,
      "balance_loss_clip": 1.04525173,
      "balance_loss_mlp": 1.01296544,
      "epoch": 0.7040221246918776,
      "flos": 17749819198080.0,
      "grad_norm": 3.595183743845393,
      "language_loss": 0.7777915,
      "learning_rate": 8.505440263524044e-07,
      "loss": 0.79967558,
      "num_input_tokens_seen": 125947485,
      "step": 5855,
      "time_per_iteration": 3.484596014022827
    },
    {
      "auxiliary_loss_clip": 0.01174882,
      "auxiliary_loss_mlp": 0.01024838,
      "balance_loss_clip": 1.0499047,
      "balance_loss_mlp": 1.01663649,
      "epoch": 0.7041423675825167,
      "flos": 16279851012480.0,
      "grad_norm": 2.38816463858539,
      "language_loss": 0.88353467,
      "learning_rate": 8.49906646381322e-07,
      "loss": 0.90553182,
      "num_input_tokens_seen": 125960320,
      "step": 5856,
      "time_per_iteration": 2.547606945037842
    },
    {
      "auxiliary_loss_clip": 0.01170961,
      "auxiliary_loss_mlp": 0.01022504,
      "balance_loss_clip": 1.04857039,
      "balance_loss_mlp": 1.01505983,
      "epoch": 0.7042626104731557,
      "flos": 25483181639040.0,
      "grad_norm": 3.288888389559155,
      "language_loss": 0.72276473,
      "learning_rate": 8.492694408784884e-07,
      "loss": 0.74469936,
      "num_input_tokens_seen": 125980575,
      "step": 5857,
      "time_per_iteration": 2.6975810527801514
    },
    {
      "auxiliary_loss_clip": 0.01172293,
      "auxiliary_loss_mlp": 0.01029775,
      "balance_loss_clip": 1.04908752,
      "balance_loss_mlp": 1.02230382,
      "epoch": 0.7043828533637949,
      "flos": 17857622891520.0,
      "grad_norm": 2.4125023006944235,
      "language_loss": 0.6315825,
      "learning_rate": 8.486324099405642e-07,
      "loss": 0.6536032,
      "num_input_tokens_seen": 125997420,
      "step": 5858,
      "time_per_iteration": 2.552983522415161
    },
    {
      "auxiliary_loss_clip": 0.011685,
      "auxiliary_loss_mlp": 0.01024298,
      "balance_loss_clip": 1.04582727,
      "balance_loss_mlp": 1.01684177,
      "epoch": 0.704503096254434,
      "flos": 29494259533440.0,
      "grad_norm": 2.0775196877755153,
      "language_loss": 0.74965453,
      "learning_rate": 8.479955536641887e-07,
      "loss": 0.77158254,
      "num_input_tokens_seen": 126018915,
      "step": 5859,
      "time_per_iteration": 2.7240161895751953
    },
    {
      "auxiliary_loss_clip": 0.01170536,
      "auxiliary_loss_mlp": 0.01023108,
      "balance_loss_clip": 1.049034,
      "balance_loss_mlp": 1.0151391,
      "epoch": 0.704623339145073,
      "flos": 30920739327360.0,
      "grad_norm": 2.886879189228161,
      "language_loss": 0.66426581,
      "learning_rate": 8.473588721459716e-07,
      "loss": 0.68620229,
      "num_input_tokens_seen": 126038825,
      "step": 5860,
      "time_per_iteration": 2.6827926635742188
    },
    {
      "auxiliary_loss_clip": 0.01178228,
      "auxiliary_loss_mlp": 0.01035468,
      "balance_loss_clip": 1.05335414,
      "balance_loss_mlp": 1.02670002,
      "epoch": 0.7047435820357122,
      "flos": 23914747296000.0,
      "grad_norm": 1.9881647716806157,
      "language_loss": 0.71003312,
      "learning_rate": 8.467223654824967e-07,
      "loss": 0.7321701,
      "num_input_tokens_seen": 126058280,
      "step": 5861,
      "time_per_iteration": 2.656920909881592
    },
    {
      "auxiliary_loss_clip": 0.01175042,
      "auxiliary_loss_mlp": 0.01023154,
      "balance_loss_clip": 1.05074763,
      "balance_loss_mlp": 1.01539338,
      "epoch": 0.7048638249263512,
      "flos": 46494010926720.0,
      "grad_norm": 2.031896643040756,
      "language_loss": 0.62153268,
      "learning_rate": 8.460860337703233e-07,
      "loss": 0.64351463,
      "num_input_tokens_seen": 126078885,
      "step": 5862,
      "time_per_iteration": 2.792757987976074
    },
    {
      "auxiliary_loss_clip": 0.01175202,
      "auxiliary_loss_mlp": 0.01026715,
      "balance_loss_clip": 1.04958832,
      "balance_loss_mlp": 1.01838183,
      "epoch": 0.7049840678169903,
      "flos": 21689219502720.0,
      "grad_norm": 1.8770523227304907,
      "language_loss": 0.70718235,
      "learning_rate": 8.454498771059797e-07,
      "loss": 0.7292015,
      "num_input_tokens_seen": 126098260,
      "step": 5863,
      "time_per_iteration": 2.6069045066833496
    },
    {
      "auxiliary_loss_clip": 0.01175481,
      "auxiliary_loss_mlp": 0.01030789,
      "balance_loss_clip": 1.04994416,
      "balance_loss_mlp": 1.02307606,
      "epoch": 0.7051043107076294,
      "flos": 18405081054720.0,
      "grad_norm": 3.158708878607183,
      "language_loss": 0.83263826,
      "learning_rate": 8.448138955859725e-07,
      "loss": 0.85470092,
      "num_input_tokens_seen": 126114845,
      "step": 5864,
      "time_per_iteration": 2.542188882827759
    },
    {
      "auxiliary_loss_clip": 0.0117508,
      "auxiliary_loss_mlp": 0.01027016,
      "balance_loss_clip": 1.05154431,
      "balance_loss_mlp": 1.01914799,
      "epoch": 0.7052245535982685,
      "flos": 19319043640320.0,
      "grad_norm": 1.9298561254013615,
      "language_loss": 0.90105391,
      "learning_rate": 8.44178089306778e-07,
      "loss": 0.92307484,
      "num_input_tokens_seen": 126132780,
      "step": 5865,
      "time_per_iteration": 2.6349892616271973
    },
    {
      "auxiliary_loss_clip": 0.0117134,
      "auxiliary_loss_mlp": 0.01026596,
      "balance_loss_clip": 1.04994416,
      "balance_loss_mlp": 1.01938939,
      "epoch": 0.7053447964889076,
      "flos": 19062138591360.0,
      "grad_norm": 1.872041292447931,
      "language_loss": 0.76939046,
      "learning_rate": 8.4354245836485e-07,
      "loss": 0.79136986,
      "num_input_tokens_seen": 126151225,
      "step": 5866,
      "time_per_iteration": 2.7318058013916016
    },
    {
      "auxiliary_loss_clip": 0.0117609,
      "auxiliary_loss_mlp": 0.01031634,
      "balance_loss_clip": 1.05003905,
      "balance_loss_mlp": 1.02328944,
      "epoch": 0.7054650393795466,
      "flos": 27379228953600.0,
      "grad_norm": 1.6035686188801823,
      "language_loss": 0.73325467,
      "learning_rate": 8.429070028566108e-07,
      "loss": 0.75533199,
      "num_input_tokens_seen": 126172535,
      "step": 5867,
      "time_per_iteration": 2.685758590698242
    },
    {
      "auxiliary_loss_clip": 0.01171652,
      "auxiliary_loss_mlp": 0.01024374,
      "balance_loss_clip": 1.04966903,
      "balance_loss_mlp": 1.01668549,
      "epoch": 0.7055852822701858,
      "flos": 16102201322880.0,
      "grad_norm": 2.5557330835791667,
      "language_loss": 0.74834609,
      "learning_rate": 8.422717228784586e-07,
      "loss": 0.77030635,
      "num_input_tokens_seen": 126189410,
      "step": 5868,
      "time_per_iteration": 2.645063877105713
    },
    {
      "auxiliary_loss_clip": 0.01173298,
      "auxiliary_loss_mlp": 0.01031761,
      "balance_loss_clip": 1.04948843,
      "balance_loss_mlp": 1.02313066,
      "epoch": 0.7057055251608249,
      "flos": 11692299744000.0,
      "grad_norm": 3.491036409369266,
      "language_loss": 0.69299644,
      "learning_rate": 8.416366185267663e-07,
      "loss": 0.71504706,
      "num_input_tokens_seen": 126206910,
      "step": 5869,
      "time_per_iteration": 3.493593215942383
    },
    {
      "auxiliary_loss_clip": 0.01168025,
      "auxiliary_loss_mlp": 0.01022834,
      "balance_loss_clip": 1.04686368,
      "balance_loss_mlp": 1.01575291,
      "epoch": 0.7058257680514639,
      "flos": 22711560399360.0,
      "grad_norm": 2.098740207796172,
      "language_loss": 0.77926421,
      "learning_rate": 8.410016898978778e-07,
      "loss": 0.80117279,
      "num_input_tokens_seen": 126224385,
      "step": 5870,
      "time_per_iteration": 2.674175977706909
    },
    {
      "auxiliary_loss_clip": 0.01166075,
      "auxiliary_loss_mlp": 0.01023063,
      "balance_loss_clip": 1.04574561,
      "balance_loss_mlp": 1.01545143,
      "epoch": 0.7059460109421031,
      "flos": 17529543043200.0,
      "grad_norm": 1.7373892727045959,
      "language_loss": 0.79149085,
      "learning_rate": 8.403669370881115e-07,
      "loss": 0.81338227,
      "num_input_tokens_seen": 126243120,
      "step": 5871,
      "time_per_iteration": 2.6125426292419434
    },
    {
      "auxiliary_loss_clip": 0.01172088,
      "auxiliary_loss_mlp": 0.01022811,
      "balance_loss_clip": 1.050282,
      "balance_loss_mlp": 1.01555753,
      "epoch": 0.7060662538327421,
      "flos": 23544687427200.0,
      "grad_norm": 3.1810682848088105,
      "language_loss": 0.78630149,
      "learning_rate": 8.397323601937587e-07,
      "loss": 0.80825043,
      "num_input_tokens_seen": 126263020,
      "step": 5872,
      "time_per_iteration": 2.68342924118042
    },
    {
      "auxiliary_loss_clip": 0.01170575,
      "auxiliary_loss_mlp": 0.01027213,
      "balance_loss_clip": 1.04854119,
      "balance_loss_mlp": 1.01970267,
      "epoch": 0.7061864967233812,
      "flos": 30260736875520.0,
      "grad_norm": 2.5106241213721114,
      "language_loss": 0.77026939,
      "learning_rate": 8.390979593110838e-07,
      "loss": 0.79224724,
      "num_input_tokens_seen": 126285150,
      "step": 5873,
      "time_per_iteration": 2.7160942554473877
    },
    {
      "auxiliary_loss_clip": 0.01174716,
      "auxiliary_loss_mlp": 0.01032818,
      "balance_loss_clip": 1.05040121,
      "balance_loss_mlp": 1.0241816,
      "epoch": 0.7063067396140204,
      "flos": 20701460424960.0,
      "grad_norm": 1.9491288488455163,
      "language_loss": 0.81567889,
      "learning_rate": 8.384637345363262e-07,
      "loss": 0.83775425,
      "num_input_tokens_seen": 126304340,
      "step": 5874,
      "time_per_iteration": 2.615093469619751
    },
    {
      "auxiliary_loss_clip": 0.01171111,
      "auxiliary_loss_mlp": 0.01027706,
      "balance_loss_clip": 1.04797125,
      "balance_loss_mlp": 1.01984131,
      "epoch": 0.7064269825046594,
      "flos": 32266168081920.0,
      "grad_norm": 2.018758332475016,
      "language_loss": 0.76505214,
      "learning_rate": 8.378296859656964e-07,
      "loss": 0.78704029,
      "num_input_tokens_seen": 126325495,
      "step": 5875,
      "time_per_iteration": 2.662187337875366
    },
    {
      "auxiliary_loss_clip": 0.01173538,
      "auxiliary_loss_mlp": 0.01028795,
      "balance_loss_clip": 1.05162239,
      "balance_loss_mlp": 1.02080166,
      "epoch": 0.7065472253952985,
      "flos": 30227124723840.0,
      "grad_norm": 2.218114491590927,
      "language_loss": 0.68208474,
      "learning_rate": 8.371958136953792e-07,
      "loss": 0.70410812,
      "num_input_tokens_seen": 126345525,
      "step": 5876,
      "time_per_iteration": 2.888139247894287
    },
    {
      "auxiliary_loss_clip": 0.01175278,
      "auxiliary_loss_mlp": 0.01025542,
      "balance_loss_clip": 1.05046546,
      "balance_loss_mlp": 1.01693535,
      "epoch": 0.7066674682859376,
      "flos": 16216720859520.0,
      "grad_norm": 2.5296118865134996,
      "language_loss": 0.66542125,
      "learning_rate": 8.365621178215326e-07,
      "loss": 0.68742943,
      "num_input_tokens_seen": 126361995,
      "step": 5877,
      "time_per_iteration": 2.5699706077575684
    },
    {
      "auxiliary_loss_clip": 0.01173472,
      "auxiliary_loss_mlp": 0.01026529,
      "balance_loss_clip": 1.05059767,
      "balance_loss_mlp": 1.01909006,
      "epoch": 0.7067877111765767,
      "flos": 14830461319680.0,
      "grad_norm": 2.266230220820261,
      "language_loss": 0.75386715,
      "learning_rate": 8.359285984402871e-07,
      "loss": 0.77586716,
      "num_input_tokens_seen": 126379260,
      "step": 5878,
      "time_per_iteration": 2.5864951610565186
    },
    {
      "auxiliary_loss_clip": 0.01168798,
      "auxiliary_loss_mlp": 0.01024181,
      "balance_loss_clip": 1.04709125,
      "balance_loss_mlp": 1.0172317,
      "epoch": 0.7069079540672157,
      "flos": 25440196037760.0,
      "grad_norm": 2.038022476870566,
      "language_loss": 0.74325609,
      "learning_rate": 8.352952556477489e-07,
      "loss": 0.76518589,
      "num_input_tokens_seen": 126397170,
      "step": 5879,
      "time_per_iteration": 3.5737524032592773
    },
    {
      "auxiliary_loss_clip": 0.01171041,
      "auxiliary_loss_mlp": 0.01024474,
      "balance_loss_clip": 1.0494616,
      "balance_loss_mlp": 1.01678491,
      "epoch": 0.7070281969578549,
      "flos": 24607751368320.0,
      "grad_norm": 2.044979659612839,
      "language_loss": 0.76897717,
      "learning_rate": 8.34662089539993e-07,
      "loss": 0.7909323,
      "num_input_tokens_seen": 126416680,
      "step": 5880,
      "time_per_iteration": 3.5850260257720947
    },
    {
      "auxiliary_loss_clip": 0.01168668,
      "auxiliary_loss_mlp": 0.01022513,
      "balance_loss_clip": 1.04818344,
      "balance_loss_mlp": 1.01507998,
      "epoch": 0.707148439848494,
      "flos": 26724469887360.0,
      "grad_norm": 2.174892894642784,
      "language_loss": 0.79538411,
      "learning_rate": 8.340291002130722e-07,
      "loss": 0.81729591,
      "num_input_tokens_seen": 126435870,
      "step": 5881,
      "time_per_iteration": 2.6380455493927
    },
    {
      "auxiliary_loss_clip": 0.01175042,
      "auxiliary_loss_mlp": 0.01030891,
      "balance_loss_clip": 1.05094659,
      "balance_loss_mlp": 1.02239728,
      "epoch": 0.707268682739133,
      "flos": 15085750256640.0,
      "grad_norm": 3.888418703393941,
      "language_loss": 0.79999477,
      "learning_rate": 8.3339628776301e-07,
      "loss": 0.82205403,
      "num_input_tokens_seen": 126454010,
      "step": 5882,
      "time_per_iteration": 3.479482650756836
    },
    {
      "auxiliary_loss_clip": 0.01170697,
      "auxiliary_loss_mlp": 0.01027604,
      "balance_loss_clip": 1.04852605,
      "balance_loss_mlp": 1.01961088,
      "epoch": 0.7073889256297722,
      "flos": 34313148345600.0,
      "grad_norm": 2.9633553508725563,
      "language_loss": 0.57390678,
      "learning_rate": 8.327636522858033e-07,
      "loss": 0.59588975,
      "num_input_tokens_seen": 126473615,
      "step": 5883,
      "time_per_iteration": 2.721792221069336
    },
    {
      "auxiliary_loss_clip": 0.01172094,
      "auxiliary_loss_mlp": 0.01027473,
      "balance_loss_clip": 1.04951441,
      "balance_loss_mlp": 1.01961696,
      "epoch": 0.7075091685204112,
      "flos": 20083940784000.0,
      "grad_norm": 2.682837591910758,
      "language_loss": 0.7748307,
      "learning_rate": 8.321311938774225e-07,
      "loss": 0.79682636,
      "num_input_tokens_seen": 126492705,
      "step": 5884,
      "time_per_iteration": 2.642468214035034
    },
    {
      "auxiliary_loss_clip": 0.01175195,
      "auxiliary_loss_mlp": 0.01027055,
      "balance_loss_clip": 1.05077052,
      "balance_loss_mlp": 1.0195632,
      "epoch": 0.7076294114110503,
      "flos": 20777124424320.0,
      "grad_norm": 1.9030572506469265,
      "language_loss": 0.80099487,
      "learning_rate": 8.314989126338104e-07,
      "loss": 0.82301736,
      "num_input_tokens_seen": 126512715,
      "step": 5885,
      "time_per_iteration": 2.6021668910980225
    },
    {
      "auxiliary_loss_clip": 0.01169549,
      "auxiliary_loss_mlp": 0.01021937,
      "balance_loss_clip": 1.04693747,
      "balance_loss_mlp": 1.01414084,
      "epoch": 0.7077496543016895,
      "flos": 17967689141760.0,
      "grad_norm": 2.0847736887591872,
      "language_loss": 0.84387922,
      "learning_rate": 8.308668086508847e-07,
      "loss": 0.86579406,
      "num_input_tokens_seen": 126530795,
      "step": 5886,
      "time_per_iteration": 2.599667549133301
    },
    {
      "auxiliary_loss_clip": 0.01173248,
      "auxiliary_loss_mlp": 0.01025632,
      "balance_loss_clip": 1.0498426,
      "balance_loss_mlp": 1.01759148,
      "epoch": 0.7078698971923285,
      "flos": 45478098564480.0,
      "grad_norm": 3.003724559027487,
      "language_loss": 0.7401017,
      "learning_rate": 8.302348820245342e-07,
      "loss": 0.7620905,
      "num_input_tokens_seen": 126553360,
      "step": 5887,
      "time_per_iteration": 2.7854390144348145
    },
    {
      "auxiliary_loss_clip": 0.01175684,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.05052805,
      "balance_loss_mlp": 1.01675451,
      "epoch": 0.7079901400829676,
      "flos": 26943704547840.0,
      "grad_norm": 2.428472466158505,
      "language_loss": 0.70111394,
      "learning_rate": 8.296031328506232e-07,
      "loss": 0.72313297,
      "num_input_tokens_seen": 126573110,
      "step": 5888,
      "time_per_iteration": 2.668137311935425
    },
    {
      "auxiliary_loss_clip": 0.01171651,
      "auxiliary_loss_mlp": 0.01028675,
      "balance_loss_clip": 1.04941404,
      "balance_loss_mlp": 1.02056623,
      "epoch": 0.7081103829736067,
      "flos": 24423206267520.0,
      "grad_norm": 1.9463512100046074,
      "language_loss": 0.75941008,
      "learning_rate": 8.289715612249857e-07,
      "loss": 0.78141344,
      "num_input_tokens_seen": 126593725,
      "step": 5889,
      "time_per_iteration": 2.606247663497925
    },
    {
      "auxiliary_loss_clip": 0.01172911,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.04974437,
      "balance_loss_mlp": 1.01722324,
      "epoch": 0.7082306258642458,
      "flos": 18543300589440.0,
      "grad_norm": 5.248648401092586,
      "language_loss": 0.77664232,
      "learning_rate": 8.283401672434305e-07,
      "loss": 0.79862344,
      "num_input_tokens_seen": 126608950,
      "step": 5890,
      "time_per_iteration": 2.6163763999938965
    },
    {
      "auxiliary_loss_clip": 0.01169855,
      "auxiliary_loss_mlp": 0.01028189,
      "balance_loss_clip": 1.04763603,
      "balance_loss_mlp": 1.0209229,
      "epoch": 0.7083508687548848,
      "flos": 23477534951040.0,
      "grad_norm": 2.7034166502832266,
      "language_loss": 0.70601219,
      "learning_rate": 8.277089510017412e-07,
      "loss": 0.72799259,
      "num_input_tokens_seen": 126629755,
      "step": 5891,
      "time_per_iteration": 2.6446900367736816
    },
    {
      "auxiliary_loss_clip": 0.011711,
      "auxiliary_loss_mlp": 0.01032614,
      "balance_loss_clip": 1.04966819,
      "balance_loss_mlp": 1.02505589,
      "epoch": 0.708471111645524,
      "flos": 22419463000320.0,
      "grad_norm": 1.83253247780486,
      "language_loss": 0.82587039,
      "learning_rate": 8.270779125956719e-07,
      "loss": 0.84790754,
      "num_input_tokens_seen": 126650135,
      "step": 5892,
      "time_per_iteration": 2.571406126022339
    },
    {
      "auxiliary_loss_clip": 0.01171605,
      "auxiliary_loss_mlp": 0.01021723,
      "balance_loss_clip": 1.05040646,
      "balance_loss_mlp": 1.01381958,
      "epoch": 0.7085913545361631,
      "flos": 20922885815040.0,
      "grad_norm": 2.065148664796075,
      "language_loss": 0.80125189,
      "learning_rate": 8.264470521209505e-07,
      "loss": 0.82318515,
      "num_input_tokens_seen": 126668500,
      "step": 5893,
      "time_per_iteration": 2.6715073585510254
    },
    {
      "auxiliary_loss_clip": 0.01171361,
      "auxiliary_loss_mlp": 0.01029145,
      "balance_loss_clip": 1.04811418,
      "balance_loss_mlp": 1.02116358,
      "epoch": 0.7087115974268021,
      "flos": 15012384727680.0,
      "grad_norm": 2.634047433171405,
      "language_loss": 0.76555288,
      "learning_rate": 8.258163696732785e-07,
      "loss": 0.78755796,
      "num_input_tokens_seen": 126686090,
      "step": 5894,
      "time_per_iteration": 3.672025442123413
    },
    {
      "auxiliary_loss_clip": 0.0117249,
      "auxiliary_loss_mlp": 0.0102446,
      "balance_loss_clip": 1.05008483,
      "balance_loss_mlp": 1.01706934,
      "epoch": 0.7088318403174413,
      "flos": 21539040739200.0,
      "grad_norm": 3.9881701109390097,
      "language_loss": 0.77014637,
      "learning_rate": 8.251858653483288e-07,
      "loss": 0.79211593,
      "num_input_tokens_seen": 126704255,
      "step": 5895,
      "time_per_iteration": 2.663529872894287
    },
    {
      "auxiliary_loss_clip": 0.01173531,
      "auxiliary_loss_mlp": 0.01026842,
      "balance_loss_clip": 1.05118001,
      "balance_loss_mlp": 1.01849699,
      "epoch": 0.7089520832080803,
      "flos": 15516785462400.0,
      "grad_norm": 2.238466763981791,
      "language_loss": 0.8580482,
      "learning_rate": 8.245555392417501e-07,
      "loss": 0.88005197,
      "num_input_tokens_seen": 126718910,
      "step": 5896,
      "time_per_iteration": 2.698425769805908
    },
    {
      "auxiliary_loss_clip": 0.01171399,
      "auxiliary_loss_mlp": 0.01025376,
      "balance_loss_clip": 1.04865479,
      "balance_loss_mlp": 1.01773441,
      "epoch": 0.7090723260987194,
      "flos": 20412667077120.0,
      "grad_norm": 2.040968898073895,
      "language_loss": 0.79063421,
      "learning_rate": 8.239253914491613e-07,
      "loss": 0.81260192,
      "num_input_tokens_seen": 126737235,
      "step": 5897,
      "time_per_iteration": 2.7223222255706787
    },
    {
      "auxiliary_loss_clip": 0.01175418,
      "auxiliary_loss_mlp": 0.01022539,
      "balance_loss_clip": 1.05091393,
      "balance_loss_mlp": 1.01412845,
      "epoch": 0.7091925689893585,
      "flos": 25668337271040.0,
      "grad_norm": 1.8523265616423514,
      "language_loss": 0.75031418,
      "learning_rate": 8.232954220661556e-07,
      "loss": 0.77229375,
      "num_input_tokens_seen": 126759970,
      "step": 5898,
      "time_per_iteration": 2.672661066055298
    },
    {
      "auxiliary_loss_clip": 0.01171903,
      "auxiliary_loss_mlp": 0.01025189,
      "balance_loss_clip": 1.0507704,
      "balance_loss_mlp": 1.01740479,
      "epoch": 0.7093128118799976,
      "flos": 24206629213440.0,
      "grad_norm": 4.023656162083417,
      "language_loss": 0.7014761,
      "learning_rate": 8.226656311882989e-07,
      "loss": 0.72344702,
      "num_input_tokens_seen": 126779280,
      "step": 5899,
      "time_per_iteration": 2.6459734439849854
    },
    {
      "auxiliary_loss_clip": 0.01170881,
      "auxiliary_loss_mlp": 0.01027858,
      "balance_loss_clip": 1.05042243,
      "balance_loss_mlp": 1.01996088,
      "epoch": 0.7094330547706367,
      "flos": 16646786398080.0,
      "grad_norm": 2.0967517751096274,
      "language_loss": 0.7645247,
      "learning_rate": 8.22036018911129e-07,
      "loss": 0.78651214,
      "num_input_tokens_seen": 126797310,
      "step": 5900,
      "time_per_iteration": 2.657661199569702
    },
    {
      "auxiliary_loss_clip": 0.01175094,
      "auxiliary_loss_mlp": 0.01026486,
      "balance_loss_clip": 1.05050814,
      "balance_loss_mlp": 1.01789117,
      "epoch": 0.7095532976612757,
      "flos": 16283370545280.0,
      "grad_norm": 2.1145137284334354,
      "language_loss": 0.8048569,
      "learning_rate": 8.214065853301599e-07,
      "loss": 0.82687259,
      "num_input_tokens_seen": 126812840,
      "step": 5901,
      "time_per_iteration": 2.591831922531128
    },
    {
      "auxiliary_loss_clip": 0.01070763,
      "auxiliary_loss_mlp": 0.01000857,
      "balance_loss_clip": 1.0127517,
      "balance_loss_mlp": 0.99975407,
      "epoch": 0.7096735405519149,
      "flos": 70722080559360.0,
      "grad_norm": 0.8048142877788184,
      "language_loss": 0.58200336,
      "learning_rate": 8.207773305408734e-07,
      "loss": 0.60271955,
      "num_input_tokens_seen": 126880060,
      "step": 5902,
      "time_per_iteration": 3.279432773590088
    },
    {
      "auxiliary_loss_clip": 0.01173706,
      "auxiliary_loss_mlp": 0.01021299,
      "balance_loss_clip": 1.04932833,
      "balance_loss_mlp": 1.01344872,
      "epoch": 0.709793783442554,
      "flos": 23621500661760.0,
      "grad_norm": 4.425411723161438,
      "language_loss": 0.79951704,
      "learning_rate": 8.201482546387288e-07,
      "loss": 0.82146716,
      "num_input_tokens_seen": 126899535,
      "step": 5903,
      "time_per_iteration": 2.598283529281616
    },
    {
      "auxiliary_loss_clip": 0.01170511,
      "auxiliary_loss_mlp": 0.01027677,
      "balance_loss_clip": 1.04932547,
      "balance_loss_mlp": 1.02009583,
      "epoch": 0.709914026333193,
      "flos": 25993472204160.0,
      "grad_norm": 1.95263865736394,
      "language_loss": 0.91744399,
      "learning_rate": 8.195193577191553e-07,
      "loss": 0.93942583,
      "num_input_tokens_seen": 126921365,
      "step": 5904,
      "time_per_iteration": 2.784304618835449
    },
    {
      "auxiliary_loss_clip": 0.01171616,
      "auxiliary_loss_mlp": 0.01025857,
      "balance_loss_clip": 1.04881334,
      "balance_loss_mlp": 1.01828754,
      "epoch": 0.7100342692238322,
      "flos": 24861531934080.0,
      "grad_norm": 2.03934025604201,
      "language_loss": 0.8436082,
      "learning_rate": 8.188906398775579e-07,
      "loss": 0.86558294,
      "num_input_tokens_seen": 126941910,
      "step": 5905,
      "time_per_iteration": 3.5190141201019287
    },
    {
      "auxiliary_loss_clip": 0.01171956,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.04727626,
      "balance_loss_mlp": 1.01764691,
      "epoch": 0.7101545121144712,
      "flos": 24932203943040.0,
      "grad_norm": 2.150514888443473,
      "language_loss": 0.6869638,
      "learning_rate": 8.18262101209311e-07,
      "loss": 0.70894432,
      "num_input_tokens_seen": 126961120,
      "step": 5906,
      "time_per_iteration": 2.7107324600219727
    },
    {
      "auxiliary_loss_clip": 0.01171213,
      "auxiliary_loss_mlp": 0.0102426,
      "balance_loss_clip": 1.04856622,
      "balance_loss_mlp": 1.01654148,
      "epoch": 0.7102747550051103,
      "flos": 23768842250880.0,
      "grad_norm": 2.33117306053922,
      "language_loss": 0.70209366,
      "learning_rate": 8.176337418097626e-07,
      "loss": 0.72404838,
      "num_input_tokens_seen": 126981590,
      "step": 5907,
      "time_per_iteration": 3.551069974899292
    },
    {
      "auxiliary_loss_clip": 0.0117065,
      "auxiliary_loss_mlp": 0.01029695,
      "balance_loss_clip": 1.04874563,
      "balance_loss_mlp": 1.02185667,
      "epoch": 0.7103949978957494,
      "flos": 15303907509120.0,
      "grad_norm": 3.2745658359039957,
      "language_loss": 0.79678792,
      "learning_rate": 8.170055617742364e-07,
      "loss": 0.81879133,
      "num_input_tokens_seen": 126998870,
      "step": 5908,
      "time_per_iteration": 3.4955248832702637
    },
    {
      "auxiliary_loss_clip": 0.01170971,
      "auxiliary_loss_mlp": 0.0103033,
      "balance_loss_clip": 1.04821622,
      "balance_loss_mlp": 1.02248645,
      "epoch": 0.7105152407863885,
      "flos": 22638805401600.0,
      "grad_norm": 1.879523840299737,
      "language_loss": 0.70948839,
      "learning_rate": 8.163775611980252e-07,
      "loss": 0.73150134,
      "num_input_tokens_seen": 127017980,
      "step": 5909,
      "time_per_iteration": 2.646961212158203
    },
    {
      "auxiliary_loss_clip": 0.01172101,
      "auxiliary_loss_mlp": 0.01026055,
      "balance_loss_clip": 1.0484755,
      "balance_loss_mlp": 1.0181694,
      "epoch": 0.7106354836770276,
      "flos": 17238594879360.0,
      "grad_norm": 2.595770406714233,
      "language_loss": 0.79031175,
      "learning_rate": 8.157497401763982e-07,
      "loss": 0.81229329,
      "num_input_tokens_seen": 127035645,
      "step": 5910,
      "time_per_iteration": 2.5398213863372803
    },
    {
      "auxiliary_loss_clip": 0.01169907,
      "auxiliary_loss_mlp": 0.01023043,
      "balance_loss_clip": 1.04772854,
      "balance_loss_mlp": 1.01547933,
      "epoch": 0.7107557265676667,
      "flos": 20193647898240.0,
      "grad_norm": 1.864726110001805,
      "language_loss": 0.77651042,
      "learning_rate": 8.151220988045935e-07,
      "loss": 0.79843998,
      "num_input_tokens_seen": 127054900,
      "step": 5911,
      "time_per_iteration": 2.6410186290740967
    },
    {
      "auxiliary_loss_clip": 0.01170292,
      "auxiliary_loss_mlp": 0.01021473,
      "balance_loss_clip": 1.04758883,
      "balance_loss_mlp": 1.01426649,
      "epoch": 0.7108759694583058,
      "flos": 21507080613120.0,
      "grad_norm": 2.025167548357377,
      "language_loss": 0.82690716,
      "learning_rate": 8.144946371778234e-07,
      "loss": 0.84882474,
      "num_input_tokens_seen": 127075010,
      "step": 5912,
      "time_per_iteration": 2.6424598693847656
    },
    {
      "auxiliary_loss_clip": 0.01172749,
      "auxiliary_loss_mlp": 0.01028357,
      "balance_loss_clip": 1.04855657,
      "balance_loss_mlp": 1.02027202,
      "epoch": 0.7109962123489448,
      "flos": 24061909317120.0,
      "grad_norm": 1.8398506869777602,
      "language_loss": 0.78107423,
      "learning_rate": 8.138673553912751e-07,
      "loss": 0.80308533,
      "num_input_tokens_seen": 127095570,
      "step": 5913,
      "time_per_iteration": 2.596544027328491
    },
    {
      "auxiliary_loss_clip": 0.01175819,
      "auxiliary_loss_mlp": 0.01030679,
      "balance_loss_clip": 1.05123806,
      "balance_loss_mlp": 1.02265656,
      "epoch": 0.711116455239584,
      "flos": 30480474326400.0,
      "grad_norm": 2.861114425836154,
      "language_loss": 0.574893,
      "learning_rate": 8.132402535401059e-07,
      "loss": 0.59695798,
      "num_input_tokens_seen": 127116825,
      "step": 5914,
      "time_per_iteration": 2.7147371768951416
    },
    {
      "auxiliary_loss_clip": 0.01169868,
      "auxiliary_loss_mlp": 0.01027724,
      "balance_loss_clip": 1.04799676,
      "balance_loss_mlp": 1.02007079,
      "epoch": 0.711236698130223,
      "flos": 25045610158080.0,
      "grad_norm": 19.19619335007377,
      "language_loss": 0.74634004,
      "learning_rate": 8.126133317194465e-07,
      "loss": 0.76831597,
      "num_input_tokens_seen": 127137015,
      "step": 5915,
      "time_per_iteration": 2.611893892288208
    },
    {
      "auxiliary_loss_clip": 0.01176469,
      "auxiliary_loss_mlp": 0.0102607,
      "balance_loss_clip": 1.04975808,
      "balance_loss_mlp": 1.01764154,
      "epoch": 0.7113569410208621,
      "flos": 24206701040640.0,
      "grad_norm": 1.848730213810065,
      "language_loss": 0.74162853,
      "learning_rate": 8.11986590024401e-07,
      "loss": 0.76365393,
      "num_input_tokens_seen": 127156755,
      "step": 5916,
      "time_per_iteration": 2.628821849822998
    },
    {
      "auxiliary_loss_clip": 0.01174795,
      "auxiliary_loss_mlp": 0.01034772,
      "balance_loss_clip": 1.04973078,
      "balance_loss_mlp": 1.02590275,
      "epoch": 0.7114771839115013,
      "flos": 35439306526080.0,
      "grad_norm": 2.22932468621017,
      "language_loss": 0.68989491,
      "learning_rate": 8.113600285500442e-07,
      "loss": 0.71199059,
      "num_input_tokens_seen": 127176965,
      "step": 5917,
      "time_per_iteration": 2.787198305130005
    },
    {
      "auxiliary_loss_clip": 0.01171683,
      "auxiliary_loss_mlp": 0.01023598,
      "balance_loss_clip": 1.04839206,
      "balance_loss_mlp": 1.01582527,
      "epoch": 0.7115974268021403,
      "flos": 21099458096640.0,
      "grad_norm": 1.8048229237972102,
      "language_loss": 0.74418747,
      "learning_rate": 8.107336473914268e-07,
      "loss": 0.76614028,
      "num_input_tokens_seen": 127195595,
      "step": 5918,
      "time_per_iteration": 2.623452663421631
    },
    {
      "auxiliary_loss_clip": 0.01071237,
      "auxiliary_loss_mlp": 0.01000947,
      "balance_loss_clip": 1.0127424,
      "balance_loss_mlp": 0.99986821,
      "epoch": 0.7117176696927794,
      "flos": 56752866616320.0,
      "grad_norm": 0.7681265249389412,
      "language_loss": 0.55680811,
      "learning_rate": 8.101074466435694e-07,
      "loss": 0.57752997,
      "num_input_tokens_seen": 127255070,
      "step": 5919,
      "time_per_iteration": 3.0741119384765625
    },
    {
      "auxiliary_loss_clip": 0.01172833,
      "auxiliary_loss_mlp": 0.01024514,
      "balance_loss_clip": 1.05143631,
      "balance_loss_mlp": 1.01641357,
      "epoch": 0.7118379125834186,
      "flos": 15925269905280.0,
      "grad_norm": 2.0872272307613264,
      "language_loss": 0.6822536,
      "learning_rate": 8.094814264014662e-07,
      "loss": 0.70422709,
      "num_input_tokens_seen": 127273825,
      "step": 5920,
      "time_per_iteration": 2.659944772720337
    },
    {
      "auxiliary_loss_clip": 0.01175015,
      "auxiliary_loss_mlp": 0.01022218,
      "balance_loss_clip": 1.04982948,
      "balance_loss_mlp": 1.01278222,
      "epoch": 0.7119581554740576,
      "flos": 20193360589440.0,
      "grad_norm": 2.201371570854312,
      "language_loss": 0.81415123,
      "learning_rate": 8.088555867600844e-07,
      "loss": 0.83612359,
      "num_input_tokens_seen": 127289990,
      "step": 5921,
      "time_per_iteration": 2.5800724029541016
    },
    {
      "auxiliary_loss_clip": 0.01170476,
      "auxiliary_loss_mlp": 0.01025654,
      "balance_loss_clip": 1.05167699,
      "balance_loss_mlp": 1.01835847,
      "epoch": 0.7120783983646967,
      "flos": 34715383822080.0,
      "grad_norm": 1.9599769946170598,
      "language_loss": 0.60348332,
      "learning_rate": 8.08229927814362e-07,
      "loss": 0.62544459,
      "num_input_tokens_seen": 127312880,
      "step": 5922,
      "time_per_iteration": 3.7159812450408936
    },
    {
      "auxiliary_loss_clip": 0.01168934,
      "auxiliary_loss_mlp": 0.01026383,
      "balance_loss_clip": 1.0485543,
      "balance_loss_mlp": 1.01904011,
      "epoch": 0.7121986412553358,
      "flos": 26359114700160.0,
      "grad_norm": 1.735583402734794,
      "language_loss": 0.64783919,
      "learning_rate": 8.076044496592134e-07,
      "loss": 0.66979241,
      "num_input_tokens_seen": 127334730,
      "step": 5923,
      "time_per_iteration": 2.728550434112549
    },
    {
      "auxiliary_loss_clip": 0.0117204,
      "auxiliary_loss_mlp": 0.01025933,
      "balance_loss_clip": 1.04846513,
      "balance_loss_mlp": 1.01796377,
      "epoch": 0.7123188841459749,
      "flos": 11145344371200.0,
      "grad_norm": 2.3264299289053683,
      "language_loss": 0.77731228,
      "learning_rate": 8.069791523895204e-07,
      "loss": 0.79929209,
      "num_input_tokens_seen": 127351180,
      "step": 5924,
      "time_per_iteration": 2.5986838340759277
    },
    {
      "auxiliary_loss_clip": 0.01173135,
      "auxiliary_loss_mlp": 0.01024541,
      "balance_loss_clip": 1.04999185,
      "balance_loss_mlp": 1.01697695,
      "epoch": 0.7124391270366139,
      "flos": 20811670329600.0,
      "grad_norm": 2.0589889457828106,
      "language_loss": 0.77632314,
      "learning_rate": 8.063540361001422e-07,
      "loss": 0.79829991,
      "num_input_tokens_seen": 127369750,
      "step": 5925,
      "time_per_iteration": 2.6586365699768066
    },
    {
      "auxiliary_loss_clip": 0.01170945,
      "auxiliary_loss_mlp": 0.010275,
      "balance_loss_clip": 1.04862332,
      "balance_loss_mlp": 1.0195545,
      "epoch": 0.7125593699272531,
      "flos": 17603734584960.0,
      "grad_norm": 2.490614523943032,
      "language_loss": 0.79540467,
      "learning_rate": 8.057291008859069e-07,
      "loss": 0.81738913,
      "num_input_tokens_seen": 127387910,
      "step": 5926,
      "time_per_iteration": 2.6417341232299805
    },
    {
      "auxiliary_loss_clip": 0.01168202,
      "auxiliary_loss_mlp": 0.01024105,
      "balance_loss_clip": 1.04838073,
      "balance_loss_mlp": 1.01675022,
      "epoch": 0.7126796128178922,
      "flos": 28654057526400.0,
      "grad_norm": 2.1399143807667085,
      "language_loss": 0.68450421,
      "learning_rate": 8.051043468416187e-07,
      "loss": 0.70642728,
      "num_input_tokens_seen": 127409160,
      "step": 5927,
      "time_per_iteration": 2.6918978691101074
    },
    {
      "auxiliary_loss_clip": 0.01169687,
      "auxiliary_loss_mlp": 0.01021199,
      "balance_loss_clip": 1.04921448,
      "balance_loss_mlp": 1.01418662,
      "epoch": 0.7127998557085312,
      "flos": 16034438315520.0,
      "grad_norm": 6.823256360671355,
      "language_loss": 0.82482302,
      "learning_rate": 8.044797740620506e-07,
      "loss": 0.8467319,
      "num_input_tokens_seen": 127427765,
      "step": 5928,
      "time_per_iteration": 2.625623941421509
    },
    {
      "auxiliary_loss_clip": 0.01173949,
      "auxiliary_loss_mlp": 0.01029846,
      "balance_loss_clip": 1.04915953,
      "balance_loss_mlp": 1.02254391,
      "epoch": 0.7129200985991703,
      "flos": 23403271582080.0,
      "grad_norm": 2.002003338866451,
      "language_loss": 0.78797853,
      "learning_rate": 8.038553826419494e-07,
      "loss": 0.81001645,
      "num_input_tokens_seen": 127446475,
      "step": 5929,
      "time_per_iteration": 2.6371071338653564
    },
    {
      "auxiliary_loss_clip": 0.0116745,
      "auxiliary_loss_mlp": 0.01021327,
      "balance_loss_clip": 1.04570055,
      "balance_loss_mlp": 1.01428449,
      "epoch": 0.7130403414898094,
      "flos": 21397445326080.0,
      "grad_norm": 1.8355749692586063,
      "language_loss": 0.80892622,
      "learning_rate": 8.032311726760364e-07,
      "loss": 0.830814,
      "num_input_tokens_seen": 127467695,
      "step": 5930,
      "time_per_iteration": 2.6222894191741943
    },
    {
      "auxiliary_loss_clip": 0.01173386,
      "auxiliary_loss_mlp": 0.01030852,
      "balance_loss_clip": 1.04910183,
      "balance_loss_mlp": 1.02261758,
      "epoch": 0.7131605843804485,
      "flos": 74739045306240.0,
      "grad_norm": 1.907260562992836,
      "language_loss": 0.68676084,
      "learning_rate": 8.026071442590022e-07,
      "loss": 0.70880318,
      "num_input_tokens_seen": 127494590,
      "step": 5931,
      "time_per_iteration": 3.9600493907928467
    },
    {
      "auxiliary_loss_clip": 0.0117254,
      "auxiliary_loss_mlp": 0.01025186,
      "balance_loss_clip": 1.05075359,
      "balance_loss_mlp": 1.01774693,
      "epoch": 0.7132808272710875,
      "flos": 18368739469440.0,
      "grad_norm": 1.9886783767575882,
      "language_loss": 0.80889636,
      "learning_rate": 8.019832974855134e-07,
      "loss": 0.83087361,
      "num_input_tokens_seen": 127512550,
      "step": 5932,
      "time_per_iteration": 2.6039605140686035
    },
    {
      "auxiliary_loss_clip": 0.01175116,
      "auxiliary_loss_mlp": 0.0102248,
      "balance_loss_clip": 1.05182552,
      "balance_loss_mlp": 1.01486874,
      "epoch": 0.7134010701617267,
      "flos": 23253380127360.0,
      "grad_norm": 3.384388898187076,
      "language_loss": 0.8301757,
      "learning_rate": 8.013596324502052e-07,
      "loss": 0.85215169,
      "num_input_tokens_seen": 127531015,
      "step": 5933,
      "time_per_iteration": 3.5165154933929443
    },
    {
      "auxiliary_loss_clip": 0.01166828,
      "auxiliary_loss_mlp": 0.01021113,
      "balance_loss_clip": 1.04796529,
      "balance_loss_mlp": 1.01373363,
      "epoch": 0.7135213130523658,
      "flos": 23653137565440.0,
      "grad_norm": 1.7801420387359945,
      "language_loss": 0.79188728,
      "learning_rate": 8.007361492476872e-07,
      "loss": 0.81376666,
      "num_input_tokens_seen": 127550340,
      "step": 5934,
      "time_per_iteration": 3.5760257244110107
    },
    {
      "auxiliary_loss_clip": 0.01173885,
      "auxiliary_loss_mlp": 0.01023901,
      "balance_loss_clip": 1.04933906,
      "balance_loss_mlp": 1.01630425,
      "epoch": 0.7136415559430048,
      "flos": 24790644443520.0,
      "grad_norm": 1.8068949091240574,
      "language_loss": 0.78898692,
      "learning_rate": 8.001128479725426e-07,
      "loss": 0.81096482,
      "num_input_tokens_seen": 127572245,
      "step": 5935,
      "time_per_iteration": 2.633873462677002
    },
    {
      "auxiliary_loss_clip": 0.01172846,
      "auxiliary_loss_mlp": 0.01028174,
      "balance_loss_clip": 1.04976749,
      "balance_loss_mlp": 1.02054477,
      "epoch": 0.713761798833644,
      "flos": 18296954138880.0,
      "grad_norm": 2.1276105604402358,
      "language_loss": 0.80852544,
      "learning_rate": 7.994897287193248e-07,
      "loss": 0.83053565,
      "num_input_tokens_seen": 127591625,
      "step": 5936,
      "time_per_iteration": 2.6034467220306396
    },
    {
      "auxiliary_loss_clip": 0.0116993,
      "auxiliary_loss_mlp": 0.01022914,
      "balance_loss_clip": 1.04753506,
      "balance_loss_mlp": 1.01478362,
      "epoch": 0.713882041724283,
      "flos": 15558262692480.0,
      "grad_norm": 3.613467285423278,
      "language_loss": 0.83779889,
      "learning_rate": 7.988667915825605e-07,
      "loss": 0.85972738,
      "num_input_tokens_seen": 127608690,
      "step": 5937,
      "time_per_iteration": 2.5926647186279297
    },
    {
      "auxiliary_loss_clip": 0.0117718,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.05336976,
      "balance_loss_mlp": 1.01790917,
      "epoch": 0.7140022846149221,
      "flos": 24061011477120.0,
      "grad_norm": 2.4397758674516505,
      "language_loss": 0.75872195,
      "learning_rate": 7.982440366567491e-07,
      "loss": 0.78075171,
      "num_input_tokens_seen": 127627180,
      "step": 5938,
      "time_per_iteration": 2.575509548187256
    },
    {
      "auxiliary_loss_clip": 0.01167762,
      "auxiliary_loss_mlp": 0.01023877,
      "balance_loss_clip": 1.04747987,
      "balance_loss_mlp": 1.0168283,
      "epoch": 0.7141225275055613,
      "flos": 27891710248320.0,
      "grad_norm": 1.696737828251887,
      "language_loss": 0.75509518,
      "learning_rate": 7.97621464036361e-07,
      "loss": 0.77701151,
      "num_input_tokens_seen": 127648940,
      "step": 5939,
      "time_per_iteration": 2.630460262298584
    },
    {
      "auxiliary_loss_clip": 0.01172191,
      "auxiliary_loss_mlp": 0.01024155,
      "balance_loss_clip": 1.04759634,
      "balance_loss_mlp": 1.01622772,
      "epoch": 0.7142427703962003,
      "flos": 19682603147520.0,
      "grad_norm": 2.5976257607281705,
      "language_loss": 0.68412471,
      "learning_rate": 7.969990738158417e-07,
      "loss": 0.70608819,
      "num_input_tokens_seen": 127667350,
      "step": 5940,
      "time_per_iteration": 2.6534900665283203
    },
    {
      "auxiliary_loss_clip": 0.01173854,
      "auxiliary_loss_mlp": 0.01027247,
      "balance_loss_clip": 1.04942775,
      "balance_loss_mlp": 1.01945972,
      "epoch": 0.7143630132868394,
      "flos": 21032377447680.0,
      "grad_norm": 2.009249464515882,
      "language_loss": 0.85568452,
      "learning_rate": 7.963768660896062e-07,
      "loss": 0.87769544,
      "num_input_tokens_seen": 127685760,
      "step": 5941,
      "time_per_iteration": 2.6458559036254883
    },
    {
      "auxiliary_loss_clip": 0.01172331,
      "auxiliary_loss_mlp": 0.01025748,
      "balance_loss_clip": 1.05019093,
      "balance_loss_mlp": 1.01779687,
      "epoch": 0.7144832561774785,
      "flos": 24129923719680.0,
      "grad_norm": 1.821872683482049,
      "language_loss": 0.82409179,
      "learning_rate": 7.957548409520432e-07,
      "loss": 0.84607255,
      "num_input_tokens_seen": 127704985,
      "step": 5942,
      "time_per_iteration": 2.6109907627105713
    },
    {
      "auxiliary_loss_clip": 0.01171398,
      "auxiliary_loss_mlp": 0.01023385,
      "balance_loss_clip": 1.04947996,
      "balance_loss_mlp": 1.01587439,
      "epoch": 0.7146034990681176,
      "flos": 16325817442560.0,
      "grad_norm": 2.2922596879684916,
      "language_loss": 0.84173131,
      "learning_rate": 7.951329984975135e-07,
      "loss": 0.86367905,
      "num_input_tokens_seen": 127721925,
      "step": 5943,
      "time_per_iteration": 2.6155576705932617
    },
    {
      "auxiliary_loss_clip": 0.01071417,
      "auxiliary_loss_mlp": 0.01001043,
      "balance_loss_clip": 1.01357365,
      "balance_loss_mlp": 0.99988699,
      "epoch": 0.7147237419587567,
      "flos": 69627164232960.0,
      "grad_norm": 0.7131840510187797,
      "language_loss": 0.54336637,
      "learning_rate": 7.94511338820349e-07,
      "loss": 0.56409097,
      "num_input_tokens_seen": 127784230,
      "step": 5944,
      "time_per_iteration": 3.1808481216430664
    },
    {
      "auxiliary_loss_clip": 0.01171859,
      "auxiliary_loss_mlp": 0.0102476,
      "balance_loss_clip": 1.04792166,
      "balance_loss_mlp": 1.01704764,
      "epoch": 0.7148439848493958,
      "flos": 22266806198400.0,
      "grad_norm": 2.41177533830607,
      "language_loss": 0.78296363,
      "learning_rate": 7.938898620148575e-07,
      "loss": 0.80492985,
      "num_input_tokens_seen": 127801990,
      "step": 5945,
      "time_per_iteration": 2.597933292388916
    },
    {
      "auxiliary_loss_clip": 0.01126363,
      "auxiliary_loss_mlp": 0.01026386,
      "balance_loss_clip": 1.04127896,
      "balance_loss_mlp": 1.01819634,
      "epoch": 0.7149642277400349,
      "flos": 17931383470080.0,
      "grad_norm": 2.377646929000803,
      "language_loss": 0.71149695,
      "learning_rate": 7.932685681753135e-07,
      "loss": 0.73302442,
      "num_input_tokens_seen": 127819270,
      "step": 5946,
      "time_per_iteration": 2.727630615234375
    },
    {
      "auxiliary_loss_clip": 0.01068207,
      "auxiliary_loss_mlp": 0.01019687,
      "balance_loss_clip": 1.03589046,
      "balance_loss_mlp": 1.01210546,
      "epoch": 0.7150844706306739,
      "flos": 31681937370240.0,
      "grad_norm": 2.004416694553908,
      "language_loss": 0.62773407,
      "learning_rate": 7.92647457395969e-07,
      "loss": 0.64861304,
      "num_input_tokens_seen": 127841095,
      "step": 5947,
      "time_per_iteration": 3.1319291591644287
    },
    {
      "auxiliary_loss_clip": 0.01173257,
      "auxiliary_loss_mlp": 0.01030339,
      "balance_loss_clip": 1.05028248,
      "balance_loss_mlp": 1.02220857,
      "epoch": 0.7152047135213131,
      "flos": 10926217451520.0,
      "grad_norm": 7.666890381106461,
      "language_loss": 0.74004233,
      "learning_rate": 7.920265297710444e-07,
      "loss": 0.76207829,
      "num_input_tokens_seen": 127858485,
      "step": 5948,
      "time_per_iteration": 3.684605836868286
    },
    {
      "auxiliary_loss_clip": 0.01131509,
      "auxiliary_loss_mlp": 0.01024718,
      "balance_loss_clip": 1.04457664,
      "balance_loss_mlp": 1.0171957,
      "epoch": 0.7153249564119522,
      "flos": 20995640812800.0,
      "grad_norm": 2.909709441622768,
      "language_loss": 0.73373276,
      "learning_rate": 7.914057853947363e-07,
      "loss": 0.75529504,
      "num_input_tokens_seen": 127877665,
      "step": 5949,
      "time_per_iteration": 2.693267822265625
    },
    {
      "auxiliary_loss_clip": 0.01109701,
      "auxiliary_loss_mlp": 0.00763609,
      "balance_loss_clip": 1.04565024,
      "balance_loss_mlp": 1.000319,
      "epoch": 0.7154451993025912,
      "flos": 24243114453120.0,
      "grad_norm": 1.7375597621614807,
      "language_loss": 0.62750876,
      "learning_rate": 7.907852243612089e-07,
      "loss": 0.6462419,
      "num_input_tokens_seen": 127898070,
      "step": 5950,
      "time_per_iteration": 2.8176615238189697
    },
    {
      "auxiliary_loss_clip": 0.01097206,
      "auxiliary_loss_mlp": 0.01023498,
      "balance_loss_clip": 1.04341078,
      "balance_loss_mlp": 1.01579094,
      "epoch": 0.7155654421932304,
      "flos": 23330947547520.0,
      "grad_norm": 2.129237969510858,
      "language_loss": 0.72717488,
      "learning_rate": 7.901648467646009e-07,
      "loss": 0.74838191,
      "num_input_tokens_seen": 127917010,
      "step": 5951,
      "time_per_iteration": 2.8082621097564697
    },
    {
      "auxiliary_loss_clip": 0.01057288,
      "auxiliary_loss_mlp": 0.0102484,
      "balance_loss_clip": 1.03717375,
      "balance_loss_mlp": 1.01702619,
      "epoch": 0.7156856850838694,
      "flos": 22711883621760.0,
      "grad_norm": 1.7828070633219828,
      "language_loss": 0.72847384,
      "learning_rate": 7.895446526990244e-07,
      "loss": 0.74929512,
      "num_input_tokens_seen": 127937025,
      "step": 5952,
      "time_per_iteration": 2.997647762298584
    },
    {
      "auxiliary_loss_clip": 0.01161476,
      "auxiliary_loss_mlp": 0.00762788,
      "balance_loss_clip": 1.04765332,
      "balance_loss_mlp": 1.00028765,
      "epoch": 0.7158059279745085,
      "flos": 19865424395520.0,
      "grad_norm": 1.533838769298876,
      "language_loss": 0.75789809,
      "learning_rate": 7.889246422585609e-07,
      "loss": 0.77714074,
      "num_input_tokens_seen": 127956410,
      "step": 5953,
      "time_per_iteration": 3.4434022903442383
    },
    {
      "auxiliary_loss_clip": 0.01150849,
      "auxiliary_loss_mlp": 0.01027124,
      "balance_loss_clip": 1.04749227,
      "balance_loss_mlp": 1.0191133,
      "epoch": 0.7159261708651476,
      "flos": 24134772055680.0,
      "grad_norm": 1.702755919960403,
      "language_loss": 0.73744446,
      "learning_rate": 7.883048155372675e-07,
      "loss": 0.75922418,
      "num_input_tokens_seen": 127974925,
      "step": 5954,
      "time_per_iteration": 2.7331957817077637
    },
    {
      "auxiliary_loss_clip": 0.01135865,
      "auxiliary_loss_mlp": 0.01023654,
      "balance_loss_clip": 1.044837,
      "balance_loss_mlp": 1.01580429,
      "epoch": 0.7160464137557867,
      "flos": 16983198201600.0,
      "grad_norm": 2.373094626403993,
      "language_loss": 0.71294338,
      "learning_rate": 7.876851726291698e-07,
      "loss": 0.73453856,
      "num_input_tokens_seen": 127993225,
      "step": 5955,
      "time_per_iteration": 2.757322311401367
    },
    {
      "auxiliary_loss_clip": 0.01119911,
      "auxiliary_loss_mlp": 0.01025663,
      "balance_loss_clip": 1.0454669,
      "balance_loss_mlp": 1.01824808,
      "epoch": 0.7161666566464258,
      "flos": 25228251838080.0,
      "grad_norm": 2.0086765949098293,
      "language_loss": 0.78298831,
      "learning_rate": 7.870657136282666e-07,
      "loss": 0.80444402,
      "num_input_tokens_seen": 128012085,
      "step": 5956,
      "time_per_iteration": 2.770124912261963
    },
    {
      "auxiliary_loss_clip": 0.01117109,
      "auxiliary_loss_mlp": 0.01026124,
      "balance_loss_clip": 1.04190993,
      "balance_loss_mlp": 1.01795769,
      "epoch": 0.7162868995370649,
      "flos": 26468390851200.0,
      "grad_norm": 1.8321019496674265,
      "language_loss": 0.82138532,
      "learning_rate": 7.86446438628531e-07,
      "loss": 0.84281766,
      "num_input_tokens_seen": 128033155,
      "step": 5957,
      "time_per_iteration": 3.668361186981201
    },
    {
      "auxiliary_loss_clip": 0.01016597,
      "auxiliary_loss_mlp": 0.01003668,
      "balance_loss_clip": 1.01290274,
      "balance_loss_mlp": 1.00236249,
      "epoch": 0.716407142427704,
      "flos": 69998912040960.0,
      "grad_norm": 0.7742963218509353,
      "language_loss": 0.56848782,
      "learning_rate": 7.858273477239059e-07,
      "loss": 0.58869052,
      "num_input_tokens_seen": 128101575,
      "step": 5958,
      "time_per_iteration": 3.6758947372436523
    },
    {
      "auxiliary_loss_clip": 0.01172093,
      "auxiliary_loss_mlp": 0.01028444,
      "balance_loss_clip": 1.0493989,
      "balance_loss_mlp": 1.02053428,
      "epoch": 0.716527385318343,
      "flos": 20740459616640.0,
      "grad_norm": 2.0112269686890576,
      "language_loss": 0.71321833,
      "learning_rate": 7.852084410083067e-07,
      "loss": 0.73522377,
      "num_input_tokens_seen": 128120395,
      "step": 5959,
      "time_per_iteration": 4.5778374671936035
    },
    {
      "auxiliary_loss_clip": 0.01112548,
      "auxiliary_loss_mlp": 0.01025142,
      "balance_loss_clip": 1.04116654,
      "balance_loss_mlp": 1.01790893,
      "epoch": 0.7166476282089821,
      "flos": 25371966153600.0,
      "grad_norm": 1.5944234510454136,
      "language_loss": 0.63444376,
      "learning_rate": 7.84589718575621e-07,
      "loss": 0.65582067,
      "num_input_tokens_seen": 128140840,
      "step": 5960,
      "time_per_iteration": 3.709197521209717
    },
    {
      "auxiliary_loss_clip": 0.01138555,
      "auxiliary_loss_mlp": 0.01021625,
      "balance_loss_clip": 1.04378605,
      "balance_loss_mlp": 1.01381683,
      "epoch": 0.7167678710996213,
      "flos": 24133730561280.0,
      "grad_norm": 2.5091354163372226,
      "language_loss": 0.68942779,
      "learning_rate": 7.83971180519708e-07,
      "loss": 0.71102959,
      "num_input_tokens_seen": 128159695,
      "step": 5961,
      "time_per_iteration": 2.626422166824341
    },
    {
      "auxiliary_loss_clip": 0.01110196,
      "auxiliary_loss_mlp": 0.01024449,
      "balance_loss_clip": 1.04338861,
      "balance_loss_mlp": 1.01577115,
      "epoch": 0.7168881139902603,
      "flos": 30226586019840.0,
      "grad_norm": 2.1756696790601766,
      "language_loss": 0.75450408,
      "learning_rate": 7.833528269344008e-07,
      "loss": 0.77585053,
      "num_input_tokens_seen": 128179600,
      "step": 5962,
      "time_per_iteration": 2.838521718978882
    },
    {
      "auxiliary_loss_clip": 0.01144386,
      "auxiliary_loss_mlp": 0.01026616,
      "balance_loss_clip": 1.04657841,
      "balance_loss_mlp": 1.01848555,
      "epoch": 0.7170083568808994,
      "flos": 14606414236800.0,
      "grad_norm": 6.561602106222439,
      "language_loss": 0.77407551,
      "learning_rate": 7.827346579135023e-07,
      "loss": 0.79578543,
      "num_input_tokens_seen": 128196940,
      "step": 5963,
      "time_per_iteration": 2.6397902965545654
    },
    {
      "auxiliary_loss_clip": 0.01095989,
      "auxiliary_loss_mlp": 0.01024303,
      "balance_loss_clip": 1.04466045,
      "balance_loss_mlp": 1.01660252,
      "epoch": 0.7171285997715385,
      "flos": 23331091201920.0,
      "grad_norm": 1.8954830914283107,
      "language_loss": 0.83597279,
      "learning_rate": 7.821166735507885e-07,
      "loss": 0.85717571,
      "num_input_tokens_seen": 128215970,
      "step": 5964,
      "time_per_iteration": 2.8152072429656982
    },
    {
      "auxiliary_loss_clip": 0.01156759,
      "auxiliary_loss_mlp": 0.01023316,
      "balance_loss_clip": 1.05009305,
      "balance_loss_mlp": 1.01584125,
      "epoch": 0.7172488426621776,
      "flos": 16543543731840.0,
      "grad_norm": 2.1775130682943535,
      "language_loss": 0.6875174,
      "learning_rate": 7.81498873940007e-07,
      "loss": 0.70931804,
      "num_input_tokens_seen": 128233185,
      "step": 5965,
      "time_per_iteration": 2.5767855644226074
    },
    {
      "auxiliary_loss_clip": 0.01087211,
      "auxiliary_loss_mlp": 0.01025048,
      "balance_loss_clip": 1.03604674,
      "balance_loss_mlp": 1.01657176,
      "epoch": 0.7173690855528166,
      "flos": 26541612725760.0,
      "grad_norm": 2.1805026986805975,
      "language_loss": 0.77634728,
      "learning_rate": 7.808812591748768e-07,
      "loss": 0.79746985,
      "num_input_tokens_seen": 128253565,
      "step": 5966,
      "time_per_iteration": 2.8375282287597656
    },
    {
      "auxiliary_loss_clip": 0.0113388,
      "auxiliary_loss_mlp": 0.01025796,
      "balance_loss_clip": 1.04370797,
      "balance_loss_mlp": 1.01768398,
      "epoch": 0.7174893284434558,
      "flos": 22784099915520.0,
      "grad_norm": 2.999702405439084,
      "language_loss": 0.65173984,
      "learning_rate": 7.802638293490915e-07,
      "loss": 0.67333657,
      "num_input_tokens_seen": 128273210,
      "step": 5967,
      "time_per_iteration": 2.700505018234253
    },
    {
      "auxiliary_loss_clip": 0.0115277,
      "auxiliary_loss_mlp": 0.01021823,
      "balance_loss_clip": 1.0467962,
      "balance_loss_mlp": 1.01390123,
      "epoch": 0.7176095713340949,
      "flos": 23293564467840.0,
      "grad_norm": 3.421970983257029,
      "language_loss": 0.77056319,
      "learning_rate": 7.796465845563123e-07,
      "loss": 0.79230917,
      "num_input_tokens_seen": 128292085,
      "step": 5968,
      "time_per_iteration": 2.703040599822998
    },
    {
      "auxiliary_loss_clip": 0.01059222,
      "auxiliary_loss_mlp": 0.01025379,
      "balance_loss_clip": 1.03350687,
      "balance_loss_mlp": 1.01773131,
      "epoch": 0.7177298142247339,
      "flos": 25591631777280.0,
      "grad_norm": 2.011478501950101,
      "language_loss": 0.79772955,
      "learning_rate": 7.790295248901766e-07,
      "loss": 0.81857556,
      "num_input_tokens_seen": 128313215,
      "step": 5969,
      "time_per_iteration": 3.2134077548980713
    },
    {
      "auxiliary_loss_clip": 0.01091049,
      "auxiliary_loss_mlp": 0.01021333,
      "balance_loss_clip": 1.04242396,
      "balance_loss_mlp": 1.01298273,
      "epoch": 0.7178500571153731,
      "flos": 31652778504960.0,
      "grad_norm": 2.181480111498677,
      "language_loss": 0.61977917,
      "learning_rate": 7.784126504442902e-07,
      "loss": 0.640903,
      "num_input_tokens_seen": 128336445,
      "step": 5970,
      "time_per_iteration": 3.5671589374542236
    },
    {
      "auxiliary_loss_clip": 0.01131013,
      "auxiliary_loss_mlp": 0.010258,
      "balance_loss_clip": 1.04631567,
      "balance_loss_mlp": 1.01830184,
      "epoch": 0.7179703000060121,
      "flos": 19427242383360.0,
      "grad_norm": 1.3293195065732526,
      "language_loss": 0.67752528,
      "learning_rate": 7.777959613122351e-07,
      "loss": 0.6990934,
      "num_input_tokens_seen": 128356270,
      "step": 5971,
      "time_per_iteration": 2.7664542198181152
    },
    {
      "auxiliary_loss_clip": 0.01100636,
      "auxiliary_loss_mlp": 0.01022368,
      "balance_loss_clip": 1.0403434,
      "balance_loss_mlp": 1.01484585,
      "epoch": 0.7180905428966512,
      "flos": 28839249072000.0,
      "grad_norm": 2.0337653794422637,
      "language_loss": 0.78084993,
      "learning_rate": 7.771794575875604e-07,
      "loss": 0.80207998,
      "num_input_tokens_seen": 128378140,
      "step": 5972,
      "time_per_iteration": 2.7913780212402344
    },
    {
      "auxiliary_loss_clip": 0.01148466,
      "auxiliary_loss_mlp": 0.00763072,
      "balance_loss_clip": 1.04484081,
      "balance_loss_mlp": 1.00043845,
      "epoch": 0.7182107857872904,
      "flos": 20047563285120.0,
      "grad_norm": 3.3072233531605963,
      "language_loss": 0.77950716,
      "learning_rate": 7.765631393637888e-07,
      "loss": 0.79862261,
      "num_input_tokens_seen": 128396335,
      "step": 5973,
      "time_per_iteration": 2.689817190170288
    },
    {
      "auxiliary_loss_clip": 0.0116189,
      "auxiliary_loss_mlp": 0.01030122,
      "balance_loss_clip": 1.04754543,
      "balance_loss_mlp": 1.02189112,
      "epoch": 0.7183310286779294,
      "flos": 22747686503040.0,
      "grad_norm": 3.8976948588732006,
      "language_loss": 0.47901413,
      "learning_rate": 7.75947006734417e-07,
      "loss": 0.50093424,
      "num_input_tokens_seen": 128414115,
      "step": 5974,
      "time_per_iteration": 4.057405471801758
    },
    {
      "auxiliary_loss_clip": 0.01110045,
      "auxiliary_loss_mlp": 0.01024506,
      "balance_loss_clip": 1.04026151,
      "balance_loss_mlp": 1.01657891,
      "epoch": 0.7184512715685685,
      "flos": 17158262112000.0,
      "grad_norm": 3.6369790564999622,
      "language_loss": 0.831626,
      "learning_rate": 7.753310597929101e-07,
      "loss": 0.85297149,
      "num_input_tokens_seen": 128430755,
      "step": 5975,
      "time_per_iteration": 2.7356767654418945
    },
    {
      "auxiliary_loss_clip": 0.01044586,
      "auxiliary_loss_mlp": 0.01002424,
      "balance_loss_clip": 1.01358771,
      "balance_loss_mlp": 1.001369,
      "epoch": 0.7185715144592076,
      "flos": 65509611448320.0,
      "grad_norm": 0.7576371763043058,
      "language_loss": 0.55072778,
      "learning_rate": 7.747152986327095e-07,
      "loss": 0.57119787,
      "num_input_tokens_seen": 128491300,
      "step": 5976,
      "time_per_iteration": 3.1749649047851562
    },
    {
      "auxiliary_loss_clip": 0.01150537,
      "auxiliary_loss_mlp": 0.01025093,
      "balance_loss_clip": 1.04667974,
      "balance_loss_mlp": 1.01787448,
      "epoch": 0.7186917573498467,
      "flos": 16180522928640.0,
      "grad_norm": 1.884035464367145,
      "language_loss": 0.6836561,
      "learning_rate": 7.740997233472228e-07,
      "loss": 0.70541233,
      "num_input_tokens_seen": 128508920,
      "step": 5977,
      "time_per_iteration": 2.581763505935669
    },
    {
      "auxiliary_loss_clip": 0.01126837,
      "auxiliary_loss_mlp": 0.01022094,
      "balance_loss_clip": 1.04346943,
      "balance_loss_mlp": 1.01472068,
      "epoch": 0.7188120002404857,
      "flos": 29242274647680.0,
      "grad_norm": 2.3851737102786554,
      "language_loss": 0.70379686,
      "learning_rate": 7.734843340298329e-07,
      "loss": 0.72528613,
      "num_input_tokens_seen": 128528745,
      "step": 5978,
      "time_per_iteration": 2.731870174407959
    },
    {
      "auxiliary_loss_clip": 0.01136173,
      "auxiliary_loss_mlp": 0.0102523,
      "balance_loss_clip": 1.04610729,
      "balance_loss_mlp": 1.01658154,
      "epoch": 0.7189322431311249,
      "flos": 33401161008000.0,
      "grad_norm": 2.6943999635376668,
      "language_loss": 0.75395614,
      "learning_rate": 7.72869130773895e-07,
      "loss": 0.77557015,
      "num_input_tokens_seen": 128549345,
      "step": 5979,
      "time_per_iteration": 2.8746120929718018
    },
    {
      "auxiliary_loss_clip": 0.01011255,
      "auxiliary_loss_mlp": 0.01004545,
      "balance_loss_clip": 1.01513565,
      "balance_loss_mlp": 1.00346053,
      "epoch": 0.719052486021764,
      "flos": 61351263792000.0,
      "grad_norm": 0.7971971494227273,
      "language_loss": 0.59358895,
      "learning_rate": 7.722541136727343e-07,
      "loss": 0.613747,
      "num_input_tokens_seen": 128605360,
      "step": 5980,
      "time_per_iteration": 3.218266248703003
    },
    {
      "auxiliary_loss_clip": 0.01160515,
      "auxiliary_loss_mlp": 0.01027667,
      "balance_loss_clip": 1.04779589,
      "balance_loss_mlp": 1.02023435,
      "epoch": 0.719172728912403,
      "flos": 15596795007360.0,
      "grad_norm": 2.407796558527547,
      "language_loss": 0.80913758,
      "learning_rate": 7.716392828196483e-07,
      "loss": 0.8310194,
      "num_input_tokens_seen": 128623160,
      "step": 5981,
      "time_per_iteration": 2.736133337020874
    },
    {
      "auxiliary_loss_clip": 0.01143092,
      "auxiliary_loss_mlp": 0.01027846,
      "balance_loss_clip": 1.0481627,
      "balance_loss_mlp": 1.01955485,
      "epoch": 0.7192929718030422,
      "flos": 15553162961280.0,
      "grad_norm": 2.3204255854521754,
      "language_loss": 0.77522326,
      "learning_rate": 7.710246383079064e-07,
      "loss": 0.79693258,
      "num_input_tokens_seen": 128638545,
      "step": 5982,
      "time_per_iteration": 2.693380117416382
    },
    {
      "auxiliary_loss_clip": 0.01139033,
      "auxiliary_loss_mlp": 0.01024534,
      "balance_loss_clip": 1.0455842,
      "balance_loss_mlp": 1.0169518,
      "epoch": 0.7194132146936812,
      "flos": 21862487733120.0,
      "grad_norm": 3.5970059443280165,
      "language_loss": 0.92314553,
      "learning_rate": 7.704101802307492e-07,
      "loss": 0.94478118,
      "num_input_tokens_seen": 128650845,
      "step": 5983,
      "time_per_iteration": 3.878540515899658
    },
    {
      "auxiliary_loss_clip": 0.01159724,
      "auxiliary_loss_mlp": 0.01029482,
      "balance_loss_clip": 1.05196297,
      "balance_loss_mlp": 1.02172709,
      "epoch": 0.7195334575843203,
      "flos": 27338900958720.0,
      "grad_norm": 2.7387419448509784,
      "language_loss": 0.87109065,
      "learning_rate": 7.697959086813912e-07,
      "loss": 0.89298266,
      "num_input_tokens_seen": 128667010,
      "step": 5984,
      "time_per_iteration": 3.5429365634918213
    },
    {
      "auxiliary_loss_clip": 0.01099301,
      "auxiliary_loss_mlp": 0.01025091,
      "balance_loss_clip": 1.04189014,
      "balance_loss_mlp": 1.01779497,
      "epoch": 0.7196537004749595,
      "flos": 18770615809920.0,
      "grad_norm": 1.7141360059225101,
      "language_loss": 0.79990715,
      "learning_rate": 7.691818237530145e-07,
      "loss": 0.82115108,
      "num_input_tokens_seen": 128685870,
      "step": 5985,
      "time_per_iteration": 2.760165214538574
    },
    {
      "auxiliary_loss_clip": 0.01151547,
      "auxiliary_loss_mlp": 0.01029238,
      "balance_loss_clip": 1.05237019,
      "balance_loss_mlp": 1.02154303,
      "epoch": 0.7197739433655985,
      "flos": 24531009960960.0,
      "grad_norm": 1.9009192289608274,
      "language_loss": 0.77194309,
      "learning_rate": 7.685679255387774e-07,
      "loss": 0.793751,
      "num_input_tokens_seen": 128704185,
      "step": 5986,
      "time_per_iteration": 3.7403669357299805
    },
    {
      "auxiliary_loss_clip": 0.01124534,
      "auxiliary_loss_mlp": 0.01033704,
      "balance_loss_clip": 1.039819,
      "balance_loss_mlp": 1.02554417,
      "epoch": 0.7198941862562376,
      "flos": 18040587793920.0,
      "grad_norm": 2.0932973209710046,
      "language_loss": 0.76882446,
      "learning_rate": 7.679542141318065e-07,
      "loss": 0.79040682,
      "num_input_tokens_seen": 128721290,
      "step": 5987,
      "time_per_iteration": 2.80513858795166
    },
    {
      "auxiliary_loss_clip": 0.0114736,
      "auxiliary_loss_mlp": 0.00761582,
      "balance_loss_clip": 1.04858112,
      "balance_loss_mlp": 1.00043571,
      "epoch": 0.7200144291468767,
      "flos": 29022393542400.0,
      "grad_norm": 2.007602909933667,
      "language_loss": 0.75943613,
      "learning_rate": 7.673406896252013e-07,
      "loss": 0.77852553,
      "num_input_tokens_seen": 128742665,
      "step": 5988,
      "time_per_iteration": 2.73661732673645
    },
    {
      "auxiliary_loss_clip": 0.01138496,
      "auxiliary_loss_mlp": 0.01025415,
      "balance_loss_clip": 1.04571187,
      "balance_loss_mlp": 1.01665306,
      "epoch": 0.7201346720375158,
      "flos": 25374264624000.0,
      "grad_norm": 2.792306923640512,
      "language_loss": 0.78300923,
      "learning_rate": 7.667273521120347e-07,
      "loss": 0.80464834,
      "num_input_tokens_seen": 128762225,
      "step": 5989,
      "time_per_iteration": 2.7278380393981934
    },
    {
      "auxiliary_loss_clip": 0.01102078,
      "auxiliary_loss_mlp": 0.01025084,
      "balance_loss_clip": 1.04176438,
      "balance_loss_mlp": 1.01734126,
      "epoch": 0.7202549149281549,
      "flos": 14355614499840.0,
      "grad_norm": 1.9365835297903755,
      "language_loss": 0.79779571,
      "learning_rate": 7.661142016853468e-07,
      "loss": 0.81906736,
      "num_input_tokens_seen": 128779585,
      "step": 5990,
      "time_per_iteration": 2.6974313259124756
    },
    {
      "auxiliary_loss_clip": 0.01120575,
      "auxiliary_loss_mlp": 0.01027398,
      "balance_loss_clip": 1.04867721,
      "balance_loss_mlp": 1.01923823,
      "epoch": 0.7203751578187939,
      "flos": 23001682550400.0,
      "grad_norm": 1.7581981606548276,
      "language_loss": 0.74847156,
      "learning_rate": 7.655012384381543e-07,
      "loss": 0.76995134,
      "num_input_tokens_seen": 128799070,
      "step": 5991,
      "time_per_iteration": 2.8810505867004395
    },
    {
      "auxiliary_loss_clip": 0.01160032,
      "auxiliary_loss_mlp": 0.01026032,
      "balance_loss_clip": 1.04754925,
      "balance_loss_mlp": 1.01773524,
      "epoch": 0.7204954007094331,
      "flos": 23692424065920.0,
      "grad_norm": 2.0516392679681004,
      "language_loss": 0.81725454,
      "learning_rate": 7.648884624634415e-07,
      "loss": 0.83911514,
      "num_input_tokens_seen": 128817620,
      "step": 5992,
      "time_per_iteration": 2.7014191150665283
    },
    {
      "auxiliary_loss_clip": 0.01086957,
      "auxiliary_loss_mlp": 0.01031728,
      "balance_loss_clip": 1.04065204,
      "balance_loss_mlp": 1.02363384,
      "epoch": 0.7206156436000721,
      "flos": 16253026531200.0,
      "grad_norm": 4.418132466429256,
      "language_loss": 0.89031488,
      "learning_rate": 7.642758738541683e-07,
      "loss": 0.91150177,
      "num_input_tokens_seen": 128834200,
      "step": 5993,
      "time_per_iteration": 2.757542133331299
    },
    {
      "auxiliary_loss_clip": 0.01046978,
      "auxiliary_loss_mlp": 0.01004364,
      "balance_loss_clip": 1.01509738,
      "balance_loss_mlp": 1.00332665,
      "epoch": 0.7207358864907112,
      "flos": 54377806504320.0,
      "grad_norm": 0.7549191890193174,
      "language_loss": 0.60725635,
      "learning_rate": 7.636634727032621e-07,
      "loss": 0.62776977,
      "num_input_tokens_seen": 128891305,
      "step": 5994,
      "time_per_iteration": 3.1726579666137695
    },
    {
      "auxiliary_loss_clip": 0.0105772,
      "auxiliary_loss_mlp": 0.0102774,
      "balance_loss_clip": 1.03986561,
      "balance_loss_mlp": 1.01856089,
      "epoch": 0.7208561293813504,
      "flos": 19135540033920.0,
      "grad_norm": 2.0051436739052866,
      "language_loss": 0.78878278,
      "learning_rate": 7.630512591036231e-07,
      "loss": 0.80963743,
      "num_input_tokens_seen": 128910615,
      "step": 5995,
      "time_per_iteration": 3.1132829189300537
    },
    {
      "auxiliary_loss_clip": 0.01115315,
      "auxiliary_loss_mlp": 0.0076281,
      "balance_loss_clip": 1.04427624,
      "balance_loss_mlp": 1.00035691,
      "epoch": 0.7209763722719894,
      "flos": 17748526308480.0,
      "grad_norm": 2.2916267417964877,
      "language_loss": 0.64832878,
      "learning_rate": 7.624392331481255e-07,
      "loss": 0.66710997,
      "num_input_tokens_seen": 128928270,
      "step": 5996,
      "time_per_iteration": 3.6295340061187744
    },
    {
      "auxiliary_loss_clip": 0.01023249,
      "auxiliary_loss_mlp": 0.01002884,
      "balance_loss_clip": 1.01186585,
      "balance_loss_mlp": 1.00179315,
      "epoch": 0.7210966151626285,
      "flos": 66819488716800.0,
      "grad_norm": 0.7440846065774371,
      "language_loss": 0.51828456,
      "learning_rate": 7.618273949296115e-07,
      "loss": 0.53854597,
      "num_input_tokens_seen": 128987780,
      "step": 5997,
      "time_per_iteration": 3.297100067138672
    },
    {
      "auxiliary_loss_clip": 0.01152057,
      "auxiliary_loss_mlp": 0.00762841,
      "balance_loss_clip": 1.04519081,
      "balance_loss_mlp": 1.00038397,
      "epoch": 0.7212168580532676,
      "flos": 21141869080320.0,
      "grad_norm": 1.918329793719467,
      "language_loss": 0.68514705,
      "learning_rate": 7.612157445408987e-07,
      "loss": 0.70429605,
      "num_input_tokens_seen": 129005590,
      "step": 5998,
      "time_per_iteration": 2.6519856452941895
    },
    {
      "auxiliary_loss_clip": 0.01132514,
      "auxiliary_loss_mlp": 0.01025292,
      "balance_loss_clip": 1.04777956,
      "balance_loss_mlp": 1.01703072,
      "epoch": 0.7213371009439067,
      "flos": 22345738335360.0,
      "grad_norm": 2.9367893348994922,
      "language_loss": 0.74379128,
      "learning_rate": 7.606042820747716e-07,
      "loss": 0.76536936,
      "num_input_tokens_seen": 129021995,
      "step": 5999,
      "time_per_iteration": 2.7421975135803223
    },
    {
      "auxiliary_loss_clip": 0.01136461,
      "auxiliary_loss_mlp": 0.01026017,
      "balance_loss_clip": 1.04775262,
      "balance_loss_mlp": 1.01839662,
      "epoch": 0.7214573438345457,
      "flos": 18515901490560.0,
      "grad_norm": 1.940113498475789,
      "language_loss": 0.85706913,
      "learning_rate": 7.599930076239889e-07,
      "loss": 0.87869382,
      "num_input_tokens_seen": 129039280,
      "step": 6000,
      "time_per_iteration": 3.6032302379608154
    },
    {
      "auxiliary_loss_clip": 0.01138056,
      "auxiliary_loss_mlp": 0.0102871,
      "balance_loss_clip": 1.04599404,
      "balance_loss_mlp": 1.02028775,
      "epoch": 0.7215775867251849,
      "flos": 35736108606720.0,
      "grad_norm": 2.3564312970048857,
      "language_loss": 0.70743728,
      "learning_rate": 7.593819212812818e-07,
      "loss": 0.72910494,
      "num_input_tokens_seen": 129060860,
      "step": 6001,
      "time_per_iteration": 2.8777294158935547
    },
    {
      "auxiliary_loss_clip": 0.01106666,
      "auxiliary_loss_mlp": 0.00762679,
      "balance_loss_clip": 1.04318368,
      "balance_loss_mlp": 1.0003612,
      "epoch": 0.721697829615824,
      "flos": 20372410909440.0,
      "grad_norm": 2.334330645538725,
      "language_loss": 0.7175163,
      "learning_rate": 7.587710231393508e-07,
      "loss": 0.73620975,
      "num_input_tokens_seen": 129079215,
      "step": 6002,
      "time_per_iteration": 2.832409143447876
    },
    {
      "auxiliary_loss_clip": 0.01135267,
      "auxiliary_loss_mlp": 0.01029272,
      "balance_loss_clip": 1.0483011,
      "balance_loss_mlp": 1.02145815,
      "epoch": 0.721818072506463,
      "flos": 20229809915520.0,
      "grad_norm": 2.2923913996011693,
      "language_loss": 0.837713,
      "learning_rate": 7.581603132908685e-07,
      "loss": 0.85935837,
      "num_input_tokens_seen": 129097185,
      "step": 6003,
      "time_per_iteration": 2.7187094688415527
    },
    {
      "auxiliary_loss_clip": 0.01125183,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.04559386,
      "balance_loss_mlp": 1.02158916,
      "epoch": 0.7219383153971022,
      "flos": 18186887888640.0,
      "grad_norm": 1.9186157608531318,
      "language_loss": 0.78752774,
      "learning_rate": 7.575497918284795e-07,
      "loss": 0.80908096,
      "num_input_tokens_seen": 129114730,
      "step": 6004,
      "time_per_iteration": 2.7251856327056885
    },
    {
      "auxiliary_loss_clip": 0.01114083,
      "auxiliary_loss_mlp": 0.01029729,
      "balance_loss_clip": 1.04314232,
      "balance_loss_mlp": 1.0203892,
      "epoch": 0.7220585582877412,
      "flos": 17342124854400.0,
      "grad_norm": 5.039976024765744,
      "language_loss": 0.7470454,
      "learning_rate": 7.569394588447984e-07,
      "loss": 0.76848352,
      "num_input_tokens_seen": 129131745,
      "step": 6005,
      "time_per_iteration": 2.6944420337677
    },
    {
      "auxiliary_loss_clip": 0.01126298,
      "auxiliary_loss_mlp": 0.01022547,
      "balance_loss_clip": 1.042979,
      "balance_loss_mlp": 1.01481652,
      "epoch": 0.7221788011783803,
      "flos": 16976338704000.0,
      "grad_norm": 2.3130062884964375,
      "language_loss": 0.78122294,
      "learning_rate": 7.563293144324146e-07,
      "loss": 0.80271143,
      "num_input_tokens_seen": 129147295,
      "step": 6006,
      "time_per_iteration": 2.7425918579101562
    },
    {
      "auxiliary_loss_clip": 0.0112055,
      "auxiliary_loss_mlp": 0.01028096,
      "balance_loss_clip": 1.04399025,
      "balance_loss_mlp": 1.02046382,
      "epoch": 0.7222990440690195,
      "flos": 26286359702400.0,
      "grad_norm": 2.590313169473484,
      "language_loss": 0.80129033,
      "learning_rate": 7.557193586838834e-07,
      "loss": 0.82277679,
      "num_input_tokens_seen": 129162660,
      "step": 6007,
      "time_per_iteration": 2.7357687950134277
    },
    {
      "auxiliary_loss_clip": 0.01149437,
      "auxiliary_loss_mlp": 0.01024161,
      "balance_loss_clip": 1.04295063,
      "balance_loss_mlp": 1.01593566,
      "epoch": 0.7224192869596585,
      "flos": 17601687509760.0,
      "grad_norm": 2.9459619839629014,
      "language_loss": 0.70274639,
      "learning_rate": 7.551095916917371e-07,
      "loss": 0.72448236,
      "num_input_tokens_seen": 129179990,
      "step": 6008,
      "time_per_iteration": 2.7165091037750244
    },
    {
      "auxiliary_loss_clip": 0.01118081,
      "auxiliary_loss_mlp": 0.01029725,
      "balance_loss_clip": 1.04581821,
      "balance_loss_mlp": 1.02008128,
      "epoch": 0.7225395298502976,
      "flos": 12932331016320.0,
      "grad_norm": 4.220648446406173,
      "language_loss": 0.66585618,
      "learning_rate": 7.545000135484758e-07,
      "loss": 0.68733424,
      "num_input_tokens_seen": 129197425,
      "step": 6009,
      "time_per_iteration": 3.7225258350372314
    },
    {
      "auxiliary_loss_clip": 0.01144806,
      "auxiliary_loss_mlp": 0.01025052,
      "balance_loss_clip": 1.04660571,
      "balance_loss_mlp": 1.01741123,
      "epoch": 0.7226597727409367,
      "flos": 29643899592960.0,
      "grad_norm": 2.5024164392767148,
      "language_loss": 0.62996793,
      "learning_rate": 7.538906243465714e-07,
      "loss": 0.65166652,
      "num_input_tokens_seen": 129217560,
      "step": 6010,
      "time_per_iteration": 3.716871976852417
    },
    {
      "auxiliary_loss_clip": 0.01135614,
      "auxiliary_loss_mlp": 0.0102902,
      "balance_loss_clip": 1.04359972,
      "balance_loss_mlp": 1.02080107,
      "epoch": 0.7227800156315758,
      "flos": 13771635183360.0,
      "grad_norm": 2.1845142707600598,
      "language_loss": 0.78628623,
      "learning_rate": 7.5328142417847e-07,
      "loss": 0.80793262,
      "num_input_tokens_seen": 129234325,
      "step": 6011,
      "time_per_iteration": 2.7342488765716553
    },
    {
      "auxiliary_loss_clip": 0.01117553,
      "auxiliary_loss_mlp": 0.01020065,
      "balance_loss_clip": 1.04257262,
      "balance_loss_mlp": 1.01286435,
      "epoch": 0.7229002585222148,
      "flos": 20301882554880.0,
      "grad_norm": 1.7135475764038142,
      "language_loss": 0.69258851,
      "learning_rate": 7.526724131365838e-07,
      "loss": 0.71396464,
      "num_input_tokens_seen": 129255280,
      "step": 6012,
      "time_per_iteration": 3.7449557781219482
    },
    {
      "auxiliary_loss_clip": 0.01150681,
      "auxiliary_loss_mlp": 0.01032411,
      "balance_loss_clip": 1.0503757,
      "balance_loss_mlp": 1.02386916,
      "epoch": 0.723020501412854,
      "flos": 16581250033920.0,
      "grad_norm": 1.9128665774176903,
      "language_loss": 0.70317757,
      "learning_rate": 7.520635913133017e-07,
      "loss": 0.72500849,
      "num_input_tokens_seen": 129273910,
      "step": 6013,
      "time_per_iteration": 2.678151845932007
    },
    {
      "auxiliary_loss_clip": 0.01106981,
      "auxiliary_loss_mlp": 0.01028967,
      "balance_loss_clip": 1.03860068,
      "balance_loss_mlp": 1.0201757,
      "epoch": 0.7231407443034931,
      "flos": 28548300908160.0,
      "grad_norm": 1.967338217675199,
      "language_loss": 0.82448554,
      "learning_rate": 7.514549588009798e-07,
      "loss": 0.84584504,
      "num_input_tokens_seen": 129294785,
      "step": 6014,
      "time_per_iteration": 2.8521246910095215
    },
    {
      "auxiliary_loss_clip": 0.01145382,
      "auxiliary_loss_mlp": 0.01027069,
      "balance_loss_clip": 1.04532635,
      "balance_loss_mlp": 1.01913595,
      "epoch": 0.7232609871941321,
      "flos": 30008536508160.0,
      "grad_norm": 2.687807758157342,
      "language_loss": 0.70729488,
      "learning_rate": 7.508465156919492e-07,
      "loss": 0.7290194,
      "num_input_tokens_seen": 129318295,
      "step": 6015,
      "time_per_iteration": 2.7411949634552
    },
    {
      "auxiliary_loss_clip": 0.01141265,
      "auxiliary_loss_mlp": 0.01026525,
      "balance_loss_clip": 1.043926,
      "balance_loss_mlp": 1.01849055,
      "epoch": 0.7233812300847713,
      "flos": 16654005031680.0,
      "grad_norm": 3.021364670174063,
      "language_loss": 0.61523253,
      "learning_rate": 7.502382620785083e-07,
      "loss": 0.63691044,
      "num_input_tokens_seen": 129334845,
      "step": 6016,
      "time_per_iteration": 2.733612060546875
    },
    {
      "auxiliary_loss_clip": 0.01048056,
      "auxiliary_loss_mlp": 0.00754297,
      "balance_loss_clip": 1.01131129,
      "balance_loss_mlp": 1.00106144,
      "epoch": 0.7235014729754103,
      "flos": 67258784050560.0,
      "grad_norm": 0.8063039044616019,
      "language_loss": 0.6255219,
      "learning_rate": 7.496301980529289e-07,
      "loss": 0.64354545,
      "num_input_tokens_seen": 129398055,
      "step": 6017,
      "time_per_iteration": 3.26717209815979
    },
    {
      "auxiliary_loss_clip": 0.01114499,
      "auxiliary_loss_mlp": 0.0102923,
      "balance_loss_clip": 1.04522038,
      "balance_loss_mlp": 1.02094531,
      "epoch": 0.7236217158660494,
      "flos": 26943237671040.0,
      "grad_norm": 2.328888787539915,
      "language_loss": 0.74547398,
      "learning_rate": 7.490223237074547e-07,
      "loss": 0.76691127,
      "num_input_tokens_seen": 129417765,
      "step": 6018,
      "time_per_iteration": 2.7686965465545654
    },
    {
      "auxiliary_loss_clip": 0.01115922,
      "auxiliary_loss_mlp": 0.01028525,
      "balance_loss_clip": 1.04064059,
      "balance_loss_mlp": 1.0206753,
      "epoch": 0.7237419587566886,
      "flos": 29423372042880.0,
      "grad_norm": 2.0958744520427706,
      "language_loss": 0.66670668,
      "learning_rate": 7.484146391342989e-07,
      "loss": 0.68815112,
      "num_input_tokens_seen": 129437560,
      "step": 6019,
      "time_per_iteration": 2.8755598068237305
    },
    {
      "auxiliary_loss_clip": 0.01156382,
      "auxiliary_loss_mlp": 0.01024027,
      "balance_loss_clip": 1.04604006,
      "balance_loss_mlp": 1.01608145,
      "epoch": 0.7238622016473276,
      "flos": 17821496787840.0,
      "grad_norm": 2.1425670931533016,
      "language_loss": 0.57013166,
      "learning_rate": 7.478071444256484e-07,
      "loss": 0.59193575,
      "num_input_tokens_seen": 129455320,
      "step": 6020,
      "time_per_iteration": 2.710054874420166
    },
    {
      "auxiliary_loss_clip": 0.01129355,
      "auxiliary_loss_mlp": 0.01027108,
      "balance_loss_clip": 1.04677343,
      "balance_loss_mlp": 1.01873326,
      "epoch": 0.7239824445379667,
      "flos": 25739117020800.0,
      "grad_norm": 1.9057393619439622,
      "language_loss": 0.79275066,
      "learning_rate": 7.471998396736579e-07,
      "loss": 0.81431532,
      "num_input_tokens_seen": 129475700,
      "step": 6021,
      "time_per_iteration": 2.799029588699341
    },
    {
      "auxiliary_loss_clip": 0.01137462,
      "auxiliary_loss_mlp": 0.01033435,
      "balance_loss_clip": 1.04620969,
      "balance_loss_mlp": 1.02531123,
      "epoch": 0.7241026874286057,
      "flos": 23148916398720.0,
      "grad_norm": 1.930827356254264,
      "language_loss": 0.76161599,
      "learning_rate": 7.465927249704549e-07,
      "loss": 0.78332496,
      "num_input_tokens_seen": 129493585,
      "step": 6022,
      "time_per_iteration": 2.6945877075195312
    },
    {
      "auxiliary_loss_clip": 0.01122684,
      "auxiliary_loss_mlp": 0.0102619,
      "balance_loss_clip": 1.04280257,
      "balance_loss_mlp": 1.01848912,
      "epoch": 0.7242229303192449,
      "flos": 20266905686400.0,
      "grad_norm": 1.92328734013886,
      "language_loss": 0.77407873,
      "learning_rate": 7.459858004081398e-07,
      "loss": 0.79556751,
      "num_input_tokens_seen": 129511555,
      "step": 6023,
      "time_per_iteration": 2.7609241008758545
    },
    {
      "auxiliary_loss_clip": 0.01059495,
      "auxiliary_loss_mlp": 0.01000856,
      "balance_loss_clip": 1.01108027,
      "balance_loss_mlp": 0.99979538,
      "epoch": 0.724343173209884,
      "flos": 62311659684480.0,
      "grad_norm": 0.6722467057558335,
      "language_loss": 0.58006114,
      "learning_rate": 7.453790660787815e-07,
      "loss": 0.60066468,
      "num_input_tokens_seen": 129579650,
      "step": 6024,
      "time_per_iteration": 3.3658950328826904
    },
    {
      "auxiliary_loss_clip": 0.01116133,
      "auxiliary_loss_mlp": 0.0102755,
      "balance_loss_clip": 1.04353833,
      "balance_loss_mlp": 1.01914024,
      "epoch": 0.724463416100523,
      "flos": 35006403813120.0,
      "grad_norm": 1.952071477570262,
      "language_loss": 0.63759917,
      "learning_rate": 7.447725220744214e-07,
      "loss": 0.65903598,
      "num_input_tokens_seen": 129601895,
      "step": 6025,
      "time_per_iteration": 2.8846688270568848
    },
    {
      "auxiliary_loss_clip": 0.01069591,
      "auxiliary_loss_mlp": 0.01028552,
      "balance_loss_clip": 1.0384295,
      "balance_loss_mlp": 1.01979017,
      "epoch": 0.7245836589911622,
      "flos": 21871968923520.0,
      "grad_norm": 2.677444548299487,
      "language_loss": 0.77405453,
      "learning_rate": 7.441661684870717e-07,
      "loss": 0.79503596,
      "num_input_tokens_seen": 129622150,
      "step": 6026,
      "time_per_iteration": 3.915194511413574
    },
    {
      "auxiliary_loss_clip": 0.01104553,
      "auxiliary_loss_mlp": 0.01025917,
      "balance_loss_clip": 1.03927207,
      "balance_loss_mlp": 1.01781678,
      "epoch": 0.7247039018818012,
      "flos": 23006494972800.0,
      "grad_norm": 1.756041443668043,
      "language_loss": 0.81815833,
      "learning_rate": 7.435600054087152e-07,
      "loss": 0.839463,
      "num_input_tokens_seen": 129644315,
      "step": 6027,
      "time_per_iteration": 3.0668983459472656
    },
    {
      "auxiliary_loss_clip": 0.01137134,
      "auxiliary_loss_mlp": 0.01027936,
      "balance_loss_clip": 1.04701078,
      "balance_loss_mlp": 1.019997,
      "epoch": 0.7248241447724403,
      "flos": 31722588587520.0,
      "grad_norm": 2.2576124149815837,
      "language_loss": 0.74267852,
      "learning_rate": 7.42954032931308e-07,
      "loss": 0.7643292,
      "num_input_tokens_seen": 129665355,
      "step": 6028,
      "time_per_iteration": 2.8181512355804443
    },
    {
      "auxiliary_loss_clip": 0.0113069,
      "auxiliary_loss_mlp": 0.01028764,
      "balance_loss_clip": 1.04380274,
      "balance_loss_mlp": 1.02077687,
      "epoch": 0.7249443876630794,
      "flos": 34896984007680.0,
      "grad_norm": 1.928254030738363,
      "language_loss": 0.74284965,
      "learning_rate": 7.423482511467733e-07,
      "loss": 0.76444411,
      "num_input_tokens_seen": 129686125,
      "step": 6029,
      "time_per_iteration": 2.8622498512268066
    },
    {
      "auxiliary_loss_clip": 0.01133203,
      "auxiliary_loss_mlp": 0.01032749,
      "balance_loss_clip": 1.04898751,
      "balance_loss_mlp": 1.02514303,
      "epoch": 0.7250646305537185,
      "flos": 26359294268160.0,
      "grad_norm": 2.796973238524974,
      "language_loss": 0.65261054,
      "learning_rate": 7.417426601470099e-07,
      "loss": 0.67427003,
      "num_input_tokens_seen": 129706485,
      "step": 6030,
      "time_per_iteration": 2.743992328643799
    },
    {
      "auxiliary_loss_clip": 0.01124276,
      "auxiliary_loss_mlp": 0.01027697,
      "balance_loss_clip": 1.04553723,
      "balance_loss_mlp": 1.0192275,
      "epoch": 0.7251848734443576,
      "flos": 30081614728320.0,
      "grad_norm": 2.4905989094324585,
      "language_loss": 0.7865606,
      "learning_rate": 7.411372600238841e-07,
      "loss": 0.80808032,
      "num_input_tokens_seen": 129727100,
      "step": 6031,
      "time_per_iteration": 2.897355794906616
    },
    {
      "auxiliary_loss_clip": 0.01131486,
      "auxiliary_loss_mlp": 0.01022779,
      "balance_loss_clip": 1.04284143,
      "balance_loss_mlp": 1.01505446,
      "epoch": 0.7253051163349967,
      "flos": 17785262943360.0,
      "grad_norm": 1.8167664631542946,
      "language_loss": 0.73962182,
      "learning_rate": 7.405320508692346e-07,
      "loss": 0.76116449,
      "num_input_tokens_seen": 129745840,
      "step": 6032,
      "time_per_iteration": 2.8155980110168457
    },
    {
      "auxiliary_loss_clip": 0.01102837,
      "auxiliary_loss_mlp": 0.00762102,
      "balance_loss_clip": 1.04076004,
      "balance_loss_mlp": 1.00039172,
      "epoch": 0.7254253592256358,
      "flos": 12641346938880.0,
      "grad_norm": 1.7428367944279393,
      "language_loss": 0.7540164,
      "learning_rate": 7.399270327748727e-07,
      "loss": 0.7726658,
      "num_input_tokens_seen": 129763500,
      "step": 6033,
      "time_per_iteration": 2.8193519115448
    },
    {
      "auxiliary_loss_clip": 0.01158422,
      "auxiliary_loss_mlp": 0.01026113,
      "balance_loss_clip": 1.04728031,
      "balance_loss_mlp": 1.01847148,
      "epoch": 0.7255456021162748,
      "flos": 27199208966400.0,
      "grad_norm": 2.0895662870874645,
      "language_loss": 0.74551785,
      "learning_rate": 7.39322205832577e-07,
      "loss": 0.76736319,
      "num_input_tokens_seen": 129784390,
      "step": 6034,
      "time_per_iteration": 3.6702284812927246
    },
    {
      "auxiliary_loss_clip": 0.01109407,
      "auxiliary_loss_mlp": 0.01027741,
      "balance_loss_clip": 1.04453671,
      "balance_loss_mlp": 1.01979589,
      "epoch": 0.725665845006914,
      "flos": 21288205088640.0,
      "grad_norm": 2.7007300783404844,
      "language_loss": 0.8095479,
      "learning_rate": 7.387175701341009e-07,
      "loss": 0.83091938,
      "num_input_tokens_seen": 129803060,
      "step": 6035,
      "time_per_iteration": 2.7690043449401855
    },
    {
      "auxiliary_loss_clip": 0.01113834,
      "auxiliary_loss_mlp": 0.01033506,
      "balance_loss_clip": 1.04257703,
      "balance_loss_mlp": 1.02432108,
      "epoch": 0.7257860878975531,
      "flos": 16033684129920.0,
      "grad_norm": 2.4834737996700187,
      "language_loss": 0.71885252,
      "learning_rate": 7.381131257711659e-07,
      "loss": 0.74032593,
      "num_input_tokens_seen": 129820165,
      "step": 6036,
      "time_per_iteration": 3.717690944671631
    },
    {
      "auxiliary_loss_clip": 0.01107946,
      "auxiliary_loss_mlp": 0.01026273,
      "balance_loss_clip": 1.04149675,
      "balance_loss_mlp": 1.01851201,
      "epoch": 0.7259063307881921,
      "flos": 12129943052160.0,
      "grad_norm": 2.0086846733395456,
      "language_loss": 0.83802783,
      "learning_rate": 7.375088728354677e-07,
      "loss": 0.85936999,
      "num_input_tokens_seen": 129835195,
      "step": 6037,
      "time_per_iteration": 2.7849440574645996
    },
    {
      "auxiliary_loss_clip": 0.01126389,
      "auxiliary_loss_mlp": 0.01024024,
      "balance_loss_clip": 1.04401231,
      "balance_loss_mlp": 1.0159353,
      "epoch": 0.7260265736788313,
      "flos": 30443845432320.0,
      "grad_norm": 2.1079552582353625,
      "language_loss": 0.67575085,
      "learning_rate": 7.369048114186691e-07,
      "loss": 0.69725502,
      "num_input_tokens_seen": 129856240,
      "step": 6038,
      "time_per_iteration": 3.6814029216766357
    },
    {
      "auxiliary_loss_clip": 0.01086666,
      "auxiliary_loss_mlp": 0.0102205,
      "balance_loss_clip": 1.03870058,
      "balance_loss_mlp": 1.01480532,
      "epoch": 0.7261468165694703,
      "flos": 21142264129920.0,
      "grad_norm": 7.29937695405641,
      "language_loss": 0.83132297,
      "learning_rate": 7.363009416124055e-07,
      "loss": 0.85241008,
      "num_input_tokens_seen": 129875565,
      "step": 6039,
      "time_per_iteration": 2.8933963775634766
    },
    {
      "auxiliary_loss_clip": 0.01124253,
      "auxiliary_loss_mlp": 0.00762583,
      "balance_loss_clip": 1.04265785,
      "balance_loss_mlp": 1.00041473,
      "epoch": 0.7262670594601094,
      "flos": 22306308180480.0,
      "grad_norm": 2.140084395241763,
      "language_loss": 0.63119543,
      "learning_rate": 7.356972635082852e-07,
      "loss": 0.65006375,
      "num_input_tokens_seen": 129894420,
      "step": 6040,
      "time_per_iteration": 3.0119943618774414
    },
    {
      "auxiliary_loss_clip": 0.01109865,
      "auxiliary_loss_mlp": 0.01025183,
      "balance_loss_clip": 1.04107618,
      "balance_loss_mlp": 1.01703167,
      "epoch": 0.7263873023507486,
      "flos": 25335049950720.0,
      "grad_norm": 2.4792417137953517,
      "language_loss": 0.75432992,
      "learning_rate": 7.35093777197884e-07,
      "loss": 0.77568042,
      "num_input_tokens_seen": 129914490,
      "step": 6041,
      "time_per_iteration": 2.7988743782043457
    },
    {
      "auxiliary_loss_clip": 0.01126677,
      "auxiliary_loss_mlp": 0.00762372,
      "balance_loss_clip": 1.04370368,
      "balance_loss_mlp": 1.00041163,
      "epoch": 0.7265075452413876,
      "flos": 23878621192320.0,
      "grad_norm": 2.482167156111018,
      "language_loss": 0.86211002,
      "learning_rate": 7.344904827727525e-07,
      "loss": 0.88100052,
      "num_input_tokens_seen": 129931670,
      "step": 6042,
      "time_per_iteration": 2.7126402854919434
    },
    {
      "auxiliary_loss_clip": 0.01140226,
      "auxiliary_loss_mlp": 0.01020346,
      "balance_loss_clip": 1.04588199,
      "balance_loss_mlp": 1.01243973,
      "epoch": 0.7266277881320267,
      "flos": 28724549967360.0,
      "grad_norm": 4.372657017025713,
      "language_loss": 0.73317266,
      "learning_rate": 7.338873803244076e-07,
      "loss": 0.75477839,
      "num_input_tokens_seen": 129946905,
      "step": 6043,
      "time_per_iteration": 2.7279417514801025
    },
    {
      "auxiliary_loss_clip": 0.01142055,
      "auxiliary_loss_mlp": 0.01026426,
      "balance_loss_clip": 1.04125726,
      "balance_loss_mlp": 1.01909518,
      "epoch": 0.7267480310226658,
      "flos": 24863507182080.0,
      "grad_norm": 2.030650400613761,
      "language_loss": 0.80984128,
      "learning_rate": 7.332844699443401e-07,
      "loss": 0.8315261,
      "num_input_tokens_seen": 129965505,
      "step": 6044,
      "time_per_iteration": 2.7593994140625
    },
    {
      "auxiliary_loss_clip": 0.01171266,
      "auxiliary_loss_mlp": 0.01026855,
      "balance_loss_clip": 1.04809642,
      "balance_loss_mlp": 1.01933002,
      "epoch": 0.7268682739133049,
      "flos": 27198490694400.0,
      "grad_norm": 2.3051937190688334,
      "language_loss": 0.75265205,
      "learning_rate": 7.326817517240121e-07,
      "loss": 0.77463317,
      "num_input_tokens_seen": 129987210,
      "step": 6045,
      "time_per_iteration": 2.6661744117736816
    },
    {
      "auxiliary_loss_clip": 0.01089077,
      "auxiliary_loss_mlp": 0.01024623,
      "balance_loss_clip": 1.04202747,
      "balance_loss_mlp": 1.0179708,
      "epoch": 0.7269885168039439,
      "flos": 33508138688640.0,
      "grad_norm": 1.7752688832031926,
      "language_loss": 0.83330548,
      "learning_rate": 7.320792257548545e-07,
      "loss": 0.85444248,
      "num_input_tokens_seen": 130008385,
      "step": 6046,
      "time_per_iteration": 2.9170045852661133
    },
    {
      "auxiliary_loss_clip": 0.01137164,
      "auxiliary_loss_mlp": 0.00762677,
      "balance_loss_clip": 1.04399633,
      "balance_loss_mlp": 1.00043249,
      "epoch": 0.7271087596945831,
      "flos": 24313750548480.0,
      "grad_norm": 2.080449492270503,
      "language_loss": 0.76224178,
      "learning_rate": 7.314768921282704e-07,
      "loss": 0.78124022,
      "num_input_tokens_seen": 130029040,
      "step": 6047,
      "time_per_iteration": 2.7603509426116943
    },
    {
      "auxiliary_loss_clip": 0.01114474,
      "auxiliary_loss_mlp": 0.01025172,
      "balance_loss_clip": 1.0426805,
      "balance_loss_mlp": 1.01710129,
      "epoch": 0.7272290025852222,
      "flos": 23805147922560.0,
      "grad_norm": 3.315274210742085,
      "language_loss": 0.72073114,
      "learning_rate": 7.30874750935633e-07,
      "loss": 0.7421276,
      "num_input_tokens_seen": 130048725,
      "step": 6048,
      "time_per_iteration": 2.737684965133667
    },
    {
      "auxiliary_loss_clip": 0.01131053,
      "auxiliary_loss_mlp": 0.01023802,
      "balance_loss_clip": 1.04202724,
      "balance_loss_mlp": 1.01596403,
      "epoch": 0.7273492454758612,
      "flos": 16720367408640.0,
      "grad_norm": 2.007901238954025,
      "language_loss": 0.79346019,
      "learning_rate": 7.30272802268286e-07,
      "loss": 0.81500876,
      "num_input_tokens_seen": 130065720,
      "step": 6049,
      "time_per_iteration": 2.835472822189331
    },
    {
      "auxiliary_loss_clip": 0.01108372,
      "auxiliary_loss_mlp": 0.010242,
      "balance_loss_clip": 1.04492891,
      "balance_loss_mlp": 1.01730633,
      "epoch": 0.7274694883665004,
      "flos": 28031330413440.0,
      "grad_norm": 1.9569689698100574,
      "language_loss": 0.76101661,
      "learning_rate": 7.29671046217547e-07,
      "loss": 0.78234231,
      "num_input_tokens_seen": 130084830,
      "step": 6050,
      "time_per_iteration": 2.7901787757873535
    },
    {
      "auxiliary_loss_clip": 0.01132866,
      "auxiliary_loss_mlp": 0.01024573,
      "balance_loss_clip": 1.0434469,
      "balance_loss_mlp": 1.01722968,
      "epoch": 0.7275897312571394,
      "flos": 30372706546560.0,
      "grad_norm": 2.09769067890971,
      "language_loss": 0.81823903,
      "learning_rate": 7.290694828746988e-07,
      "loss": 0.83981335,
      "num_input_tokens_seen": 130104495,
      "step": 6051,
      "time_per_iteration": 2.8067708015441895
    },
    {
      "auxiliary_loss_clip": 0.01093108,
      "auxiliary_loss_mlp": 0.01026578,
      "balance_loss_clip": 1.03941715,
      "balance_loss_mlp": 1.01866269,
      "epoch": 0.7277099741477785,
      "flos": 19204775498880.0,
      "grad_norm": 2.505911271607699,
      "language_loss": 0.85863084,
      "learning_rate": 7.284681123310004e-07,
      "loss": 0.87982768,
      "num_input_tokens_seen": 130123210,
      "step": 6052,
      "time_per_iteration": 3.875277280807495
    },
    {
      "auxiliary_loss_clip": 0.01130694,
      "auxiliary_loss_mlp": 0.01024834,
      "balance_loss_clip": 1.04267883,
      "balance_loss_mlp": 1.01674223,
      "epoch": 0.7278302170384175,
      "flos": 20667884186880.0,
      "grad_norm": 1.7542110481119277,
      "language_loss": 0.79704893,
      "learning_rate": 7.27866934677678e-07,
      "loss": 0.81860423,
      "num_input_tokens_seen": 130142880,
      "step": 6053,
      "time_per_iteration": 2.6985037326812744
    },
    {
      "auxiliary_loss_clip": 0.01112324,
      "auxiliary_loss_mlp": 0.01028874,
      "balance_loss_clip": 1.04107988,
      "balance_loss_mlp": 1.02081561,
      "epoch": 0.7279504599290567,
      "flos": 19093200877440.0,
      "grad_norm": 1.834067799619196,
      "language_loss": 0.78337079,
      "learning_rate": 7.272659500059297e-07,
      "loss": 0.80478281,
      "num_input_tokens_seen": 130160220,
      "step": 6054,
      "time_per_iteration": 2.728536605834961
    },
    {
      "auxiliary_loss_clip": 0.01121225,
      "auxiliary_loss_mlp": 0.01027866,
      "balance_loss_clip": 1.04799986,
      "balance_loss_mlp": 1.01978314,
      "epoch": 0.7280707028196958,
      "flos": 19062174504960.0,
      "grad_norm": 6.375764154389191,
      "language_loss": 0.80240059,
      "learning_rate": 7.266651584069264e-07,
      "loss": 0.82389146,
      "num_input_tokens_seen": 130177885,
      "step": 6055,
      "time_per_iteration": 2.763233184814453
    },
    {
      "auxiliary_loss_clip": 0.0114165,
      "auxiliary_loss_mlp": 0.01027401,
      "balance_loss_clip": 1.04606259,
      "balance_loss_mlp": 1.01981354,
      "epoch": 0.7281909457103348,
      "flos": 37196308293120.0,
      "grad_norm": 1.9084029024754094,
      "language_loss": 0.5685603,
      "learning_rate": 7.260645599718045e-07,
      "loss": 0.59025079,
      "num_input_tokens_seen": 130204240,
      "step": 6056,
      "time_per_iteration": 2.78006911277771
    },
    {
      "auxiliary_loss_clip": 0.01124196,
      "auxiliary_loss_mlp": 0.01028666,
      "balance_loss_clip": 1.0452975,
      "balance_loss_mlp": 1.01965928,
      "epoch": 0.728311188600974,
      "flos": 20667094087680.0,
      "grad_norm": 2.326972941307115,
      "language_loss": 0.67386711,
      "learning_rate": 7.254641547916767e-07,
      "loss": 0.69539571,
      "num_input_tokens_seen": 130221735,
      "step": 6057,
      "time_per_iteration": 2.766164779663086
    },
    {
      "auxiliary_loss_clip": 0.0112399,
      "auxiliary_loss_mlp": 0.01031016,
      "balance_loss_clip": 1.04083443,
      "balance_loss_mlp": 1.02290726,
      "epoch": 0.728431431491613,
      "flos": 28840685616000.0,
      "grad_norm": 1.9052149582828992,
      "language_loss": 0.69372529,
      "learning_rate": 7.248639429576226e-07,
      "loss": 0.71527541,
      "num_input_tokens_seen": 130241190,
      "step": 6058,
      "time_per_iteration": 2.8117458820343018
    },
    {
      "auxiliary_loss_clip": 0.01155417,
      "auxiliary_loss_mlp": 0.01023326,
      "balance_loss_clip": 1.04612017,
      "balance_loss_mlp": 1.01509511,
      "epoch": 0.7285516743822521,
      "flos": 25991856092160.0,
      "grad_norm": 1.645231318846922,
      "language_loss": 0.71962297,
      "learning_rate": 7.242639245606959e-07,
      "loss": 0.74141037,
      "num_input_tokens_seen": 130260980,
      "step": 6059,
      "time_per_iteration": 2.649947166442871
    },
    {
      "auxiliary_loss_clip": 0.01100957,
      "auxiliary_loss_mlp": 0.01029071,
      "balance_loss_clip": 1.04200387,
      "balance_loss_mlp": 1.02063668,
      "epoch": 0.7286719172728913,
      "flos": 16399721675520.0,
      "grad_norm": 1.8804573667717075,
      "language_loss": 0.82461774,
      "learning_rate": 7.236640996919168e-07,
      "loss": 0.84591806,
      "num_input_tokens_seen": 130280025,
      "step": 6060,
      "time_per_iteration": 2.800928831100464
    },
    {
      "auxiliary_loss_clip": 0.01118762,
      "auxiliary_loss_mlp": 0.01029533,
      "balance_loss_clip": 1.0386914,
      "balance_loss_mlp": 1.02122414,
      "epoch": 0.7287921601635303,
      "flos": 22018161277440.0,
      "grad_norm": 2.1297225511818927,
      "language_loss": 0.70419228,
      "learning_rate": 7.230644684422782e-07,
      "loss": 0.72567523,
      "num_input_tokens_seen": 130300255,
      "step": 6061,
      "time_per_iteration": 3.6006972789764404
    },
    {
      "auxiliary_loss_clip": 0.01158335,
      "auxiliary_loss_mlp": 0.01023972,
      "balance_loss_clip": 1.04962921,
      "balance_loss_mlp": 1.01589572,
      "epoch": 0.7289124030541694,
      "flos": 24600927784320.0,
      "grad_norm": 1.7739342213788882,
      "language_loss": 0.8172968,
      "learning_rate": 7.224650309027451e-07,
      "loss": 0.83911991,
      "num_input_tokens_seen": 130320005,
      "step": 6062,
      "time_per_iteration": 3.6484897136688232
    },
    {
      "auxiliary_loss_clip": 0.01062909,
      "auxiliary_loss_mlp": 0.01026076,
      "balance_loss_clip": 1.04041493,
      "balance_loss_mlp": 1.01835132,
      "epoch": 0.7290326459448085,
      "flos": 21393638484480.0,
      "grad_norm": 1.9258548607116066,
      "language_loss": 0.68795985,
      "learning_rate": 7.218657871642506e-07,
      "loss": 0.70884973,
      "num_input_tokens_seen": 130338810,
      "step": 6063,
      "time_per_iteration": 3.801042079925537
    },
    {
      "auxiliary_loss_clip": 0.01114471,
      "auxiliary_loss_mlp": 0.01024437,
      "balance_loss_clip": 1.04208827,
      "balance_loss_mlp": 1.01592529,
      "epoch": 0.7291528888354476,
      "flos": 18587686821120.0,
      "grad_norm": 2.1371092167661554,
      "language_loss": 0.62571847,
      "learning_rate": 7.212667373177012e-07,
      "loss": 0.6471076,
      "num_input_tokens_seen": 130353805,
      "step": 6064,
      "time_per_iteration": 2.735034465789795
    },
    {
      "auxiliary_loss_clip": 0.0114164,
      "auxiliary_loss_mlp": 0.01026465,
      "balance_loss_clip": 1.04214263,
      "balance_loss_mlp": 1.01872265,
      "epoch": 0.7292731317260867,
      "flos": 18951066760320.0,
      "grad_norm": 1.8872797692465453,
      "language_loss": 0.75053895,
      "learning_rate": 7.206678814539704e-07,
      "loss": 0.77222002,
      "num_input_tokens_seen": 130372105,
      "step": 6065,
      "time_per_iteration": 2.692308187484741
    },
    {
      "auxiliary_loss_clip": 0.01106553,
      "auxiliary_loss_mlp": 0.01027141,
      "balance_loss_clip": 1.043051,
      "balance_loss_mlp": 1.0197823,
      "epoch": 0.7293933746167258,
      "flos": 21067569797760.0,
      "grad_norm": 2.307465785141914,
      "language_loss": 0.72990513,
      "learning_rate": 7.20069219663904e-07,
      "loss": 0.75124204,
      "num_input_tokens_seen": 130391990,
      "step": 6066,
      "time_per_iteration": 2.759010076522827
    },
    {
      "auxiliary_loss_clip": 0.01108277,
      "auxiliary_loss_mlp": 0.01024906,
      "balance_loss_clip": 1.04161394,
      "balance_loss_mlp": 1.01750302,
      "epoch": 0.7295136175073649,
      "flos": 22453326547200.0,
      "grad_norm": 1.7280037690376018,
      "language_loss": 0.79705667,
      "learning_rate": 7.1947075203832e-07,
      "loss": 0.81838846,
      "num_input_tokens_seen": 130411970,
      "step": 6067,
      "time_per_iteration": 2.7676851749420166
    },
    {
      "auxiliary_loss_clip": 0.01033488,
      "auxiliary_loss_mlp": 0.01002095,
      "balance_loss_clip": 1.03697395,
      "balance_loss_mlp": 1.0010041,
      "epoch": 0.7296338603980039,
      "flos": 56125506648960.0,
      "grad_norm": 0.8684187682471607,
      "language_loss": 0.60063928,
      "learning_rate": 7.188724786680049e-07,
      "loss": 0.6209951,
      "num_input_tokens_seen": 130472440,
      "step": 6068,
      "time_per_iteration": 3.3244030475616455
    },
    {
      "auxiliary_loss_clip": 0.01113628,
      "auxiliary_loss_mlp": 0.01027492,
      "balance_loss_clip": 1.0405457,
      "balance_loss_mlp": 1.01936817,
      "epoch": 0.7297541032886431,
      "flos": 25228287751680.0,
      "grad_norm": 1.6874330878375077,
      "language_loss": 0.75673217,
      "learning_rate": 7.182743996437162e-07,
      "loss": 0.77814341,
      "num_input_tokens_seen": 130491975,
      "step": 6069,
      "time_per_iteration": 2.7453136444091797
    },
    {
      "auxiliary_loss_clip": 0.0112068,
      "auxiliary_loss_mlp": 0.01027538,
      "balance_loss_clip": 1.04310012,
      "balance_loss_mlp": 1.01884174,
      "epoch": 0.7298743461792822,
      "flos": 26467600752000.0,
      "grad_norm": 4.44276524920708,
      "language_loss": 0.6887536,
      "learning_rate": 7.176765150561819e-07,
      "loss": 0.71023577,
      "num_input_tokens_seen": 130510580,
      "step": 6070,
      "time_per_iteration": 2.7165894508361816
    },
    {
      "auxiliary_loss_clip": 0.01138224,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04452944,
      "balance_loss_mlp": 1.0187515,
      "epoch": 0.7299945890699212,
      "flos": 19569053278080.0,
      "grad_norm": 2.411040991956311,
      "language_loss": 0.7989105,
      "learning_rate": 7.170788249961002e-07,
      "loss": 0.82056046,
      "num_input_tokens_seen": 130529090,
      "step": 6071,
      "time_per_iteration": 2.6356208324432373
    },
    {
      "auxiliary_loss_clip": 0.01118493,
      "auxiliary_loss_mlp": 0.0102441,
      "balance_loss_clip": 1.03990316,
      "balance_loss_mlp": 1.01652467,
      "epoch": 0.7301148319605604,
      "flos": 22928963466240.0,
      "grad_norm": 2.2280136244940825,
      "language_loss": 0.88124871,
      "learning_rate": 7.164813295541418e-07,
      "loss": 0.90267777,
      "num_input_tokens_seen": 130548655,
      "step": 6072,
      "time_per_iteration": 2.708411455154419
    },
    {
      "auxiliary_loss_clip": 0.01154579,
      "auxiliary_loss_mlp": 0.00762615,
      "balance_loss_clip": 1.04374087,
      "balance_loss_mlp": 1.00036931,
      "epoch": 0.7302350748511994,
      "flos": 25369703596800.0,
      "grad_norm": 1.7215093467655724,
      "language_loss": 0.70061243,
      "learning_rate": 7.15884028820944e-07,
      "loss": 0.71978438,
      "num_input_tokens_seen": 130567710,
      "step": 6073,
      "time_per_iteration": 2.5794787406921387
    },
    {
      "auxiliary_loss_clip": 0.01172197,
      "auxiliary_loss_mlp": 0.00762206,
      "balance_loss_clip": 1.04856694,
      "balance_loss_mlp": 1.00032258,
      "epoch": 0.7303553177418385,
      "flos": 27819170732160.0,
      "grad_norm": 2.3240126499469467,
      "language_loss": 0.60785758,
      "learning_rate": 7.152869228871185e-07,
      "loss": 0.62720162,
      "num_input_tokens_seen": 130590195,
      "step": 6074,
      "time_per_iteration": 2.636364698410034
    },
    {
      "auxiliary_loss_clip": 0.0112871,
      "auxiliary_loss_mlp": 0.01023734,
      "balance_loss_clip": 1.04829144,
      "balance_loss_mlp": 1.01562774,
      "epoch": 0.7304755606324776,
      "flos": 24426510318720.0,
      "grad_norm": 2.6276938901828615,
      "language_loss": 0.72262788,
      "learning_rate": 7.146900118432457e-07,
      "loss": 0.74415231,
      "num_input_tokens_seen": 130609940,
      "step": 6075,
      "time_per_iteration": 2.718867540359497
    },
    {
      "auxiliary_loss_clip": 0.01086185,
      "auxiliary_loss_mlp": 0.00762641,
      "balance_loss_clip": 1.03793788,
      "balance_loss_mlp": 1.00040531,
      "epoch": 0.7305958035231167,
      "flos": 23840483927040.0,
      "grad_norm": 1.9834088334599076,
      "language_loss": 0.85839343,
      "learning_rate": 7.140932957798753e-07,
      "loss": 0.87688166,
      "num_input_tokens_seen": 130628380,
      "step": 6076,
      "time_per_iteration": 2.8692574501037598
    },
    {
      "auxiliary_loss_clip": 0.0113195,
      "auxiliary_loss_mlp": 0.01029847,
      "balance_loss_clip": 1.04672301,
      "balance_loss_mlp": 1.0220747,
      "epoch": 0.7307160464137558,
      "flos": 16726939597440.0,
      "grad_norm": 2.5368909430034057,
      "language_loss": 0.71122634,
      "learning_rate": 7.134967747875309e-07,
      "loss": 0.73284429,
      "num_input_tokens_seen": 130646590,
      "step": 6077,
      "time_per_iteration": 2.713362216949463
    },
    {
      "auxiliary_loss_clip": 0.01140705,
      "auxiliary_loss_mlp": 0.01034214,
      "balance_loss_clip": 1.0441103,
      "balance_loss_mlp": 1.02647769,
      "epoch": 0.7308362893043949,
      "flos": 21798280172160.0,
      "grad_norm": 1.817680328197745,
      "language_loss": 0.81895643,
      "learning_rate": 7.129004489567014e-07,
      "loss": 0.84070563,
      "num_input_tokens_seen": 130664070,
      "step": 6078,
      "time_per_iteration": 3.6647896766662598
    },
    {
      "auxiliary_loss_clip": 0.0114512,
      "auxiliary_loss_mlp": 0.01025259,
      "balance_loss_clip": 1.04605293,
      "balance_loss_mlp": 1.01731968,
      "epoch": 0.730956532195034,
      "flos": 10707377840640.0,
      "grad_norm": 2.394091707482937,
      "language_loss": 0.78432679,
      "learning_rate": 7.123043183778512e-07,
      "loss": 0.80603063,
      "num_input_tokens_seen": 130681400,
      "step": 6079,
      "time_per_iteration": 2.705822229385376
    },
    {
      "auxiliary_loss_clip": 0.01135402,
      "auxiliary_loss_mlp": 0.01031378,
      "balance_loss_clip": 1.04913723,
      "balance_loss_mlp": 1.02294397,
      "epoch": 0.731076775085673,
      "flos": 19791987039360.0,
      "grad_norm": 1.9061653008492108,
      "language_loss": 0.65289915,
      "learning_rate": 7.117083831414114e-07,
      "loss": 0.67456692,
      "num_input_tokens_seen": 130700675,
      "step": 6080,
      "time_per_iteration": 2.757951259613037
    },
    {
      "auxiliary_loss_clip": 0.01135703,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.04552341,
      "balance_loss_mlp": 1.01623535,
      "epoch": 0.7311970179763122,
      "flos": 20447033414400.0,
      "grad_norm": 3.040748406249425,
      "language_loss": 0.69684726,
      "learning_rate": 7.11112643337787e-07,
      "loss": 0.71844792,
      "num_input_tokens_seen": 130719720,
      "step": 6081,
      "time_per_iteration": 2.6531012058258057
    },
    {
      "auxiliary_loss_clip": 0.01129201,
      "auxiliary_loss_mlp": 0.00762688,
      "balance_loss_clip": 1.04554164,
      "balance_loss_mlp": 1.00040054,
      "epoch": 0.7313172608669513,
      "flos": 18513818501760.0,
      "grad_norm": 3.6295511061364896,
      "language_loss": 0.76315963,
      "learning_rate": 7.10517099057349e-07,
      "loss": 0.7820785,
      "num_input_tokens_seen": 130736670,
      "step": 6082,
      "time_per_iteration": 2.7091174125671387
    },
    {
      "auxiliary_loss_clip": 0.01081108,
      "auxiliary_loss_mlp": 0.01032946,
      "balance_loss_clip": 1.03550863,
      "balance_loss_mlp": 1.02430952,
      "epoch": 0.7314375037575903,
      "flos": 16180738410240.0,
      "grad_norm": 9.15932052474182,
      "language_loss": 0.61594158,
      "learning_rate": 7.099217503904411e-07,
      "loss": 0.6370821,
      "num_input_tokens_seen": 130754525,
      "step": 6083,
      "time_per_iteration": 2.785996437072754
    },
    {
      "auxiliary_loss_clip": 0.01127374,
      "auxiliary_loss_mlp": 0.0102329,
      "balance_loss_clip": 1.04813886,
      "balance_loss_mlp": 1.01591992,
      "epoch": 0.7315577466482295,
      "flos": 17967940536960.0,
      "grad_norm": 1.7804352428748682,
      "language_loss": 0.898085,
      "learning_rate": 7.093265974273788e-07,
      "loss": 0.91959155,
      "num_input_tokens_seen": 130772420,
      "step": 6084,
      "time_per_iteration": 2.705113649368286
    },
    {
      "auxiliary_loss_clip": 0.01123288,
      "auxiliary_loss_mlp": 0.01025541,
      "balance_loss_clip": 1.03804016,
      "balance_loss_mlp": 1.0175308,
      "epoch": 0.7316779895388685,
      "flos": 18405440190720.0,
      "grad_norm": 2.6181306938300506,
      "language_loss": 0.72729433,
      "learning_rate": 7.087316402584447e-07,
      "loss": 0.74878263,
      "num_input_tokens_seen": 130791245,
      "step": 6085,
      "time_per_iteration": 2.775045871734619
    },
    {
      "auxiliary_loss_clip": 0.01106663,
      "auxiliary_loss_mlp": 0.01026754,
      "balance_loss_clip": 1.0394218,
      "balance_loss_mlp": 1.01933599,
      "epoch": 0.7317982324295076,
      "flos": 17928294900480.0,
      "grad_norm": 1.8923619204828082,
      "language_loss": 0.86167639,
      "learning_rate": 7.081368789738953e-07,
      "loss": 0.88301063,
      "num_input_tokens_seen": 130808445,
      "step": 6086,
      "time_per_iteration": 3.624467134475708
    },
    {
      "auxiliary_loss_clip": 0.01155727,
      "auxiliary_loss_mlp": 0.01025266,
      "balance_loss_clip": 1.04422665,
      "balance_loss_mlp": 1.01793158,
      "epoch": 0.7319184753201466,
      "flos": 27229840289280.0,
      "grad_norm": 22.613462038270402,
      "language_loss": 0.77931517,
      "learning_rate": 7.075423136639537e-07,
      "loss": 0.80112505,
      "num_input_tokens_seen": 130827700,
      "step": 6087,
      "time_per_iteration": 2.760847568511963
    },
    {
      "auxiliary_loss_clip": 0.01135629,
      "auxiliary_loss_mlp": 0.01025805,
      "balance_loss_clip": 1.04348397,
      "balance_loss_mlp": 1.01772857,
      "epoch": 0.7320387182107858,
      "flos": 37448544574080.0,
      "grad_norm": 2.2922386803469377,
      "language_loss": 0.74616981,
      "learning_rate": 7.069479444188149e-07,
      "loss": 0.76778412,
      "num_input_tokens_seen": 130848290,
      "step": 6088,
      "time_per_iteration": 3.74074387550354
    },
    {
      "auxiliary_loss_clip": 0.01112714,
      "auxiliary_loss_mlp": 0.01026945,
      "balance_loss_clip": 1.04209876,
      "balance_loss_mlp": 1.01936948,
      "epoch": 0.7321589611014249,
      "flos": 17859023521920.0,
      "grad_norm": 1.7023766760317023,
      "language_loss": 0.82421935,
      "learning_rate": 7.063537713286453e-07,
      "loss": 0.84561592,
      "num_input_tokens_seen": 130865970,
      "step": 6089,
      "time_per_iteration": 3.6059422492980957
    },
    {
      "auxiliary_loss_clip": 0.01144308,
      "auxiliary_loss_mlp": 0.01023492,
      "balance_loss_clip": 1.04559696,
      "balance_loss_mlp": 1.01497471,
      "epoch": 0.7322792039920639,
      "flos": 26100593539200.0,
      "grad_norm": 1.8222338717315534,
      "language_loss": 0.80884135,
      "learning_rate": 7.057597944835803e-07,
      "loss": 0.83051932,
      "num_input_tokens_seen": 130885245,
      "step": 6090,
      "time_per_iteration": 2.7068095207214355
    },
    {
      "auxiliary_loss_clip": 0.01110756,
      "auxiliary_loss_mlp": 0.01024924,
      "balance_loss_clip": 1.04202509,
      "balance_loss_mlp": 1.01728845,
      "epoch": 0.7323994468827031,
      "flos": 25369093065600.0,
      "grad_norm": 2.198378914384834,
      "language_loss": 0.74528158,
      "learning_rate": 7.051660139737253e-07,
      "loss": 0.7666384,
      "num_input_tokens_seen": 130903465,
      "step": 6091,
      "time_per_iteration": 2.795067071914673
    },
    {
      "auxiliary_loss_clip": 0.01133909,
      "auxiliary_loss_mlp": 0.01023495,
      "balance_loss_clip": 1.04401028,
      "balance_loss_mlp": 1.01568127,
      "epoch": 0.7325196897733421,
      "flos": 26907075653760.0,
      "grad_norm": 5.443557305817921,
      "language_loss": 0.76648992,
      "learning_rate": 7.045724298891565e-07,
      "loss": 0.788064,
      "num_input_tokens_seen": 130922935,
      "step": 6092,
      "time_per_iteration": 2.798827648162842
    },
    {
      "auxiliary_loss_clip": 0.01129286,
      "auxiliary_loss_mlp": 0.01026433,
      "balance_loss_clip": 1.04455137,
      "balance_loss_mlp": 1.01842177,
      "epoch": 0.7326399326639812,
      "flos": 25775781828480.0,
      "grad_norm": 2.138078520630335,
      "language_loss": 0.69173634,
      "learning_rate": 7.039790423199192e-07,
      "loss": 0.71329349,
      "num_input_tokens_seen": 130942575,
      "step": 6093,
      "time_per_iteration": 2.688420534133911
    },
    {
      "auxiliary_loss_clip": 0.01107543,
      "auxiliary_loss_mlp": 0.01026471,
      "balance_loss_clip": 1.04238522,
      "balance_loss_mlp": 1.01792979,
      "epoch": 0.7327601755546204,
      "flos": 21032269706880.0,
      "grad_norm": 2.5797868750676534,
      "language_loss": 0.77866793,
      "learning_rate": 7.033858513560322e-07,
      "loss": 0.80000806,
      "num_input_tokens_seen": 130958870,
      "step": 6094,
      "time_per_iteration": 2.712519645690918
    },
    {
      "auxiliary_loss_clip": 0.01101864,
      "auxiliary_loss_mlp": 0.01024233,
      "balance_loss_clip": 1.04329228,
      "balance_loss_mlp": 1.01653481,
      "epoch": 0.7328804184452594,
      "flos": 16289224462080.0,
      "grad_norm": 2.8326898681962644,
      "language_loss": 0.76333469,
      "learning_rate": 7.027928570874794e-07,
      "loss": 0.78459561,
      "num_input_tokens_seen": 130977060,
      "step": 6095,
      "time_per_iteration": 2.7243151664733887
    },
    {
      "auxiliary_loss_clip": 0.01169195,
      "auxiliary_loss_mlp": 0.01024907,
      "balance_loss_clip": 1.04696071,
      "balance_loss_mlp": 1.01708126,
      "epoch": 0.7330006613358985,
      "flos": 17858233422720.0,
      "grad_norm": 1.984758868404364,
      "language_loss": 0.85901451,
      "learning_rate": 7.022000596042194e-07,
      "loss": 0.88095546,
      "num_input_tokens_seen": 130994160,
      "step": 6096,
      "time_per_iteration": 2.6078507900238037
    },
    {
      "auxiliary_loss_clip": 0.01122408,
      "auxiliary_loss_mlp": 0.01028481,
      "balance_loss_clip": 1.04316258,
      "balance_loss_mlp": 1.02019584,
      "epoch": 0.7331209042265376,
      "flos": 22492074343680.0,
      "grad_norm": 2.583856420739808,
      "language_loss": 0.82266253,
      "learning_rate": 7.016074589961784e-07,
      "loss": 0.8441714,
      "num_input_tokens_seen": 131012725,
      "step": 6097,
      "time_per_iteration": 2.7029035091400146
    },
    {
      "auxiliary_loss_clip": 0.01136757,
      "auxiliary_loss_mlp": 0.01027685,
      "balance_loss_clip": 1.0427177,
      "balance_loss_mlp": 1.02042556,
      "epoch": 0.7332411471171767,
      "flos": 33072757937280.0,
      "grad_norm": 30.890333859095858,
      "language_loss": 0.67569625,
      "learning_rate": 7.01015055353253e-07,
      "loss": 0.69734073,
      "num_input_tokens_seen": 131035150,
      "step": 6098,
      "time_per_iteration": 2.8401973247528076
    },
    {
      "auxiliary_loss_clip": 0.01130011,
      "auxiliary_loss_mlp": 0.01029315,
      "balance_loss_clip": 1.04632068,
      "balance_loss_mlp": 1.02088726,
      "epoch": 0.7333613900078157,
      "flos": 22743017735040.0,
      "grad_norm": 1.917715605036885,
      "language_loss": 0.7828722,
      "learning_rate": 7.004228487653123e-07,
      "loss": 0.80446541,
      "num_input_tokens_seen": 131055955,
      "step": 6099,
      "time_per_iteration": 2.7090423107147217
    },
    {
      "auxiliary_loss_clip": 0.01146378,
      "auxiliary_loss_mlp": 0.00762011,
      "balance_loss_clip": 1.04549098,
      "balance_loss_mlp": 1.00039458,
      "epoch": 0.7334816328984549,
      "flos": 22346133384960.0,
      "grad_norm": 1.9446216493640966,
      "language_loss": 0.78231573,
      "learning_rate": 6.998308393221906e-07,
      "loss": 0.80139965,
      "num_input_tokens_seen": 131074360,
      "step": 6100,
      "time_per_iteration": 2.7102935314178467
    },
    {
      "auxiliary_loss_clip": 0.01119432,
      "auxiliary_loss_mlp": 0.01022741,
      "balance_loss_clip": 1.04305696,
      "balance_loss_mlp": 1.01439667,
      "epoch": 0.733601875789094,
      "flos": 20736149984640.0,
      "grad_norm": 2.8424091063056394,
      "language_loss": 0.71571791,
      "learning_rate": 6.992390271136977e-07,
      "loss": 0.73713964,
      "num_input_tokens_seen": 131090070,
      "step": 6101,
      "time_per_iteration": 2.679774522781372
    },
    {
      "auxiliary_loss_clip": 0.0114379,
      "auxiliary_loss_mlp": 0.0102099,
      "balance_loss_clip": 1.04572821,
      "balance_loss_mlp": 1.0138998,
      "epoch": 0.733722118679733,
      "flos": 22564362464640.0,
      "grad_norm": 3.864284708394104,
      "language_loss": 0.85609221,
      "learning_rate": 6.986474122296094e-07,
      "loss": 0.87774003,
      "num_input_tokens_seen": 131109185,
      "step": 6102,
      "time_per_iteration": 2.671635866165161
    },
    {
      "auxiliary_loss_clip": 0.01145413,
      "auxiliary_loss_mlp": 0.01026533,
      "balance_loss_clip": 1.0488677,
      "balance_loss_mlp": 1.01819444,
      "epoch": 0.7338423615703722,
      "flos": 20084192179200.0,
      "grad_norm": 2.207974291331288,
      "language_loss": 0.72538805,
      "learning_rate": 6.980559947596751e-07,
      "loss": 0.74710751,
      "num_input_tokens_seen": 131127725,
      "step": 6103,
      "time_per_iteration": 2.7635998725891113
    },
    {
      "auxiliary_loss_clip": 0.01091251,
      "auxiliary_loss_mlp": 0.01026726,
      "balance_loss_clip": 1.04111338,
      "balance_loss_mlp": 1.01903689,
      "epoch": 0.7339626044610112,
      "flos": 21687675217920.0,
      "grad_norm": 2.3203109239366526,
      "language_loss": 0.76097494,
      "learning_rate": 6.974647747936109e-07,
      "loss": 0.78215468,
      "num_input_tokens_seen": 131146110,
      "step": 6104,
      "time_per_iteration": 3.747626304626465
    },
    {
      "auxiliary_loss_clip": 0.01101893,
      "auxiliary_loss_mlp": 0.01025553,
      "balance_loss_clip": 1.04344797,
      "balance_loss_mlp": 1.0182668,
      "epoch": 0.7340828473516503,
      "flos": 15268248282240.0,
      "grad_norm": 2.003799735967241,
      "language_loss": 0.82522464,
      "learning_rate": 6.968737524211039e-07,
      "loss": 0.8464992,
      "num_input_tokens_seen": 131162920,
      "step": 6105,
      "time_per_iteration": 2.773092746734619
    },
    {
      "auxiliary_loss_clip": 0.01132953,
      "auxiliary_loss_mlp": 0.00762555,
      "balance_loss_clip": 1.04486322,
      "balance_loss_mlp": 1.00042284,
      "epoch": 0.7342030902422895,
      "flos": 22930112701440.0,
      "grad_norm": 2.100730364209436,
      "language_loss": 0.80141091,
      "learning_rate": 6.962829277318132e-07,
      "loss": 0.82036602,
      "num_input_tokens_seen": 131182515,
      "step": 6106,
      "time_per_iteration": 2.7884409427642822
    },
    {
      "auxiliary_loss_clip": 0.01117682,
      "auxiliary_loss_mlp": 0.01027962,
      "balance_loss_clip": 1.04702985,
      "balance_loss_mlp": 1.02045512,
      "epoch": 0.7343233331329285,
      "flos": 25847890381440.0,
      "grad_norm": 1.9740561683973052,
      "language_loss": 0.8408947,
      "learning_rate": 6.956923008153652e-07,
      "loss": 0.86235118,
      "num_input_tokens_seen": 131202280,
      "step": 6107,
      "time_per_iteration": 2.7975220680236816
    },
    {
      "auxiliary_loss_clip": 0.01117568,
      "auxiliary_loss_mlp": 0.00761525,
      "balance_loss_clip": 1.03988791,
      "balance_loss_mlp": 1.00034595,
      "epoch": 0.7344435760235676,
      "flos": 18478985287680.0,
      "grad_norm": 2.2129094081323135,
      "language_loss": 0.83737397,
      "learning_rate": 6.951018717613593e-07,
      "loss": 0.85616487,
      "num_input_tokens_seen": 131221295,
      "step": 6108,
      "time_per_iteration": 2.7541661262512207
    },
    {
      "auxiliary_loss_clip": 0.01101479,
      "auxiliary_loss_mlp": 0.01027576,
      "balance_loss_clip": 1.04029655,
      "balance_loss_mlp": 1.02004218,
      "epoch": 0.7345638189142067,
      "flos": 17640040256640.0,
      "grad_norm": 1.9733168981462734,
      "language_loss": 0.78599524,
      "learning_rate": 6.945116406593614e-07,
      "loss": 0.80728579,
      "num_input_tokens_seen": 131240150,
      "step": 6109,
      "time_per_iteration": 2.74922776222229
    },
    {
      "auxiliary_loss_clip": 0.01147937,
      "auxiliary_loss_mlp": 0.01023818,
      "balance_loss_clip": 1.04628277,
      "balance_loss_mlp": 1.01568758,
      "epoch": 0.7346840618048458,
      "flos": 20260225756800.0,
      "grad_norm": 2.8619033433024064,
      "language_loss": 0.74397802,
      "learning_rate": 6.939216075989089e-07,
      "loss": 0.76569557,
      "num_input_tokens_seen": 131258080,
      "step": 6110,
      "time_per_iteration": 2.7471375465393066
    },
    {
      "auxiliary_loss_clip": 0.01118671,
      "auxiliary_loss_mlp": 0.01024094,
      "balance_loss_clip": 1.04299021,
      "balance_loss_mlp": 1.01640463,
      "epoch": 0.7348043046954849,
      "flos": 29023183641600.0,
      "grad_norm": 1.7816720035135278,
      "language_loss": 0.66248775,
      "learning_rate": 6.933317726695109e-07,
      "loss": 0.68391538,
      "num_input_tokens_seen": 131279310,
      "step": 6111,
      "time_per_iteration": 2.7961294651031494
    },
    {
      "auxiliary_loss_clip": 0.01141732,
      "auxiliary_loss_mlp": 0.010243,
      "balance_loss_clip": 1.04633749,
      "balance_loss_mlp": 1.01697719,
      "epoch": 0.734924547586124,
      "flos": 17931203902080.0,
      "grad_norm": 3.670649684456973,
      "language_loss": 0.79847825,
      "learning_rate": 6.92742135960644e-07,
      "loss": 0.82013857,
      "num_input_tokens_seen": 131297010,
      "step": 6112,
      "time_per_iteration": 3.55216121673584
    },
    {
      "auxiliary_loss_clip": 0.01051665,
      "auxiliary_loss_mlp": 0.01001894,
      "balance_loss_clip": 1.03608775,
      "balance_loss_mlp": 1.00077915,
      "epoch": 0.7350447904767631,
      "flos": 63588319850880.0,
      "grad_norm": 1.5853861611856406,
      "language_loss": 0.55631518,
      "learning_rate": 6.921526975617556e-07,
      "loss": 0.57685077,
      "num_input_tokens_seen": 131356470,
      "step": 6113,
      "time_per_iteration": 4.1499762535095215
    },
    {
      "auxiliary_loss_clip": 0.01138721,
      "auxiliary_loss_mlp": 0.01023482,
      "balance_loss_clip": 1.04385138,
      "balance_loss_mlp": 1.01597798,
      "epoch": 0.7351650333674021,
      "flos": 21580015178880.0,
      "grad_norm": 1.846558911631822,
      "language_loss": 0.75469613,
      "learning_rate": 6.915634575622631e-07,
      "loss": 0.77631813,
      "num_input_tokens_seen": 131374985,
      "step": 6114,
      "time_per_iteration": 2.6687378883361816
    },
    {
      "auxiliary_loss_clip": 0.01139568,
      "auxiliary_loss_mlp": 0.0102566,
      "balance_loss_clip": 1.04675484,
      "balance_loss_mlp": 1.01771426,
      "epoch": 0.7352852762580413,
      "flos": 18186349184640.0,
      "grad_norm": 11.696193373826949,
      "language_loss": 0.71104795,
      "learning_rate": 6.909744160515532e-07,
      "loss": 0.73270023,
      "num_input_tokens_seen": 131393125,
      "step": 6115,
      "time_per_iteration": 3.5338871479034424
    },
    {
      "auxiliary_loss_clip": 0.01130993,
      "auxiliary_loss_mlp": 0.01019747,
      "balance_loss_clip": 1.04332376,
      "balance_loss_mlp": 1.01151848,
      "epoch": 0.7354055191486804,
      "flos": 38910073063680.0,
      "grad_norm": 1.873666128253589,
      "language_loss": 0.69715691,
      "learning_rate": 6.903855731189849e-07,
      "loss": 0.71866435,
      "num_input_tokens_seen": 131415760,
      "step": 6116,
      "time_per_iteration": 2.837110996246338
    },
    {
      "auxiliary_loss_clip": 0.01145278,
      "auxiliary_loss_mlp": 0.01030921,
      "balance_loss_clip": 1.05186367,
      "balance_loss_mlp": 1.02205801,
      "epoch": 0.7355257620393194,
      "flos": 16289978647680.0,
      "grad_norm": 2.4067746589701366,
      "language_loss": 0.81711304,
      "learning_rate": 6.897969288538825e-07,
      "loss": 0.83887506,
      "num_input_tokens_seen": 131433705,
      "step": 6117,
      "time_per_iteration": 2.6885316371917725
    },
    {
      "auxiliary_loss_clip": 0.01131611,
      "auxiliary_loss_mlp": 0.01018655,
      "balance_loss_clip": 1.0414412,
      "balance_loss_mlp": 1.01118088,
      "epoch": 0.7356460049299585,
      "flos": 18114240631680.0,
      "grad_norm": 1.8329350710856809,
      "language_loss": 0.8140412,
      "learning_rate": 6.892084833455452e-07,
      "loss": 0.83554387,
      "num_input_tokens_seen": 131453275,
      "step": 6118,
      "time_per_iteration": 2.855443239212036
    },
    {
      "auxiliary_loss_clip": 0.01119865,
      "auxiliary_loss_mlp": 0.01027973,
      "balance_loss_clip": 1.0444324,
      "balance_loss_mlp": 1.02046609,
      "epoch": 0.7357662478205976,
      "flos": 21325193118720.0,
      "grad_norm": 1.5350780906606105,
      "language_loss": 0.83764815,
      "learning_rate": 6.886202366832384e-07,
      "loss": 0.85912651,
      "num_input_tokens_seen": 131474960,
      "step": 6119,
      "time_per_iteration": 2.7947239875793457
    },
    {
      "auxiliary_loss_clip": 0.01118331,
      "auxiliary_loss_mlp": 0.01028453,
      "balance_loss_clip": 1.04326653,
      "balance_loss_mlp": 1.0203228,
      "epoch": 0.7358864907112367,
      "flos": 14246841139200.0,
      "grad_norm": 2.3169699942708872,
      "language_loss": 0.7403487,
      "learning_rate": 6.880321889561987e-07,
      "loss": 0.7618165,
      "num_input_tokens_seen": 131492935,
      "step": 6120,
      "time_per_iteration": 2.8030641078948975
    },
    {
      "auxiliary_loss_clip": 0.01096579,
      "auxiliary_loss_mlp": 0.01025116,
      "balance_loss_clip": 1.03840518,
      "balance_loss_mlp": 1.01697981,
      "epoch": 0.7360067336018757,
      "flos": 22309684058880.0,
      "grad_norm": 2.4397994940286547,
      "language_loss": 0.65404248,
      "learning_rate": 6.874443402536338e-07,
      "loss": 0.67525935,
      "num_input_tokens_seen": 131512025,
      "step": 6121,
      "time_per_iteration": 2.8395659923553467
    },
    {
      "auxiliary_loss_clip": 0.01118633,
      "auxiliary_loss_mlp": 0.0102558,
      "balance_loss_clip": 1.04214573,
      "balance_loss_mlp": 1.01770651,
      "epoch": 0.7361269764925149,
      "flos": 25554607833600.0,
      "grad_norm": 2.4680028564194156,
      "language_loss": 0.80588537,
      "learning_rate": 6.868566906647177e-07,
      "loss": 0.82732755,
      "num_input_tokens_seen": 131532975,
      "step": 6122,
      "time_per_iteration": 2.7720370292663574
    },
    {
      "auxiliary_loss_clip": 0.01159122,
      "auxiliary_loss_mlp": 0.01028963,
      "balance_loss_clip": 1.05077517,
      "balance_loss_mlp": 1.02122879,
      "epoch": 0.736247219383154,
      "flos": 20376505059840.0,
      "grad_norm": 5.1090832474812835,
      "language_loss": 0.83528101,
      "learning_rate": 6.862692402785984e-07,
      "loss": 0.85716188,
      "num_input_tokens_seen": 131553225,
      "step": 6123,
      "time_per_iteration": 2.6585018634796143
    },
    {
      "auxiliary_loss_clip": 0.0105636,
      "auxiliary_loss_mlp": 0.0100233,
      "balance_loss_clip": 1.02222681,
      "balance_loss_mlp": 1.00076854,
      "epoch": 0.736367462273793,
      "flos": 70339525735680.0,
      "grad_norm": 0.6837577935618885,
      "language_loss": 0.49599856,
      "learning_rate": 6.856819891843899e-07,
      "loss": 0.51658547,
      "num_input_tokens_seen": 131617930,
      "step": 6124,
      "time_per_iteration": 3.3713464736938477
    },
    {
      "auxiliary_loss_clip": 0.01131879,
      "auxiliary_loss_mlp": 0.01033473,
      "balance_loss_clip": 1.04425788,
      "balance_loss_mlp": 1.02540898,
      "epoch": 0.7364877051644322,
      "flos": 22412711243520.0,
      "grad_norm": 4.8411342319881365,
      "language_loss": 0.72082388,
      "learning_rate": 6.8509493747118e-07,
      "loss": 0.74247742,
      "num_input_tokens_seen": 131636740,
      "step": 6125,
      "time_per_iteration": 2.7608630657196045
    },
    {
      "auxiliary_loss_clip": 0.01118675,
      "auxiliary_loss_mlp": 0.01026708,
      "balance_loss_clip": 1.04177773,
      "balance_loss_mlp": 1.01851797,
      "epoch": 0.7366079480550712,
      "flos": 12130266274560.0,
      "grad_norm": 2.2392384323775847,
      "language_loss": 0.88188308,
      "learning_rate": 6.845080852280221e-07,
      "loss": 0.90333688,
      "num_input_tokens_seen": 131653810,
      "step": 6126,
      "time_per_iteration": 2.705085277557373
    },
    {
      "auxiliary_loss_clip": 0.01095799,
      "auxiliary_loss_mlp": 0.00761512,
      "balance_loss_clip": 1.0386405,
      "balance_loss_mlp": 1.00046837,
      "epoch": 0.7367281909457103,
      "flos": 15049336844160.0,
      "grad_norm": 1.9240093938188605,
      "language_loss": 0.74211836,
      "learning_rate": 6.839214325439409e-07,
      "loss": 0.76069146,
      "num_input_tokens_seen": 131671505,
      "step": 6127,
      "time_per_iteration": 2.8049843311309814
    },
    {
      "auxiliary_loss_clip": 0.01156831,
      "auxiliary_loss_mlp": 0.0102521,
      "balance_loss_clip": 1.04843628,
      "balance_loss_mlp": 1.01785517,
      "epoch": 0.7368484338363495,
      "flos": 23510752053120.0,
      "grad_norm": 3.2393570604735915,
      "language_loss": 0.71843779,
      "learning_rate": 6.833349795079327e-07,
      "loss": 0.74025822,
      "num_input_tokens_seen": 131690615,
      "step": 6128,
      "time_per_iteration": 2.668116807937622
    },
    {
      "auxiliary_loss_clip": 0.01085721,
      "auxiliary_loss_mlp": 0.01028976,
      "balance_loss_clip": 1.04294324,
      "balance_loss_mlp": 1.02061939,
      "epoch": 0.7369686767269885,
      "flos": 27417833095680.0,
      "grad_norm": 2.05170692003616,
      "language_loss": 0.68641555,
      "learning_rate": 6.827487262089613e-07,
      "loss": 0.70756245,
      "num_input_tokens_seen": 131711120,
      "step": 6129,
      "time_per_iteration": 3.8482823371887207
    },
    {
      "auxiliary_loss_clip": 0.01052537,
      "auxiliary_loss_mlp": 0.01002065,
      "balance_loss_clip": 1.02559376,
      "balance_loss_mlp": 1.00020564,
      "epoch": 0.7370889196176276,
      "flos": 70293343824000.0,
      "grad_norm": 1.0329190280795493,
      "language_loss": 0.56804663,
      "learning_rate": 6.821626727359606e-07,
      "loss": 0.58859259,
      "num_input_tokens_seen": 131776680,
      "step": 6130,
      "time_per_iteration": 3.3542566299438477
    },
    {
      "auxiliary_loss_clip": 0.01100163,
      "auxiliary_loss_mlp": 0.01023234,
      "balance_loss_clip": 1.04068851,
      "balance_loss_mlp": 1.01473451,
      "epoch": 0.7372091625082667,
      "flos": 18040839189120.0,
      "grad_norm": 3.3610062558500586,
      "language_loss": 0.77388227,
      "learning_rate": 6.815768191778348e-07,
      "loss": 0.79511619,
      "num_input_tokens_seen": 131794760,
      "step": 6131,
      "time_per_iteration": 2.7204692363739014
    },
    {
      "auxiliary_loss_clip": 0.0110919,
      "auxiliary_loss_mlp": 0.01029018,
      "balance_loss_clip": 1.03965545,
      "balance_loss_mlp": 1.02061367,
      "epoch": 0.7373294053989058,
      "flos": 33726331854720.0,
      "grad_norm": 1.8777349474853389,
      "language_loss": 0.73060054,
      "learning_rate": 6.809911656234569e-07,
      "loss": 0.75198263,
      "num_input_tokens_seen": 131816735,
      "step": 6132,
      "time_per_iteration": 2.9463934898376465
    },
    {
      "auxiliary_loss_clip": 0.01126247,
      "auxiliary_loss_mlp": 0.01024254,
      "balance_loss_clip": 1.04175782,
      "balance_loss_mlp": 1.01644564,
      "epoch": 0.7374496482895448,
      "flos": 21506326427520.0,
      "grad_norm": 2.4885816289831895,
      "language_loss": 0.78266072,
      "learning_rate": 6.804057121616707e-07,
      "loss": 0.80416578,
      "num_input_tokens_seen": 131834940,
      "step": 6133,
      "time_per_iteration": 2.81569766998291
    },
    {
      "auxiliary_loss_clip": 0.01135962,
      "auxiliary_loss_mlp": 0.01024259,
      "balance_loss_clip": 1.04575753,
      "balance_loss_mlp": 1.01648021,
      "epoch": 0.737569891180184,
      "flos": 24936908624640.0,
      "grad_norm": 5.606441699477932,
      "language_loss": 0.72459263,
      "learning_rate": 6.798204588812888e-07,
      "loss": 0.74619484,
      "num_input_tokens_seen": 131854355,
      "step": 6134,
      "time_per_iteration": 2.7892887592315674
    },
    {
      "auxiliary_loss_clip": 0.01122319,
      "auxiliary_loss_mlp": 0.01024869,
      "balance_loss_clip": 1.04029608,
      "balance_loss_mlp": 1.01686406,
      "epoch": 0.7376901340708231,
      "flos": 20664544222080.0,
      "grad_norm": 2.237084588576956,
      "language_loss": 0.75777221,
      "learning_rate": 6.792354058710937e-07,
      "loss": 0.77924412,
      "num_input_tokens_seen": 131871825,
      "step": 6135,
      "time_per_iteration": 2.7677061557769775
    },
    {
      "auxiliary_loss_clip": 0.0114219,
      "auxiliary_loss_mlp": 0.01023065,
      "balance_loss_clip": 1.0432533,
      "balance_loss_mlp": 1.01593637,
      "epoch": 0.7378103769614621,
      "flos": 23805794367360.0,
      "grad_norm": 2.251108083651031,
      "language_loss": 0.64908212,
      "learning_rate": 6.786505532198374e-07,
      "loss": 0.67073464,
      "num_input_tokens_seen": 131890770,
      "step": 6136,
      "time_per_iteration": 2.716571569442749
    },
    {
      "auxiliary_loss_clip": 0.01138339,
      "auxiliary_loss_mlp": 0.01023045,
      "balance_loss_clip": 1.0459578,
      "balance_loss_mlp": 1.01505172,
      "epoch": 0.7379306198521013,
      "flos": 22237216369920.0,
      "grad_norm": 1.9473724571789859,
      "language_loss": 0.85364711,
      "learning_rate": 6.780659010162411e-07,
      "loss": 0.87526095,
      "num_input_tokens_seen": 131909720,
      "step": 6137,
      "time_per_iteration": 2.7203707695007324
    },
    {
      "auxiliary_loss_clip": 0.01079657,
      "auxiliary_loss_mlp": 0.0102344,
      "balance_loss_clip": 1.03901672,
      "balance_loss_mlp": 1.01599503,
      "epoch": 0.7380508627427403,
      "flos": 14903108576640.0,
      "grad_norm": 1.9298688633065875,
      "language_loss": 0.83303487,
      "learning_rate": 6.774814493489975e-07,
      "loss": 0.85406584,
      "num_input_tokens_seen": 131927395,
      "step": 6138,
      "time_per_iteration": 3.656581163406372
    },
    {
      "auxiliary_loss_clip": 0.01085283,
      "auxiliary_loss_mlp": 0.01023645,
      "balance_loss_clip": 1.04004812,
      "balance_loss_mlp": 1.01628995,
      "epoch": 0.7381711056333794,
      "flos": 21685843624320.0,
      "grad_norm": 1.8986737216670369,
      "language_loss": 0.66095412,
      "learning_rate": 6.768971983067655e-07,
      "loss": 0.68204343,
      "num_input_tokens_seen": 131947725,
      "step": 6139,
      "time_per_iteration": 3.7473952770233154
    },
    {
      "auxiliary_loss_clip": 0.01049304,
      "auxiliary_loss_mlp": 0.01001769,
      "balance_loss_clip": 1.01337087,
      "balance_loss_mlp": 1.00066078,
      "epoch": 0.7382913485240186,
      "flos": 52404263596800.0,
      "grad_norm": 1.0123054578277788,
      "language_loss": 0.67707026,
      "learning_rate": 6.763131479781772e-07,
      "loss": 0.69758093,
      "num_input_tokens_seen": 131997485,
      "step": 6140,
      "time_per_iteration": 3.901002883911133
    },
    {
      "auxiliary_loss_clip": 0.01110798,
      "auxiliary_loss_mlp": 0.01026047,
      "balance_loss_clip": 1.04404771,
      "balance_loss_mlp": 1.01824176,
      "epoch": 0.7384115914146576,
      "flos": 21798818876160.0,
      "grad_norm": 1.7829262663108056,
      "language_loss": 0.7606883,
      "learning_rate": 6.757292984518316e-07,
      "loss": 0.78205681,
      "num_input_tokens_seen": 132016885,
      "step": 6141,
      "time_per_iteration": 2.720144748687744
    },
    {
      "auxiliary_loss_clip": 0.010314,
      "auxiliary_loss_mlp": 0.01001468,
      "balance_loss_clip": 1.03396702,
      "balance_loss_mlp": 1.00041866,
      "epoch": 0.7385318343052967,
      "flos": 61494331662720.0,
      "grad_norm": 0.7465674927909343,
      "language_loss": 0.56355214,
      "learning_rate": 6.751456498162981e-07,
      "loss": 0.58388078,
      "num_input_tokens_seen": 132075920,
      "step": 6142,
      "time_per_iteration": 3.305420398712158
    },
    {
      "auxiliary_loss_clip": 0.01167463,
      "auxiliary_loss_mlp": 0.01025742,
      "balance_loss_clip": 1.04564941,
      "balance_loss_mlp": 1.01844668,
      "epoch": 0.7386520771959358,
      "flos": 17013757697280.0,
      "grad_norm": 2.301976888333584,
      "language_loss": 0.85760248,
      "learning_rate": 6.745622021601174e-07,
      "loss": 0.87953448,
      "num_input_tokens_seen": 132092945,
      "step": 6143,
      "time_per_iteration": 2.6255412101745605
    },
    {
      "auxiliary_loss_clip": 0.01127958,
      "auxiliary_loss_mlp": 0.0102524,
      "balance_loss_clip": 1.04505455,
      "balance_loss_mlp": 1.01785803,
      "epoch": 0.7387723200865749,
      "flos": 18770759464320.0,
      "grad_norm": 2.2856924409240102,
      "language_loss": 0.69954634,
      "learning_rate": 6.739789555717954e-07,
      "loss": 0.72107834,
      "num_input_tokens_seen": 132109920,
      "step": 6144,
      "time_per_iteration": 2.6837756633758545
    },
    {
      "auxiliary_loss_clip": 0.01122799,
      "auxiliary_loss_mlp": 0.01024164,
      "balance_loss_clip": 1.04162776,
      "balance_loss_mlp": 1.01660061,
      "epoch": 0.738892562977214,
      "flos": 22525542840960.0,
      "grad_norm": 1.9375489847697394,
      "language_loss": 0.77674001,
      "learning_rate": 6.733959101398124e-07,
      "loss": 0.79820967,
      "num_input_tokens_seen": 132128050,
      "step": 6145,
      "time_per_iteration": 2.7996435165405273
    },
    {
      "auxiliary_loss_clip": 0.01088653,
      "auxiliary_loss_mlp": 0.01020342,
      "balance_loss_clip": 1.04008222,
      "balance_loss_mlp": 1.01317728,
      "epoch": 0.7390128058678531,
      "flos": 21501478091520.0,
      "grad_norm": 1.64527152463566,
      "language_loss": 0.81556094,
      "learning_rate": 6.728130659526143e-07,
      "loss": 0.83665091,
      "num_input_tokens_seen": 132145860,
      "step": 6146,
      "time_per_iteration": 2.7540194988250732
    },
    {
      "auxiliary_loss_clip": 0.01145369,
      "auxiliary_loss_mlp": 0.00762615,
      "balance_loss_clip": 1.04216099,
      "balance_loss_mlp": 1.00034153,
      "epoch": 0.7391330487584922,
      "flos": 25776176878080.0,
      "grad_norm": 2.292952546005351,
      "language_loss": 0.71256196,
      "learning_rate": 6.7223042309862e-07,
      "loss": 0.73164177,
      "num_input_tokens_seen": 132166060,
      "step": 6147,
      "time_per_iteration": 2.7756972312927246
    },
    {
      "auxiliary_loss_clip": 0.01142346,
      "auxiliary_loss_mlp": 0.01022266,
      "balance_loss_clip": 1.04479969,
      "balance_loss_mlp": 1.01476502,
      "epoch": 0.7392532916491312,
      "flos": 28366736636160.0,
      "grad_norm": 1.9980687997433848,
      "language_loss": 0.73367321,
      "learning_rate": 6.716479816662144e-07,
      "loss": 0.75531942,
      "num_input_tokens_seen": 132187790,
      "step": 6148,
      "time_per_iteration": 2.7691636085510254
    },
    {
      "auxiliary_loss_clip": 0.01121833,
      "auxiliary_loss_mlp": 0.01025502,
      "balance_loss_clip": 1.04065013,
      "balance_loss_mlp": 1.01762187,
      "epoch": 0.7393735345397703,
      "flos": 23585877348480.0,
      "grad_norm": 1.957811190470067,
      "language_loss": 0.73169255,
      "learning_rate": 6.710657417437531e-07,
      "loss": 0.75316596,
      "num_input_tokens_seen": 132207495,
      "step": 6149,
      "time_per_iteration": 2.7502496242523193
    },
    {
      "auxiliary_loss_clip": 0.01159539,
      "auxiliary_loss_mlp": 0.01029748,
      "balance_loss_clip": 1.04630554,
      "balance_loss_mlp": 1.02210045,
      "epoch": 0.7394937774304094,
      "flos": 19974772373760.0,
      "grad_norm": 2.188634272972611,
      "language_loss": 0.79994851,
      "learning_rate": 6.704837034195628e-07,
      "loss": 0.82184136,
      "num_input_tokens_seen": 132225960,
      "step": 6150,
      "time_per_iteration": 2.60860538482666
    },
    {
      "auxiliary_loss_clip": 0.0106792,
      "auxiliary_loss_mlp": 0.0102932,
      "balance_loss_clip": 1.03736854,
      "balance_loss_mlp": 1.02132702,
      "epoch": 0.7396140203210485,
      "flos": 23478037741440.0,
      "grad_norm": 2.3646481340923797,
      "language_loss": 0.85201979,
      "learning_rate": 6.699018667819376e-07,
      "loss": 0.87299216,
      "num_input_tokens_seen": 132245360,
      "step": 6151,
      "time_per_iteration": 2.8657937049865723
    },
    {
      "auxiliary_loss_clip": 0.01082608,
      "auxiliary_loss_mlp": 0.010231,
      "balance_loss_clip": 1.04029059,
      "balance_loss_mlp": 1.014624,
      "epoch": 0.7397342632116876,
      "flos": 25555433846400.0,
      "grad_norm": 1.610154773420632,
      "language_loss": 0.73207831,
      "learning_rate": 6.693202319191415e-07,
      "loss": 0.75313544,
      "num_input_tokens_seen": 132267095,
      "step": 6152,
      "time_per_iteration": 2.871896505355835
    },
    {
      "auxiliary_loss_clip": 0.0110734,
      "auxiliary_loss_mlp": 0.01031028,
      "balance_loss_clip": 1.04329181,
      "balance_loss_mlp": 1.02304673,
      "epoch": 0.7398545061023267,
      "flos": 24755021130240.0,
      "grad_norm": 3.996355462608115,
      "language_loss": 0.74988079,
      "learning_rate": 6.687387989194084e-07,
      "loss": 0.77126443,
      "num_input_tokens_seen": 132286610,
      "step": 6153,
      "time_per_iteration": 2.7552666664123535
    },
    {
      "auxiliary_loss_clip": 0.01138915,
      "auxiliary_loss_mlp": 0.01023798,
      "balance_loss_clip": 1.04397178,
      "balance_loss_mlp": 1.01612043,
      "epoch": 0.7399747489929658,
      "flos": 16508602776960.0,
      "grad_norm": 1.8529665103606354,
      "language_loss": 0.79547703,
      "learning_rate": 6.681575678709404e-07,
      "loss": 0.81710416,
      "num_input_tokens_seen": 132305300,
      "step": 6154,
      "time_per_iteration": 2.7054882049560547
    },
    {
      "auxiliary_loss_clip": 0.01118179,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.04250669,
      "balance_loss_mlp": 1.01960194,
      "epoch": 0.7400949918836048,
      "flos": 24097065753600.0,
      "grad_norm": 1.8429439474882992,
      "language_loss": 0.70683289,
      "learning_rate": 6.67576538861911e-07,
      "loss": 0.72828531,
      "num_input_tokens_seen": 132323875,
      "step": 6155,
      "time_per_iteration": 3.6710920333862305
    },
    {
      "auxiliary_loss_clip": 0.01086543,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.04035115,
      "balance_loss_mlp": 1.01752543,
      "epoch": 0.740215234774244,
      "flos": 21802517976960.0,
      "grad_norm": 1.7879955971714927,
      "language_loss": 0.82237381,
      "learning_rate": 6.669957119804612e-07,
      "loss": 0.84349072,
      "num_input_tokens_seen": 132345510,
      "step": 6156,
      "time_per_iteration": 2.9242210388183594
    },
    {
      "auxiliary_loss_clip": 0.01143613,
      "auxiliary_loss_mlp": 0.01024364,
      "balance_loss_clip": 1.04389095,
      "balance_loss_mlp": 1.01583767,
      "epoch": 0.7403354776648831,
      "flos": 18733196816640.0,
      "grad_norm": 3.4196158941979635,
      "language_loss": 0.73066431,
      "learning_rate": 6.66415087314702e-07,
      "loss": 0.75234401,
      "num_input_tokens_seen": 132360465,
      "step": 6157,
      "time_per_iteration": 2.696746826171875
    },
    {
      "auxiliary_loss_clip": 0.0113951,
      "auxiliary_loss_mlp": 0.01023836,
      "balance_loss_clip": 1.04510593,
      "balance_loss_mlp": 1.01579821,
      "epoch": 0.7404557205555221,
      "flos": 16909581277440.0,
      "grad_norm": 2.1127541179534552,
      "language_loss": 0.7272976,
      "learning_rate": 6.65834664952714e-07,
      "loss": 0.74893105,
      "num_input_tokens_seen": 132377915,
      "step": 6158,
      "time_per_iteration": 2.659782886505127
    },
    {
      "auxiliary_loss_clip": 0.01091228,
      "auxiliary_loss_mlp": 0.01021282,
      "balance_loss_clip": 1.04103029,
      "balance_loss_mlp": 1.01351511,
      "epoch": 0.7405759634461613,
      "flos": 21214408596480.0,
      "grad_norm": 1.8552834868857224,
      "language_loss": 0.76000369,
      "learning_rate": 6.652544449825457e-07,
      "loss": 0.78112882,
      "num_input_tokens_seen": 132398170,
      "step": 6159,
      "time_per_iteration": 2.842963933944702
    },
    {
      "auxiliary_loss_clip": 0.01133897,
      "auxiliary_loss_mlp": 0.01029759,
      "balance_loss_clip": 1.04468107,
      "balance_loss_mlp": 1.02133703,
      "epoch": 0.7406962063368003,
      "flos": 20480106862080.0,
      "grad_norm": 7.732343258383676,
      "language_loss": 0.76724029,
      "learning_rate": 6.646744274922182e-07,
      "loss": 0.78887683,
      "num_input_tokens_seen": 132416615,
      "step": 6160,
      "time_per_iteration": 2.6948182582855225
    },
    {
      "auxiliary_loss_clip": 0.01116177,
      "auxiliary_loss_mlp": 0.01025478,
      "balance_loss_clip": 1.03984261,
      "balance_loss_mlp": 1.01806951,
      "epoch": 0.7408164492274394,
      "flos": 19791915212160.0,
      "grad_norm": 3.203079961470167,
      "language_loss": 0.75617409,
      "learning_rate": 6.640946125697171e-07,
      "loss": 0.77759063,
      "num_input_tokens_seen": 132434145,
      "step": 6161,
      "time_per_iteration": 2.779266119003296
    },
    {
      "auxiliary_loss_clip": 0.01149386,
      "auxiliary_loss_mlp": 0.01027175,
      "balance_loss_clip": 1.04496992,
      "balance_loss_mlp": 1.01878273,
      "epoch": 0.7409366921180786,
      "flos": 29204855654400.0,
      "grad_norm": 3.8713434227954027,
      "language_loss": 0.75774002,
      "learning_rate": 6.635150003030017e-07,
      "loss": 0.77950561,
      "num_input_tokens_seen": 132452670,
      "step": 6162,
      "time_per_iteration": 2.7314722537994385
    },
    {
      "auxiliary_loss_clip": 0.01111091,
      "auxiliary_loss_mlp": 0.00761992,
      "balance_loss_clip": 1.0407784,
      "balance_loss_mlp": 1.00049257,
      "epoch": 0.7410569350087176,
      "flos": 22930004960640.0,
      "grad_norm": 2.249395091053292,
      "language_loss": 0.85867631,
      "learning_rate": 6.629355907799981e-07,
      "loss": 0.87740719,
      "num_input_tokens_seen": 132472475,
      "step": 6163,
      "time_per_iteration": 3.6651535034179688
    },
    {
      "auxiliary_loss_clip": 0.01111213,
      "auxiliary_loss_mlp": 0.01027459,
      "balance_loss_clip": 1.04058766,
      "balance_loss_mlp": 1.01963878,
      "epoch": 0.7411771778993567,
      "flos": 30440397726720.0,
      "grad_norm": 1.6662830716435624,
      "language_loss": 0.69107801,
      "learning_rate": 6.623563840886015e-07,
      "loss": 0.71246469,
      "num_input_tokens_seen": 132493400,
      "step": 6164,
      "time_per_iteration": 2.8790481090545654
    },
    {
      "auxiliary_loss_clip": 0.01140996,
      "auxiliary_loss_mlp": 0.0102213,
      "balance_loss_clip": 1.0444963,
      "balance_loss_mlp": 1.01500964,
      "epoch": 0.7412974207899958,
      "flos": 20522050968960.0,
      "grad_norm": 1.8414862912077774,
      "language_loss": 0.69966877,
      "learning_rate": 6.617773803166795e-07,
      "loss": 0.72130001,
      "num_input_tokens_seen": 132511725,
      "step": 6165,
      "time_per_iteration": 4.481919050216675
    },
    {
      "auxiliary_loss_clip": 0.01173568,
      "auxiliary_loss_mlp": 0.01026017,
      "balance_loss_clip": 1.04831839,
      "balance_loss_mlp": 1.01726162,
      "epoch": 0.7414176636806349,
      "flos": 22090700793600.0,
      "grad_norm": 2.5170921568574727,
      "language_loss": 0.82291615,
      "learning_rate": 6.611985795520634e-07,
      "loss": 0.84491199,
      "num_input_tokens_seen": 132530270,
      "step": 6166,
      "time_per_iteration": 2.689535617828369
    },
    {
      "auxiliary_loss_clip": 0.01102046,
      "auxiliary_loss_mlp": 0.01026835,
      "balance_loss_clip": 1.0401175,
      "balance_loss_mlp": 1.01824629,
      "epoch": 0.7415379065712739,
      "flos": 25155245445120.0,
      "grad_norm": 2.322670119719809,
      "language_loss": 0.77775931,
      "learning_rate": 6.606199818825588e-07,
      "loss": 0.79904807,
      "num_input_tokens_seen": 132550725,
      "step": 6167,
      "time_per_iteration": 2.865100860595703
    },
    {
      "auxiliary_loss_clip": 0.0113333,
      "auxiliary_loss_mlp": 0.01023533,
      "balance_loss_clip": 1.04127443,
      "balance_loss_mlp": 1.01642191,
      "epoch": 0.7416581494619131,
      "flos": 16871731320960.0,
      "grad_norm": 4.821907592993478,
      "language_loss": 0.82222235,
      "learning_rate": 6.600415873959377e-07,
      "loss": 0.84379095,
      "num_input_tokens_seen": 132568600,
      "step": 6168,
      "time_per_iteration": 2.677199602127075
    },
    {
      "auxiliary_loss_clip": 0.01109687,
      "auxiliary_loss_mlp": 0.01021081,
      "balance_loss_clip": 1.0394789,
      "balance_loss_mlp": 1.01367211,
      "epoch": 0.7417783923525522,
      "flos": 28438881102720.0,
      "grad_norm": 2.1471928090864276,
      "language_loss": 0.64813411,
      "learning_rate": 6.594633961799437e-07,
      "loss": 0.66944176,
      "num_input_tokens_seen": 132587640,
      "step": 6169,
      "time_per_iteration": 2.8511974811553955
    },
    {
      "auxiliary_loss_clip": 0.01104241,
      "auxiliary_loss_mlp": 0.01032904,
      "balance_loss_clip": 1.03842926,
      "balance_loss_mlp": 1.02486348,
      "epoch": 0.7418986352431912,
      "flos": 20084299920000.0,
      "grad_norm": 3.674272509424072,
      "language_loss": 0.81570601,
      "learning_rate": 6.588854083222857e-07,
      "loss": 0.8370775,
      "num_input_tokens_seen": 132607075,
      "step": 6170,
      "time_per_iteration": 2.692763328552246
    },
    {
      "auxiliary_loss_clip": 0.01127664,
      "auxiliary_loss_mlp": 0.01027225,
      "balance_loss_clip": 1.04456103,
      "balance_loss_mlp": 1.0189755,
      "epoch": 0.7420188781338304,
      "flos": 18259571059200.0,
      "grad_norm": 2.008983067786684,
      "language_loss": 0.80573666,
      "learning_rate": 6.583076239106444e-07,
      "loss": 0.82728553,
      "num_input_tokens_seen": 132625580,
      "step": 6171,
      "time_per_iteration": 2.761301040649414
    },
    {
      "auxiliary_loss_clip": 0.01116702,
      "auxiliary_loss_mlp": 0.01027123,
      "balance_loss_clip": 1.0430007,
      "balance_loss_mlp": 1.01903415,
      "epoch": 0.7421391210244694,
      "flos": 13771994319360.0,
      "grad_norm": 2.748430205865753,
      "language_loss": 0.75633496,
      "learning_rate": 6.577300430326707e-07,
      "loss": 0.7777732,
      "num_input_tokens_seen": 132640525,
      "step": 6172,
      "time_per_iteration": 2.7006819248199463
    },
    {
      "auxiliary_loss_clip": 0.01167631,
      "auxiliary_loss_mlp": 0.01024903,
      "balance_loss_clip": 1.04648852,
      "balance_loss_mlp": 1.01711226,
      "epoch": 0.7422593639151085,
      "flos": 15961683317760.0,
      "grad_norm": 2.3088333629687647,
      "language_loss": 0.72239816,
      "learning_rate": 6.571526657759821e-07,
      "loss": 0.74432361,
      "num_input_tokens_seen": 132656265,
      "step": 6173,
      "time_per_iteration": 2.6652657985687256
    },
    {
      "auxiliary_loss_clip": 0.01120776,
      "auxiliary_loss_mlp": 0.01020618,
      "balance_loss_clip": 1.04070199,
      "balance_loss_mlp": 1.01319396,
      "epoch": 0.7423796068057477,
      "flos": 30114400867200.0,
      "grad_norm": 1.9779338113242975,
      "language_loss": 0.71071577,
      "learning_rate": 6.565754922281663e-07,
      "loss": 0.73212969,
      "num_input_tokens_seen": 132678510,
      "step": 6174,
      "time_per_iteration": 2.7831921577453613
    },
    {
      "auxiliary_loss_clip": 0.01133247,
      "auxiliary_loss_mlp": 0.01024657,
      "balance_loss_clip": 1.04197121,
      "balance_loss_mlp": 1.01633668,
      "epoch": 0.7424998496963867,
      "flos": 20521907314560.0,
      "grad_norm": 2.117246307129085,
      "language_loss": 0.78169453,
      "learning_rate": 6.559985224767801e-07,
      "loss": 0.80327356,
      "num_input_tokens_seen": 132696385,
      "step": 6175,
      "time_per_iteration": 2.7517569065093994
    },
    {
      "auxiliary_loss_clip": 0.01104661,
      "auxiliary_loss_mlp": 0.01030125,
      "balance_loss_clip": 1.03889382,
      "balance_loss_mlp": 1.02190852,
      "epoch": 0.7426200925870258,
      "flos": 21871573873920.0,
      "grad_norm": 2.43499855016908,
      "language_loss": 0.75204939,
      "learning_rate": 6.55421756609349e-07,
      "loss": 0.77339721,
      "num_input_tokens_seen": 132714640,
      "step": 6176,
      "time_per_iteration": 2.900099515914917
    },
    {
      "auxiliary_loss_clip": 0.01151127,
      "auxiliary_loss_mlp": 0.00762756,
      "balance_loss_clip": 1.04654908,
      "balance_loss_mlp": 1.0005163,
      "epoch": 0.7427403354776649,
      "flos": 26432049265920.0,
      "grad_norm": 4.07310087962105,
      "language_loss": 0.78807247,
      "learning_rate": 6.54845194713369e-07,
      "loss": 0.80721134,
      "num_input_tokens_seen": 132735590,
      "step": 6177,
      "time_per_iteration": 2.6789262294769287
    },
    {
      "auxiliary_loss_clip": 0.01111546,
      "auxiliary_loss_mlp": 0.01023808,
      "balance_loss_clip": 1.04350114,
      "balance_loss_mlp": 1.01580906,
      "epoch": 0.742860578368304,
      "flos": 19898390102400.0,
      "grad_norm": 2.344887258077172,
      "language_loss": 0.80236548,
      "learning_rate": 6.542688368763034e-07,
      "loss": 0.82371902,
      "num_input_tokens_seen": 132753995,
      "step": 6178,
      "time_per_iteration": 2.7756569385528564
    },
    {
      "auxiliary_loss_clip": 0.01136358,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.04368901,
      "balance_loss_mlp": 1.01711369,
      "epoch": 0.742980821258943,
      "flos": 24827201510400.0,
      "grad_norm": 2.2711132318962424,
      "language_loss": 0.77246451,
      "learning_rate": 6.536926831855854e-07,
      "loss": 0.79407895,
      "num_input_tokens_seen": 132773160,
      "step": 6179,
      "time_per_iteration": 2.7174813747406006
    },
    {
      "auxiliary_loss_clip": 0.01130156,
      "auxiliary_loss_mlp": 0.01025483,
      "balance_loss_clip": 1.0432713,
      "balance_loss_mlp": 1.0184381,
      "epoch": 0.7431010641495821,
      "flos": 25228646887680.0,
      "grad_norm": 2.396281075833185,
      "language_loss": 0.73377317,
      "learning_rate": 6.531167337286165e-07,
      "loss": 0.75532961,
      "num_input_tokens_seen": 132793180,
      "step": 6180,
      "time_per_iteration": 3.690572500228882
    },
    {
      "auxiliary_loss_clip": 0.01078071,
      "auxiliary_loss_mlp": 0.01021922,
      "balance_loss_clip": 1.03761649,
      "balance_loss_mlp": 1.01429224,
      "epoch": 0.7432213070402213,
      "flos": 21762369550080.0,
      "grad_norm": 2.576075882059599,
      "language_loss": 0.79773593,
      "learning_rate": 6.52540988592768e-07,
      "loss": 0.81873584,
      "num_input_tokens_seen": 132814200,
      "step": 6181,
      "time_per_iteration": 3.0057036876678467
    },
    {
      "auxiliary_loss_clip": 0.01140701,
      "auxiliary_loss_mlp": 0.01019923,
      "balance_loss_clip": 1.04503393,
      "balance_loss_mlp": 1.01259756,
      "epoch": 0.7433415499308603,
      "flos": 14793832425600.0,
      "grad_norm": 4.953331479803862,
      "language_loss": 0.83761382,
      "learning_rate": 6.519654478653814e-07,
      "loss": 0.85922003,
      "num_input_tokens_seen": 132832565,
      "step": 6182,
      "time_per_iteration": 2.9611401557922363
    },
    {
      "auxiliary_loss_clip": 0.0105623,
      "auxiliary_loss_mlp": 0.01001329,
      "balance_loss_clip": 1.02345562,
      "balance_loss_mlp": 0.9994573,
      "epoch": 0.7434617928214994,
      "flos": 67155577297920.0,
      "grad_norm": 0.7575686402439353,
      "language_loss": 0.56059164,
      "learning_rate": 6.51390111633763e-07,
      "loss": 0.58116728,
      "num_input_tokens_seen": 132897840,
      "step": 6183,
      "time_per_iteration": 3.3343698978424072
    },
    {
      "auxiliary_loss_clip": 0.0109524,
      "auxiliary_loss_mlp": 0.01026851,
      "balance_loss_clip": 1.0442996,
      "balance_loss_mlp": 1.0186553,
      "epoch": 0.7435820357121385,
      "flos": 27377576928000.0,
      "grad_norm": 1.5724329447131375,
      "language_loss": 0.76328731,
      "learning_rate": 6.508149799851932e-07,
      "loss": 0.78450823,
      "num_input_tokens_seen": 132919505,
      "step": 6184,
      "time_per_iteration": 2.8249361515045166
    },
    {
      "auxiliary_loss_clip": 0.01157738,
      "auxiliary_loss_mlp": 0.01021942,
      "balance_loss_clip": 1.04697669,
      "balance_loss_mlp": 1.0147326,
      "epoch": 0.7437022786027776,
      "flos": 23987645948160.0,
      "grad_norm": 2.260153276238788,
      "language_loss": 0.61139017,
      "learning_rate": 6.502400530069183e-07,
      "loss": 0.63318706,
      "num_input_tokens_seen": 132939390,
      "step": 6185,
      "time_per_iteration": 3.0074615478515625
    },
    {
      "auxiliary_loss_clip": 0.01162942,
      "auxiliary_loss_mlp": 0.01027111,
      "balance_loss_clip": 1.05030692,
      "balance_loss_mlp": 1.01832783,
      "epoch": 0.7438225214934167,
      "flos": 21866761451520.0,
      "grad_norm": 1.9651520428097318,
      "language_loss": 0.68309629,
      "learning_rate": 6.496653307861535e-07,
      "loss": 0.70499682,
      "num_input_tokens_seen": 132960060,
      "step": 6186,
      "time_per_iteration": 2.78716778755188
    },
    {
      "auxiliary_loss_clip": 0.01128036,
      "auxiliary_loss_mlp": 0.00762496,
      "balance_loss_clip": 1.0428406,
      "balance_loss_mlp": 1.00052333,
      "epoch": 0.7439427643840558,
      "flos": 20230097224320.0,
      "grad_norm": 2.4550637975503857,
      "language_loss": 0.65930176,
      "learning_rate": 6.490908134100857e-07,
      "loss": 0.6782071,
      "num_input_tokens_seen": 132978525,
      "step": 6187,
      "time_per_iteration": 2.7486941814422607
    },
    {
      "auxiliary_loss_clip": 0.01153101,
      "auxiliary_loss_mlp": 0.01023445,
      "balance_loss_clip": 1.0478518,
      "balance_loss_mlp": 1.01580405,
      "epoch": 0.7440630072746949,
      "flos": 20849915335680.0,
      "grad_norm": 2.3089756679128266,
      "language_loss": 0.69603616,
      "learning_rate": 6.48516500965866e-07,
      "loss": 0.71780163,
      "num_input_tokens_seen": 132998460,
      "step": 6188,
      "time_per_iteration": 2.727031707763672
    },
    {
      "auxiliary_loss_clip": 0.01157287,
      "auxiliary_loss_mlp": 0.00762025,
      "balance_loss_clip": 1.04381549,
      "balance_loss_mlp": 1.00040603,
      "epoch": 0.7441832501653339,
      "flos": 26503762769280.0,
      "grad_norm": 1.569740865835259,
      "language_loss": 0.81467569,
      "learning_rate": 6.479423935406192e-07,
      "loss": 0.83386874,
      "num_input_tokens_seen": 133018445,
      "step": 6189,
      "time_per_iteration": 3.628443717956543
    },
    {
      "auxiliary_loss_clip": 0.01058498,
      "auxiliary_loss_mlp": 0.01001967,
      "balance_loss_clip": 1.02232099,
      "balance_loss_mlp": 1.00036955,
      "epoch": 0.7443034930559731,
      "flos": 68602848088320.0,
      "grad_norm": 0.8242856300291519,
      "language_loss": 0.62034142,
      "learning_rate": 6.473684912214357e-07,
      "loss": 0.64094603,
      "num_input_tokens_seen": 133082005,
      "step": 6190,
      "time_per_iteration": 4.301899671554565
    },
    {
      "auxiliary_loss_clip": 0.010998,
      "auxiliary_loss_mlp": 0.01028847,
      "balance_loss_clip": 1.04276264,
      "balance_loss_mlp": 1.02001941,
      "epoch": 0.7444237359466122,
      "flos": 18654982951680.0,
      "grad_norm": 2.413482875601263,
      "language_loss": 0.69984132,
      "learning_rate": 6.467947940953778e-07,
      "loss": 0.72112775,
      "num_input_tokens_seen": 133100530,
      "step": 6191,
      "time_per_iteration": 2.802932024002075
    },
    {
      "auxiliary_loss_clip": 0.01113731,
      "auxiliary_loss_mlp": 0.01024204,
      "balance_loss_clip": 1.04111564,
      "balance_loss_mlp": 1.01669931,
      "epoch": 0.7445439788372512,
      "flos": 22817604326400.0,
      "grad_norm": 6.4890215239844675,
      "language_loss": 0.72804385,
      "learning_rate": 6.462213022494732e-07,
      "loss": 0.74942315,
      "num_input_tokens_seen": 133119775,
      "step": 6192,
      "time_per_iteration": 3.638993263244629
    },
    {
      "auxiliary_loss_clip": 0.01071665,
      "auxiliary_loss_mlp": 0.01002024,
      "balance_loss_clip": 1.01837564,
      "balance_loss_mlp": 1.00030708,
      "epoch": 0.7446642217278904,
      "flos": 67045690615680.0,
      "grad_norm": 1.1295436658095528,
      "language_loss": 0.60999024,
      "learning_rate": 6.456480157707201e-07,
      "loss": 0.63072717,
      "num_input_tokens_seen": 133184550,
      "step": 6193,
      "time_per_iteration": 3.1930720806121826
    },
    {
      "auxiliary_loss_clip": 0.01099029,
      "auxiliary_loss_mlp": 0.01026172,
      "balance_loss_clip": 1.03849745,
      "balance_loss_mlp": 1.01862597,
      "epoch": 0.7447844646185294,
      "flos": 17417465631360.0,
      "grad_norm": 2.4577506304582672,
      "language_loss": 0.84802395,
      "learning_rate": 6.450749347460866e-07,
      "loss": 0.86927593,
      "num_input_tokens_seen": 133201525,
      "step": 6194,
      "time_per_iteration": 2.8209798336029053
    },
    {
      "auxiliary_loss_clip": 0.01119957,
      "auxiliary_loss_mlp": 0.01020935,
      "balance_loss_clip": 1.04128575,
      "balance_loss_mlp": 1.01360917,
      "epoch": 0.7449047075091685,
      "flos": 26615876094720.0,
      "grad_norm": 2.118000869202387,
      "language_loss": 0.79062569,
      "learning_rate": 6.445020592625083e-07,
      "loss": 0.81203455,
      "num_input_tokens_seen": 133222175,
      "step": 6195,
      "time_per_iteration": 2.7879233360290527
    },
    {
      "auxiliary_loss_clip": 0.01139951,
      "auxiliary_loss_mlp": 0.0103298,
      "balance_loss_clip": 1.04753399,
      "balance_loss_mlp": 1.02538645,
      "epoch": 0.7450249503998077,
      "flos": 14170458867840.0,
      "grad_norm": 2.7649321440309214,
      "language_loss": 0.80109596,
      "learning_rate": 6.4392938940689e-07,
      "loss": 0.82282531,
      "num_input_tokens_seen": 133237590,
      "step": 6196,
      "time_per_iteration": 2.73224139213562
    },
    {
      "auxiliary_loss_clip": 0.01155595,
      "auxiliary_loss_mlp": 0.01021054,
      "balance_loss_clip": 1.04928529,
      "balance_loss_mlp": 1.01348448,
      "epoch": 0.7451451932904467,
      "flos": 19606687752960.0,
      "grad_norm": 2.579627673082488,
      "language_loss": 0.71141088,
      "learning_rate": 6.433569252661049e-07,
      "loss": 0.73317742,
      "num_input_tokens_seen": 133255590,
      "step": 6197,
      "time_per_iteration": 2.670529842376709
    },
    {
      "auxiliary_loss_clip": 0.01156866,
      "auxiliary_loss_mlp": 0.00761396,
      "balance_loss_clip": 1.04746723,
      "balance_loss_mlp": 1.00039673,
      "epoch": 0.7452654361810858,
      "flos": 12495405980160.0,
      "grad_norm": 1.8625627386220516,
      "language_loss": 0.71451122,
      "learning_rate": 6.427846669269952e-07,
      "loss": 0.73369384,
      "num_input_tokens_seen": 133273210,
      "step": 6198,
      "time_per_iteration": 2.626400947570801
    },
    {
      "auxiliary_loss_clip": 0.01113283,
      "auxiliary_loss_mlp": 0.01023719,
      "balance_loss_clip": 1.04998112,
      "balance_loss_mlp": 1.01541662,
      "epoch": 0.7453856790717249,
      "flos": 22127329687680.0,
      "grad_norm": 2.1121111136258834,
      "language_loss": 0.82545215,
      "learning_rate": 6.422126144763729e-07,
      "loss": 0.8468222,
      "num_input_tokens_seen": 133292600,
      "step": 6199,
      "time_per_iteration": 2.796731948852539
    },
    {
      "auxiliary_loss_clip": 0.01145122,
      "auxiliary_loss_mlp": 0.01028133,
      "balance_loss_clip": 1.04820621,
      "balance_loss_mlp": 1.01987755,
      "epoch": 0.745505921962364,
      "flos": 20010682995840.0,
      "grad_norm": 2.119453896201903,
      "language_loss": 0.7687934,
      "learning_rate": 6.416407680010174e-07,
      "loss": 0.79052591,
      "num_input_tokens_seen": 133306960,
      "step": 6200,
      "time_per_iteration": 2.6780428886413574
    },
    {
      "auxiliary_loss_clip": 0.01122051,
      "auxiliary_loss_mlp": 0.00761775,
      "balance_loss_clip": 1.04595721,
      "balance_loss_mlp": 1.00052726,
      "epoch": 0.745626164853003,
      "flos": 24677884673280.0,
      "grad_norm": 2.4573078829342268,
      "language_loss": 0.81166387,
      "learning_rate": 6.410691275876774e-07,
      "loss": 0.83050215,
      "num_input_tokens_seen": 133326380,
      "step": 6201,
      "time_per_iteration": 2.854618787765503
    },
    {
      "auxiliary_loss_clip": 0.01115267,
      "auxiliary_loss_mlp": 0.01025199,
      "balance_loss_clip": 1.04206455,
      "balance_loss_mlp": 1.01784933,
      "epoch": 0.7457464077436422,
      "flos": 14538830797440.0,
      "grad_norm": 2.340247344944224,
      "language_loss": 0.76605982,
      "learning_rate": 6.404976933230704e-07,
      "loss": 0.7874645,
      "num_input_tokens_seen": 133342900,
      "step": 6202,
      "time_per_iteration": 2.7294363975524902
    },
    {
      "auxiliary_loss_clip": 0.0108858,
      "auxiliary_loss_mlp": 0.01030618,
      "balance_loss_clip": 1.04252768,
      "balance_loss_mlp": 1.02204096,
      "epoch": 0.7458666506342813,
      "flos": 34021194600960.0,
      "grad_norm": 2.2380361978696364,
      "language_loss": 0.72710633,
      "learning_rate": 6.399264652938813e-07,
      "loss": 0.74829829,
      "num_input_tokens_seen": 133363805,
      "step": 6203,
      "time_per_iteration": 3.0772581100463867
    },
    {
      "auxiliary_loss_clip": 0.01160723,
      "auxiliary_loss_mlp": 0.01026131,
      "balance_loss_clip": 1.0466435,
      "balance_loss_mlp": 1.01847172,
      "epoch": 0.7459868935249203,
      "flos": 24279025075200.0,
      "grad_norm": 1.918141045449746,
      "language_loss": 0.74359453,
      "learning_rate": 6.393554435867679e-07,
      "loss": 0.76546311,
      "num_input_tokens_seen": 133384655,
      "step": 6204,
      "time_per_iteration": 2.724607467651367
    },
    {
      "auxiliary_loss_clip": 0.01094837,
      "auxiliary_loss_mlp": 0.01022511,
      "balance_loss_clip": 1.04000497,
      "balance_loss_mlp": 1.01542377,
      "epoch": 0.7461071364155595,
      "flos": 21908777385600.0,
      "grad_norm": 2.695230563633683,
      "language_loss": 0.83293676,
      "learning_rate": 6.387846282883502e-07,
      "loss": 0.85411024,
      "num_input_tokens_seen": 133401185,
      "step": 6205,
      "time_per_iteration": 2.846846580505371
    },
    {
      "auxiliary_loss_clip": 0.01128874,
      "auxiliary_loss_mlp": 0.01026783,
      "balance_loss_clip": 1.04222059,
      "balance_loss_mlp": 1.01886737,
      "epoch": 0.7462273793061985,
      "flos": 22889712879360.0,
      "grad_norm": 2.9757107248418984,
      "language_loss": 0.7695877,
      "learning_rate": 6.38214019485223e-07,
      "loss": 0.79114425,
      "num_input_tokens_seen": 133420010,
      "step": 6206,
      "time_per_iteration": 2.6895651817321777
    },
    {
      "auxiliary_loss_clip": 0.01125378,
      "auxiliary_loss_mlp": 0.0102373,
      "balance_loss_clip": 1.04320157,
      "balance_loss_mlp": 1.01622844,
      "epoch": 0.7463476221968376,
      "flos": 19968451580160.0,
      "grad_norm": 1.8483333283317718,
      "language_loss": 0.71450019,
      "learning_rate": 6.376436172639461e-07,
      "loss": 0.73599124,
      "num_input_tokens_seen": 133437855,
      "step": 6207,
      "time_per_iteration": 4.030192613601685
    },
    {
      "auxiliary_loss_clip": 0.01154115,
      "auxiliary_loss_mlp": 0.00762407,
      "balance_loss_clip": 1.04849792,
      "balance_loss_mlp": 1.00048113,
      "epoch": 0.7464678650874768,
      "flos": 16836610798080.0,
      "grad_norm": 2.453351918352618,
      "language_loss": 0.64895761,
      "learning_rate": 6.370734217110487e-07,
      "loss": 0.66812277,
      "num_input_tokens_seen": 133456600,
      "step": 6208,
      "time_per_iteration": 2.707571268081665
    },
    {
      "auxiliary_loss_clip": 0.01150431,
      "auxiliary_loss_mlp": 0.01025423,
      "balance_loss_clip": 1.0497961,
      "balance_loss_mlp": 1.01728094,
      "epoch": 0.7465881079781158,
      "flos": 48100869843840.0,
      "grad_norm": 1.5732824832871715,
      "language_loss": 0.64467013,
      "learning_rate": 6.36503432913031e-07,
      "loss": 0.66642869,
      "num_input_tokens_seen": 133479745,
      "step": 6209,
      "time_per_iteration": 2.976313829421997
    },
    {
      "auxiliary_loss_clip": 0.01144724,
      "auxiliary_loss_mlp": 0.01025741,
      "balance_loss_clip": 1.0450511,
      "balance_loss_mlp": 1.01724124,
      "epoch": 0.7467083508687549,
      "flos": 19677359761920.0,
      "grad_norm": 2.086069545384096,
      "language_loss": 0.69148755,
      "learning_rate": 6.359336509563569e-07,
      "loss": 0.71319222,
      "num_input_tokens_seen": 133495765,
      "step": 6210,
      "time_per_iteration": 2.6439907550811768
    },
    {
      "auxiliary_loss_clip": 0.01172395,
      "auxiliary_loss_mlp": 0.0102908,
      "balance_loss_clip": 1.04904151,
      "balance_loss_mlp": 1.021433,
      "epoch": 0.7468285937593939,
      "flos": 17895436934400.0,
      "grad_norm": 2.001671173896852,
      "language_loss": 0.8075639,
      "learning_rate": 6.353640759274641e-07,
      "loss": 0.82957864,
      "num_input_tokens_seen": 133514655,
      "step": 6211,
      "time_per_iteration": 2.5889322757720947
    },
    {
      "auxiliary_loss_clip": 0.01143387,
      "auxiliary_loss_mlp": 0.01023955,
      "balance_loss_clip": 1.04226542,
      "balance_loss_mlp": 1.01665044,
      "epoch": 0.7469488366500331,
      "flos": 23141446369920.0,
      "grad_norm": 6.506208321267708,
      "language_loss": 0.74728549,
      "learning_rate": 6.347947079127556e-07,
      "loss": 0.76895893,
      "num_input_tokens_seen": 133532555,
      "step": 6212,
      "time_per_iteration": 2.685576915740967
    },
    {
      "auxiliary_loss_clip": 0.01138649,
      "auxiliary_loss_mlp": 0.00762101,
      "balance_loss_clip": 1.04732311,
      "balance_loss_mlp": 1.00040102,
      "epoch": 0.7470690795406721,
      "flos": 16690849407360.0,
      "grad_norm": 2.3060485098422454,
      "language_loss": 0.77259743,
      "learning_rate": 6.342255469986053e-07,
      "loss": 0.791605,
      "num_input_tokens_seen": 133551300,
      "step": 6213,
      "time_per_iteration": 2.609556198120117
    },
    {
      "auxiliary_loss_clip": 0.01082621,
      "auxiliary_loss_mlp": 0.00762346,
      "balance_loss_clip": 1.04153132,
      "balance_loss_mlp": 1.00045967,
      "epoch": 0.7471893224313112,
      "flos": 25192700352000.0,
      "grad_norm": 2.261502159154436,
      "language_loss": 0.76431543,
      "learning_rate": 6.336565932713533e-07,
      "loss": 0.78276515,
      "num_input_tokens_seen": 133570725,
      "step": 6214,
      "time_per_iteration": 2.882030487060547
    },
    {
      "auxiliary_loss_clip": 0.01154082,
      "auxiliary_loss_mlp": 0.01026328,
      "balance_loss_clip": 1.04827476,
      "balance_loss_mlp": 1.01835918,
      "epoch": 0.7473095653219504,
      "flos": 22526225199360.0,
      "grad_norm": 2.466127592391125,
      "language_loss": 0.78031707,
      "learning_rate": 6.330878468173088e-07,
      "loss": 0.80212116,
      "num_input_tokens_seen": 133590790,
      "step": 6215,
      "time_per_iteration": 3.6874840259552
    },
    {
      "auxiliary_loss_clip": 0.01133087,
      "auxiliary_loss_mlp": 0.01024853,
      "balance_loss_clip": 1.04161692,
      "balance_loss_mlp": 1.01674056,
      "epoch": 0.7474298082125894,
      "flos": 18113989236480.0,
      "grad_norm": 1.8958897113383955,
      "language_loss": 0.72704339,
      "learning_rate": 6.32519307722752e-07,
      "loss": 0.74862283,
      "num_input_tokens_seen": 133608685,
      "step": 6216,
      "time_per_iteration": 3.6172356605529785
    },
    {
      "auxiliary_loss_clip": 0.01042306,
      "auxiliary_loss_mlp": 0.0100042,
      "balance_loss_clip": 1.01237881,
      "balance_loss_mlp": 0.99924576,
      "epoch": 0.7475500511032285,
      "flos": 62086535193600.0,
      "grad_norm": 0.8151869763241563,
      "language_loss": 0.54954928,
      "learning_rate": 6.31950976073929e-07,
      "loss": 0.56997657,
      "num_input_tokens_seen": 133662775,
      "step": 6217,
      "time_per_iteration": 4.186375617980957
    },
    {
      "auxiliary_loss_clip": 0.01150836,
      "auxiliary_loss_mlp": 0.01023003,
      "balance_loss_clip": 1.04659879,
      "balance_loss_mlp": 1.01517737,
      "epoch": 0.7476702939938676,
      "flos": 17785586165760.0,
      "grad_norm": 3.129328748158278,
      "language_loss": 0.81056798,
      "learning_rate": 6.31382851957055e-07,
      "loss": 0.83230639,
      "num_input_tokens_seen": 133679595,
      "step": 6218,
      "time_per_iteration": 2.6966307163238525
    },
    {
      "auxiliary_loss_clip": 0.01143284,
      "auxiliary_loss_mlp": 0.01025715,
      "balance_loss_clip": 1.04664779,
      "balance_loss_mlp": 1.01853025,
      "epoch": 0.7477905368845067,
      "flos": 27927944092800.0,
      "grad_norm": 2.0951046108011764,
      "language_loss": 0.71943986,
      "learning_rate": 6.308149354583143e-07,
      "loss": 0.74112988,
      "num_input_tokens_seen": 133699000,
      "step": 6219,
      "time_per_iteration": 2.7296793460845947
    },
    {
      "auxiliary_loss_clip": 0.01151039,
      "auxiliary_loss_mlp": 0.01022076,
      "balance_loss_clip": 1.04556155,
      "balance_loss_mlp": 1.01385701,
      "epoch": 0.7479107797751458,
      "flos": 26870374932480.0,
      "grad_norm": 2.354651896121605,
      "language_loss": 0.81787497,
      "learning_rate": 6.302472266638586e-07,
      "loss": 0.83960611,
      "num_input_tokens_seen": 133719540,
      "step": 6220,
      "time_per_iteration": 2.770827054977417
    },
    {
      "auxiliary_loss_clip": 0.01169101,
      "auxiliary_loss_mlp": 0.01029916,
      "balance_loss_clip": 1.05185246,
      "balance_loss_mlp": 1.02163649,
      "epoch": 0.7480310226657849,
      "flos": 33943375785600.0,
      "grad_norm": 2.8527374611666354,
      "language_loss": 0.70227021,
      "learning_rate": 6.296797256598101e-07,
      "loss": 0.72426039,
      "num_input_tokens_seen": 133741020,
      "step": 6221,
      "time_per_iteration": 2.7240071296691895
    },
    {
      "auxiliary_loss_clip": 0.01145755,
      "auxiliary_loss_mlp": 0.01030597,
      "balance_loss_clip": 1.04512835,
      "balance_loss_mlp": 1.0225203,
      "epoch": 0.748151265556424,
      "flos": 24826555065600.0,
      "grad_norm": 2.042881925545529,
      "language_loss": 0.81608081,
      "learning_rate": 6.291124325322576e-07,
      "loss": 0.83784437,
      "num_input_tokens_seen": 133761145,
      "step": 6222,
      "time_per_iteration": 2.7080652713775635
    },
    {
      "auxiliary_loss_clip": 0.01101571,
      "auxiliary_loss_mlp": 0.01027632,
      "balance_loss_clip": 1.04445124,
      "balance_loss_mlp": 1.02041364,
      "epoch": 0.748271508447063,
      "flos": 38399351535360.0,
      "grad_norm": 1.6600263612305632,
      "language_loss": 0.6263712,
      "learning_rate": 6.285453473672595e-07,
      "loss": 0.64766324,
      "num_input_tokens_seen": 133783715,
      "step": 6223,
      "time_per_iteration": 2.9626505374908447
    },
    {
      "auxiliary_loss_clip": 0.01170961,
      "auxiliary_loss_mlp": 0.01022789,
      "balance_loss_clip": 1.0482496,
      "balance_loss_mlp": 1.01523662,
      "epoch": 0.7483917513377022,
      "flos": 21541842000000.0,
      "grad_norm": 3.1320602382694314,
      "language_loss": 0.7544024,
      "learning_rate": 6.279784702508415e-07,
      "loss": 0.77633989,
      "num_input_tokens_seen": 133804465,
      "step": 6224,
      "time_per_iteration": 2.648803949356079
    },
    {
      "auxiliary_loss_clip": 0.01030056,
      "auxiliary_loss_mlp": 0.01002303,
      "balance_loss_clip": 1.01256323,
      "balance_loss_mlp": 1.00102794,
      "epoch": 0.7485119942283412,
      "flos": 62314532772480.0,
      "grad_norm": 0.9633675547331322,
      "language_loss": 0.5860076,
      "learning_rate": 6.274118012689979e-07,
      "loss": 0.60633123,
      "num_input_tokens_seen": 133866365,
      "step": 6225,
      "time_per_iteration": 3.3909029960632324
    },
    {
      "auxiliary_loss_clip": 0.01146476,
      "auxiliary_loss_mlp": 0.0102599,
      "balance_loss_clip": 1.04845905,
      "balance_loss_mlp": 1.01801515,
      "epoch": 0.7486322371189803,
      "flos": 29937613104000.0,
      "grad_norm": 1.687800964435685,
      "language_loss": 0.68222857,
      "learning_rate": 6.268453405076943e-07,
      "loss": 0.70395327,
      "num_input_tokens_seen": 133888760,
      "step": 6226,
      "time_per_iteration": 2.7588014602661133
    },
    {
      "auxiliary_loss_clip": 0.01122706,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.03984571,
      "balance_loss_mlp": 1.01864028,
      "epoch": 0.7487524800096195,
      "flos": 18949414734720.0,
      "grad_norm": 5.908761863150356,
      "language_loss": 0.82275516,
      "learning_rate": 6.262790880528592e-07,
      "loss": 0.84424478,
      "num_input_tokens_seen": 133906380,
      "step": 6227,
      "time_per_iteration": 2.6854777336120605
    },
    {
      "auxiliary_loss_clip": 0.01158107,
      "auxiliary_loss_mlp": 0.0102909,
      "balance_loss_clip": 1.0474726,
      "balance_loss_mlp": 1.02039993,
      "epoch": 0.7488727229002585,
      "flos": 18697393935360.0,
      "grad_norm": 3.0745549035357356,
      "language_loss": 0.79199988,
      "learning_rate": 6.257130439903951e-07,
      "loss": 0.81387186,
      "num_input_tokens_seen": 133922875,
      "step": 6228,
      "time_per_iteration": 2.649928092956543
    },
    {
      "auxiliary_loss_clip": 0.01159775,
      "auxiliary_loss_mlp": 0.01025904,
      "balance_loss_clip": 1.04937887,
      "balance_loss_mlp": 1.01782763,
      "epoch": 0.7489929657908976,
      "flos": 23623368168960.0,
      "grad_norm": 2.0580225771593477,
      "language_loss": 0.81079161,
      "learning_rate": 6.251472084061695e-07,
      "loss": 0.8326484,
      "num_input_tokens_seen": 133941795,
      "step": 6229,
      "time_per_iteration": 2.668076992034912
    },
    {
      "auxiliary_loss_clip": 0.0112005,
      "auxiliary_loss_mlp": 0.01023229,
      "balance_loss_clip": 1.04176235,
      "balance_loss_mlp": 1.01557302,
      "epoch": 0.7491132086815367,
      "flos": 20551533056640.0,
      "grad_norm": 2.1627065697870353,
      "language_loss": 0.88833469,
      "learning_rate": 6.245815813860191e-07,
      "loss": 0.90976745,
      "num_input_tokens_seen": 133957305,
      "step": 6230,
      "time_per_iteration": 2.6718363761901855
    },
    {
      "auxiliary_loss_clip": 0.01113423,
      "auxiliary_loss_mlp": 0.01028496,
      "balance_loss_clip": 1.04094279,
      "balance_loss_mlp": 1.01951122,
      "epoch": 0.7492334515721758,
      "flos": 23003011353600.0,
      "grad_norm": 3.6884199933222988,
      "language_loss": 0.70257711,
      "learning_rate": 6.240161630157495e-07,
      "loss": 0.72399628,
      "num_input_tokens_seen": 133976660,
      "step": 6231,
      "time_per_iteration": 2.8082168102264404
    },
    {
      "auxiliary_loss_clip": 0.01123649,
      "auxiliary_loss_mlp": 0.01028282,
      "balance_loss_clip": 1.04358542,
      "balance_loss_mlp": 1.02001143,
      "epoch": 0.7493536944628149,
      "flos": 16398823835520.0,
      "grad_norm": 2.9111853770256584,
      "language_loss": 0.69783318,
      "learning_rate": 6.23450953381133e-07,
      "loss": 0.71935248,
      "num_input_tokens_seen": 133994750,
      "step": 6232,
      "time_per_iteration": 3.6570537090301514
    },
    {
      "auxiliary_loss_clip": 0.01113879,
      "auxiliary_loss_mlp": 0.01021484,
      "balance_loss_clip": 1.0461899,
      "balance_loss_mlp": 1.01354432,
      "epoch": 0.749473937353454,
      "flos": 15338561155200.0,
      "grad_norm": 1.933050977788271,
      "language_loss": 0.67531025,
      "learning_rate": 6.228859525679131e-07,
      "loss": 0.69666392,
      "num_input_tokens_seen": 134009165,
      "step": 6233,
      "time_per_iteration": 2.76340913772583
    },
    {
      "auxiliary_loss_clip": 0.01118929,
      "auxiliary_loss_mlp": 0.01027288,
      "balance_loss_clip": 1.04056156,
      "balance_loss_mlp": 1.01996875,
      "epoch": 0.7495941802440931,
      "flos": 18951138587520.0,
      "grad_norm": 2.700811295775148,
      "language_loss": 0.79625964,
      "learning_rate": 6.223211606617986e-07,
      "loss": 0.81772184,
      "num_input_tokens_seen": 134027585,
      "step": 6234,
      "time_per_iteration": 2.6944146156311035
    },
    {
      "auxiliary_loss_clip": 0.01117732,
      "auxiliary_loss_mlp": 0.01023998,
      "balance_loss_clip": 1.04391146,
      "balance_loss_mlp": 1.0170455,
      "epoch": 0.7497144231347321,
      "flos": 22492469393280.0,
      "grad_norm": 2.7462231081797923,
      "language_loss": 0.84186316,
      "learning_rate": 6.217565777484701e-07,
      "loss": 0.86328042,
      "num_input_tokens_seen": 134046680,
      "step": 6235,
      "time_per_iteration": 2.6753246784210205
    },
    {
      "auxiliary_loss_clip": 0.01147443,
      "auxiliary_loss_mlp": 0.01022856,
      "balance_loss_clip": 1.04542017,
      "balance_loss_mlp": 1.01477993,
      "epoch": 0.7498346660253713,
      "flos": 24243509502720.0,
      "grad_norm": 1.9456640702587673,
      "language_loss": 0.80193198,
      "learning_rate": 6.211922039135722e-07,
      "loss": 0.82363498,
      "num_input_tokens_seen": 134066825,
      "step": 6236,
      "time_per_iteration": 2.673220634460449
    },
    {
      "auxiliary_loss_clip": 0.01144152,
      "auxiliary_loss_mlp": 0.0102185,
      "balance_loss_clip": 1.04443872,
      "balance_loss_mlp": 1.01423812,
      "epoch": 0.7499549089160104,
      "flos": 24387080163840.0,
      "grad_norm": 1.8918972325394172,
      "language_loss": 0.80942154,
      "learning_rate": 6.206280392427201e-07,
      "loss": 0.83108157,
      "num_input_tokens_seen": 134086410,
      "step": 6237,
      "time_per_iteration": 2.6590187549591064
    },
    {
      "auxiliary_loss_clip": 0.01142112,
      "auxiliary_loss_mlp": 0.01025135,
      "balance_loss_clip": 1.04588318,
      "balance_loss_mlp": 1.01752901,
      "epoch": 0.7500751518066494,
      "flos": 34057320704640.0,
      "grad_norm": 1.8433327701862232,
      "language_loss": 0.74030244,
      "learning_rate": 6.200640838214983e-07,
      "loss": 0.76197487,
      "num_input_tokens_seen": 134109185,
      "step": 6238,
      "time_per_iteration": 2.791802406311035
    },
    {
      "auxiliary_loss_clip": 0.01168562,
      "auxiliary_loss_mlp": 0.01021353,
      "balance_loss_clip": 1.04709256,
      "balance_loss_mlp": 1.01416206,
      "epoch": 0.7501953946972886,
      "flos": 18843586289280.0,
      "grad_norm": 2.141580552923594,
      "language_loss": 0.66680992,
      "learning_rate": 6.195003377354578e-07,
      "loss": 0.68870914,
      "num_input_tokens_seen": 134128455,
      "step": 6239,
      "time_per_iteration": 2.707071304321289
    },
    {
      "auxiliary_loss_clip": 0.01131503,
      "auxiliary_loss_mlp": 0.00762252,
      "balance_loss_clip": 1.0443368,
      "balance_loss_mlp": 1.00051761,
      "epoch": 0.7503156375879276,
      "flos": 20257675891200.0,
      "grad_norm": 3.1030916415775898,
      "language_loss": 0.7351988,
      "learning_rate": 6.189368010701183e-07,
      "loss": 0.75413632,
      "num_input_tokens_seen": 134145515,
      "step": 6240,
      "time_per_iteration": 2.6625936031341553
    },
    {
      "auxiliary_loss_clip": 0.01109325,
      "auxiliary_loss_mlp": 0.0103173,
      "balance_loss_clip": 1.03967834,
      "balance_loss_mlp": 1.024261,
      "epoch": 0.7504358804785667,
      "flos": 13480040574720.0,
      "grad_norm": 2.055812101262743,
      "language_loss": 0.76638269,
      "learning_rate": 6.183734739109683e-07,
      "loss": 0.78779328,
      "num_input_tokens_seen": 134163335,
      "step": 6241,
      "time_per_iteration": 3.654130697250366
    },
    {
      "auxiliary_loss_clip": 0.01092539,
      "auxiliary_loss_mlp": 0.00762616,
      "balance_loss_clip": 1.03860712,
      "balance_loss_mlp": 1.00042748,
      "epoch": 0.7505561233692057,
      "flos": 29461042431360.0,
      "grad_norm": 14.996826809905484,
      "language_loss": 0.6881308,
      "learning_rate": 6.178103563434629e-07,
      "loss": 0.70668232,
      "num_input_tokens_seen": 134182335,
      "step": 6242,
      "time_per_iteration": 3.8241066932678223
    },
    {
      "auxiliary_loss_clip": 0.0113353,
      "auxiliary_loss_mlp": 0.01030068,
      "balance_loss_clip": 1.04687512,
      "balance_loss_mlp": 1.02190793,
      "epoch": 0.7506763662598449,
      "flos": 20302457172480.0,
      "grad_norm": 1.7851525762667957,
      "language_loss": 0.84059298,
      "learning_rate": 6.172474484530283e-07,
      "loss": 0.86222887,
      "num_input_tokens_seen": 134201070,
      "step": 6243,
      "time_per_iteration": 3.584737539291382
    },
    {
      "auxiliary_loss_clip": 0.011454,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.04651177,
      "balance_loss_mlp": 1.01710749,
      "epoch": 0.750796609150484,
      "flos": 37230961939200.0,
      "grad_norm": 1.824343064063788,
      "language_loss": 0.75942975,
      "learning_rate": 6.166847503250563e-07,
      "loss": 0.7811383,
      "num_input_tokens_seen": 134223310,
      "step": 6244,
      "time_per_iteration": 2.8485147953033447
    },
    {
      "auxiliary_loss_clip": 0.01114578,
      "auxiliary_loss_mlp": 0.0102316,
      "balance_loss_clip": 1.0416851,
      "balance_loss_mlp": 1.01558983,
      "epoch": 0.750916852041123,
      "flos": 19609417186560.0,
      "grad_norm": 3.9596007127051336,
      "language_loss": 0.79173505,
      "learning_rate": 6.161222620449078e-07,
      "loss": 0.8131125,
      "num_input_tokens_seen": 134242085,
      "step": 6245,
      "time_per_iteration": 2.716600179672241
    },
    {
      "auxiliary_loss_clip": 0.01054407,
      "auxiliary_loss_mlp": 0.01029234,
      "balance_loss_clip": 1.03620291,
      "balance_loss_mlp": 1.02085972,
      "epoch": 0.7510370949317622,
      "flos": 25112690807040.0,
      "grad_norm": 2.1418917261176804,
      "language_loss": 0.80197358,
      "learning_rate": 6.155599836979117e-07,
      "loss": 0.82280993,
      "num_input_tokens_seen": 134260770,
      "step": 6246,
      "time_per_iteration": 3.2010226249694824
    },
    {
      "auxiliary_loss_clip": 0.01124609,
      "auxiliary_loss_mlp": 0.01029373,
      "balance_loss_clip": 1.04052544,
      "balance_loss_mlp": 1.02074552,
      "epoch": 0.7511573378224012,
      "flos": 19062282245760.0,
      "grad_norm": 2.179979748524003,
      "language_loss": 0.81749642,
      "learning_rate": 6.149979153693649e-07,
      "loss": 0.83903617,
      "num_input_tokens_seen": 134278025,
      "step": 6247,
      "time_per_iteration": 3.0300748348236084
    },
    {
      "auxiliary_loss_clip": 0.01104918,
      "auxiliary_loss_mlp": 0.00761961,
      "balance_loss_clip": 1.03860068,
      "balance_loss_mlp": 1.00047767,
      "epoch": 0.7512775807130403,
      "flos": 19937676602880.0,
      "grad_norm": 2.042800540807133,
      "language_loss": 0.77011478,
      "learning_rate": 6.144360571445343e-07,
      "loss": 0.78878355,
      "num_input_tokens_seen": 134297170,
      "step": 6248,
      "time_per_iteration": 2.7922611236572266
    },
    {
      "auxiliary_loss_clip": 0.0112039,
      "auxiliary_loss_mlp": 0.01029193,
      "balance_loss_clip": 1.04075098,
      "balance_loss_mlp": 1.02122974,
      "epoch": 0.7513978236036795,
      "flos": 20739920912640.0,
      "grad_norm": 2.2200315389300855,
      "language_loss": 0.80300879,
      "learning_rate": 6.138744091086509e-07,
      "loss": 0.82450461,
      "num_input_tokens_seen": 134316755,
      "step": 6249,
      "time_per_iteration": 2.7995831966400146
    },
    {
      "auxiliary_loss_clip": 0.01127659,
      "auxiliary_loss_mlp": 0.01029165,
      "balance_loss_clip": 1.04667187,
      "balance_loss_mlp": 1.02126169,
      "epoch": 0.7515180664943185,
      "flos": 27563163523200.0,
      "grad_norm": 5.386449458908482,
      "language_loss": 0.7287904,
      "learning_rate": 6.133129713469183e-07,
      "loss": 0.75035858,
      "num_input_tokens_seen": 134335960,
      "step": 6250,
      "time_per_iteration": 2.7790064811706543
    },
    {
      "auxiliary_loss_clip": 0.01101972,
      "auxiliary_loss_mlp": 0.01025809,
      "balance_loss_clip": 1.04054892,
      "balance_loss_mlp": 1.01808417,
      "epoch": 0.7516383093849576,
      "flos": 33803181002880.0,
      "grad_norm": 1.8496594922045644,
      "language_loss": 0.64387482,
      "learning_rate": 6.127517439445053e-07,
      "loss": 0.66515261,
      "num_input_tokens_seen": 134356805,
      "step": 6251,
      "time_per_iteration": 2.8659279346466064
    },
    {
      "auxiliary_loss_clip": 0.01165625,
      "auxiliary_loss_mlp": 0.01025996,
      "balance_loss_clip": 1.0481019,
      "balance_loss_mlp": 1.01893842,
      "epoch": 0.7517585522755967,
      "flos": 29746172592000.0,
      "grad_norm": 2.500883575826162,
      "language_loss": 0.82055604,
      "learning_rate": 6.121907269865498e-07,
      "loss": 0.84247226,
      "num_input_tokens_seen": 134376295,
      "step": 6252,
      "time_per_iteration": 2.6614830493927
    },
    {
      "auxiliary_loss_clip": 0.01070608,
      "auxiliary_loss_mlp": 0.00753941,
      "balance_loss_clip": 1.01172066,
      "balance_loss_mlp": 1.00105321,
      "epoch": 0.7518787951662358,
      "flos": 69807974319360.0,
      "grad_norm": 0.9307676550486553,
      "language_loss": 0.67210388,
      "learning_rate": 6.116299205581577e-07,
      "loss": 0.69034934,
      "num_input_tokens_seen": 134431125,
      "step": 6253,
      "time_per_iteration": 3.1465699672698975
    },
    {
      "auxiliary_loss_clip": 0.01095857,
      "auxiliary_loss_mlp": 0.01029782,
      "balance_loss_clip": 1.0388,
      "balance_loss_mlp": 1.0208354,
      "epoch": 0.7519990380568748,
      "flos": 34203225749760.0,
      "grad_norm": 1.8297462407962495,
      "language_loss": 0.68223596,
      "learning_rate": 6.110693247444018e-07,
      "loss": 0.7034924,
      "num_input_tokens_seen": 134452960,
      "step": 6254,
      "time_per_iteration": 2.9922027587890625
    },
    {
      "auxiliary_loss_clip": 0.01120767,
      "auxiliary_loss_mlp": 0.0102253,
      "balance_loss_clip": 1.04223144,
      "balance_loss_mlp": 1.01507902,
      "epoch": 0.752119280947514,
      "flos": 21725704742400.0,
      "grad_norm": 1.894040711346433,
      "language_loss": 0.82531941,
      "learning_rate": 6.105089396303258e-07,
      "loss": 0.84675229,
      "num_input_tokens_seen": 134471350,
      "step": 6255,
      "time_per_iteration": 2.6888463497161865
    },
    {
      "auxiliary_loss_clip": 0.01090694,
      "auxiliary_loss_mlp": 0.01024536,
      "balance_loss_clip": 1.03954387,
      "balance_loss_mlp": 1.01620305,
      "epoch": 0.7522395238381531,
      "flos": 32742774668160.0,
      "grad_norm": 1.9007534976800502,
      "language_loss": 0.75879073,
      "learning_rate": 6.099487653009383e-07,
      "loss": 0.77994299,
      "num_input_tokens_seen": 134490695,
      "step": 6256,
      "time_per_iteration": 2.922938108444214
    },
    {
      "auxiliary_loss_clip": 0.01121163,
      "auxiliary_loss_mlp": 0.01022306,
      "balance_loss_clip": 1.03891075,
      "balance_loss_mlp": 1.01532614,
      "epoch": 0.7523597667287921,
      "flos": 23476026579840.0,
      "grad_norm": 2.201230930576844,
      "language_loss": 0.83513552,
      "learning_rate": 6.093888018412192e-07,
      "loss": 0.85657018,
      "num_input_tokens_seen": 134506885,
      "step": 6257,
      "time_per_iteration": 2.7111246585845947
    },
    {
      "auxiliary_loss_clip": 0.01034592,
      "auxiliary_loss_mlp": 0.01006607,
      "balance_loss_clip": 1.01284838,
      "balance_loss_mlp": 1.00559986,
      "epoch": 0.7524800096194313,
      "flos": 67346730501120.0,
      "grad_norm": 0.7047944065062205,
      "language_loss": 0.54649466,
      "learning_rate": 6.088290493361125e-07,
      "loss": 0.56690663,
      "num_input_tokens_seen": 134571770,
      "step": 6258,
      "time_per_iteration": 4.344357490539551
    },
    {
      "auxiliary_loss_clip": 0.0113182,
      "auxiliary_loss_mlp": 0.01024738,
      "balance_loss_clip": 1.04643977,
      "balance_loss_mlp": 1.01636386,
      "epoch": 0.7526002525100703,
      "flos": 13006055681280.0,
      "grad_norm": 2.245125987223392,
      "language_loss": 0.71696693,
      "learning_rate": 6.082695078705322e-07,
      "loss": 0.73853248,
      "num_input_tokens_seen": 134589250,
      "step": 6259,
      "time_per_iteration": 2.816645622253418
    },
    {
      "auxiliary_loss_clip": 0.01119199,
      "auxiliary_loss_mlp": 0.01029979,
      "balance_loss_clip": 1.04284179,
      "balance_loss_mlp": 1.02177191,
      "epoch": 0.7527204954007094,
      "flos": 21397229844480.0,
      "grad_norm": 3.311682756546937,
      "language_loss": 0.68749702,
      "learning_rate": 6.077101775293618e-07,
      "loss": 0.70898879,
      "num_input_tokens_seen": 134608075,
      "step": 6260,
      "time_per_iteration": 3.167104959487915
    },
    {
      "auxiliary_loss_clip": 0.01084854,
      "auxiliary_loss_mlp": 0.01021951,
      "balance_loss_clip": 1.03963554,
      "balance_loss_mlp": 1.01382685,
      "epoch": 0.7528407382913486,
      "flos": 18947188091520.0,
      "grad_norm": 2.7422471348385216,
      "language_loss": 0.82782394,
      "learning_rate": 6.071510583974504e-07,
      "loss": 0.84889197,
      "num_input_tokens_seen": 134623260,
      "step": 6261,
      "time_per_iteration": 2.870253562927246
    },
    {
      "auxiliary_loss_clip": 0.01092158,
      "auxiliary_loss_mlp": 0.00762236,
      "balance_loss_clip": 1.03712487,
      "balance_loss_mlp": 1.00043249,
      "epoch": 0.7529609811819876,
      "flos": 15231798956160.0,
      "grad_norm": 2.006027547516186,
      "language_loss": 0.72247815,
      "learning_rate": 6.065921505596161e-07,
      "loss": 0.74102205,
      "num_input_tokens_seen": 134641540,
      "step": 6262,
      "time_per_iteration": 2.8320395946502686
    },
    {
      "auxiliary_loss_clip": 0.01158557,
      "auxiliary_loss_mlp": 0.01025274,
      "balance_loss_clip": 1.04838598,
      "balance_loss_mlp": 1.01770985,
      "epoch": 0.7530812240726267,
      "flos": 19354487385600.0,
      "grad_norm": 1.6449572662123395,
      "language_loss": 0.77011716,
      "learning_rate": 6.060334541006445e-07,
      "loss": 0.79195541,
      "num_input_tokens_seen": 134660035,
      "step": 6263,
      "time_per_iteration": 2.6428050994873047
    },
    {
      "auxiliary_loss_clip": 0.01073451,
      "auxiliary_loss_mlp": 0.01027897,
      "balance_loss_clip": 1.03570724,
      "balance_loss_mlp": 1.02062201,
      "epoch": 0.7532014669632658,
      "flos": 27748247328000.0,
      "grad_norm": 1.653953883427503,
      "language_loss": 0.68887699,
      "learning_rate": 6.05474969105289e-07,
      "loss": 0.70989048,
      "num_input_tokens_seen": 134683025,
      "step": 6264,
      "time_per_iteration": 2.9242565631866455
    },
    {
      "auxiliary_loss_clip": 0.01098896,
      "auxiliary_loss_mlp": 0.01026803,
      "balance_loss_clip": 1.03972912,
      "balance_loss_mlp": 1.01883936,
      "epoch": 0.7533217098539049,
      "flos": 14137421333760.0,
      "grad_norm": 2.638414031708759,
      "language_loss": 0.74145424,
      "learning_rate": 6.049166956582725e-07,
      "loss": 0.76271117,
      "num_input_tokens_seen": 134701290,
      "step": 6265,
      "time_per_iteration": 2.7487142086029053
    },
    {
      "auxiliary_loss_clip": 0.01143642,
      "auxiliary_loss_mlp": 0.0102026,
      "balance_loss_clip": 1.04396808,
      "balance_loss_mlp": 1.01323247,
      "epoch": 0.753441952744544,
      "flos": 26429068437120.0,
      "grad_norm": 2.077866498556448,
      "language_loss": 0.87224758,
      "learning_rate": 6.043586338442841e-07,
      "loss": 0.89388657,
      "num_input_tokens_seen": 134720345,
      "step": 6266,
      "time_per_iteration": 2.732408046722412
    },
    {
      "auxiliary_loss_clip": 0.01153619,
      "auxiliary_loss_mlp": 0.01020251,
      "balance_loss_clip": 1.04705918,
      "balance_loss_mlp": 1.01368248,
      "epoch": 0.7535621956351831,
      "flos": 23878621192320.0,
      "grad_norm": 1.5587777046317903,
      "language_loss": 0.7292847,
      "learning_rate": 6.038007837479815e-07,
      "loss": 0.75102341,
      "num_input_tokens_seen": 134741450,
      "step": 6267,
      "time_per_iteration": 3.5834462642669678
    },
    {
      "auxiliary_loss_clip": 0.01128369,
      "auxiliary_loss_mlp": 0.01025178,
      "balance_loss_clip": 1.0434773,
      "balance_loss_mlp": 1.01803136,
      "epoch": 0.7536824385258222,
      "flos": 21795873960960.0,
      "grad_norm": 2.18961738853899,
      "language_loss": 0.63996196,
      "learning_rate": 6.032431454539897e-07,
      "loss": 0.66149747,
      "num_input_tokens_seen": 134760295,
      "step": 6268,
      "time_per_iteration": 3.8717870712280273
    },
    {
      "auxiliary_loss_clip": 0.01152253,
      "auxiliary_loss_mlp": 0.01024621,
      "balance_loss_clip": 1.04680574,
      "balance_loss_mlp": 1.01708984,
      "epoch": 0.7538026814164612,
      "flos": 28911644933760.0,
      "grad_norm": 2.048010531008894,
      "language_loss": 0.8174141,
      "learning_rate": 6.026857190469014e-07,
      "loss": 0.83918285,
      "num_input_tokens_seen": 134782050,
      "step": 6269,
      "time_per_iteration": 3.9531612396240234
    },
    {
      "auxiliary_loss_clip": 0.01123674,
      "auxiliary_loss_mlp": 0.00762765,
      "balance_loss_clip": 1.04666305,
      "balance_loss_mlp": 1.00039029,
      "epoch": 0.7539229243071004,
      "flos": 21104701482240.0,
      "grad_norm": 2.571283132002157,
      "language_loss": 0.73863024,
      "learning_rate": 6.0212850461128e-07,
      "loss": 0.75749457,
      "num_input_tokens_seen": 134801170,
      "step": 6270,
      "time_per_iteration": 2.8298556804656982
    },
    {
      "auxiliary_loss_clip": 0.01111496,
      "auxiliary_loss_mlp": 0.01021466,
      "balance_loss_clip": 1.03623152,
      "balance_loss_mlp": 1.0127728,
      "epoch": 0.7540431671977395,
      "flos": 15158469340800.0,
      "grad_norm": 2.5517926682027094,
      "language_loss": 0.74945951,
      "learning_rate": 6.015715022316516e-07,
      "loss": 0.77078903,
      "num_input_tokens_seen": 134819150,
      "step": 6271,
      "time_per_iteration": 2.6864209175109863
    },
    {
      "auxiliary_loss_clip": 0.01144199,
      "auxiliary_loss_mlp": 0.01023713,
      "balance_loss_clip": 1.04440677,
      "balance_loss_mlp": 1.01601863,
      "epoch": 0.7541634100883785,
      "flos": 18770579896320.0,
      "grad_norm": 2.665444751347551,
      "language_loss": 0.77968812,
      "learning_rate": 6.010147119925154e-07,
      "loss": 0.80136716,
      "num_input_tokens_seen": 134836905,
      "step": 6272,
      "time_per_iteration": 2.7326548099517822
    },
    {
      "auxiliary_loss_clip": 0.01146132,
      "auxiliary_loss_mlp": 0.00761852,
      "balance_loss_clip": 1.04690206,
      "balance_loss_mlp": 1.00047946,
      "epoch": 0.7542836529790176,
      "flos": 20594770053120.0,
      "grad_norm": 1.981115648488531,
      "language_loss": 0.66702867,
      "learning_rate": 6.004581339783348e-07,
      "loss": 0.68610847,
      "num_input_tokens_seen": 134855225,
      "step": 6273,
      "time_per_iteration": 2.6692683696746826
    },
    {
      "auxiliary_loss_clip": 0.01144263,
      "auxiliary_loss_mlp": 0.01025151,
      "balance_loss_clip": 1.04427052,
      "balance_loss_mlp": 1.01669347,
      "epoch": 0.7544038958696567,
      "flos": 19095104298240.0,
      "grad_norm": 3.5638733016203634,
      "language_loss": 0.68270016,
      "learning_rate": 5.999017682735425e-07,
      "loss": 0.70439422,
      "num_input_tokens_seen": 134871615,
      "step": 6274,
      "time_per_iteration": 2.618882417678833
    },
    {
      "auxiliary_loss_clip": 0.01115186,
      "auxiliary_loss_mlp": 0.01032505,
      "balance_loss_clip": 1.03845096,
      "balance_loss_mlp": 1.02468455,
      "epoch": 0.7545241387602958,
      "flos": 31723306859520.0,
      "grad_norm": 2.589224621474695,
      "language_loss": 0.66638052,
      "learning_rate": 5.993456149625387e-07,
      "loss": 0.68785739,
      "num_input_tokens_seen": 134892765,
      "step": 6275,
      "time_per_iteration": 2.8151180744171143
    },
    {
      "auxiliary_loss_clip": 0.01138121,
      "auxiliary_loss_mlp": 0.01025573,
      "balance_loss_clip": 1.04764605,
      "balance_loss_mlp": 1.0172044,
      "epoch": 0.7546443816509348,
      "flos": 20296495514880.0,
      "grad_norm": 2.0442441063008463,
      "language_loss": 0.82288587,
      "learning_rate": 5.987896741296909e-07,
      "loss": 0.84452283,
      "num_input_tokens_seen": 134910505,
      "step": 6276,
      "time_per_iteration": 2.679577112197876
    },
    {
      "auxiliary_loss_clip": 0.01155315,
      "auxiliary_loss_mlp": 0.01027209,
      "balance_loss_clip": 1.04893136,
      "balance_loss_mlp": 1.0191865,
      "epoch": 0.754764624541574,
      "flos": 23696159080320.0,
      "grad_norm": 2.2751583495876004,
      "language_loss": 0.78524506,
      "learning_rate": 5.982339458593361e-07,
      "loss": 0.80707026,
      "num_input_tokens_seen": 134930445,
      "step": 6277,
      "time_per_iteration": 2.6637253761291504
    },
    {
      "auxiliary_loss_clip": 0.01154656,
      "auxiliary_loss_mlp": 0.01025486,
      "balance_loss_clip": 1.04346251,
      "balance_loss_mlp": 1.01816034,
      "epoch": 0.7548848674322131,
      "flos": 25337204766720.0,
      "grad_norm": 1.5681579313856135,
      "language_loss": 0.83959258,
      "learning_rate": 5.976784302357767e-07,
      "loss": 0.86139399,
      "num_input_tokens_seen": 134951010,
      "step": 6278,
      "time_per_iteration": 2.7015557289123535
    },
    {
      "auxiliary_loss_clip": 0.01111805,
      "auxiliary_loss_mlp": 0.0102979,
      "balance_loss_clip": 1.04285562,
      "balance_loss_mlp": 1.02270889,
      "epoch": 0.7550051103228521,
      "flos": 19573147428480.0,
      "grad_norm": 3.776790826241854,
      "language_loss": 0.73676753,
      "learning_rate": 5.971231273432855e-07,
      "loss": 0.75818348,
      "num_input_tokens_seen": 134970495,
      "step": 6279,
      "time_per_iteration": 2.737666368484497
    },
    {
      "auxiliary_loss_clip": 0.01041807,
      "auxiliary_loss_mlp": 0.01002717,
      "balance_loss_clip": 1.01198995,
      "balance_loss_mlp": 1.00147104,
      "epoch": 0.7551253532134913,
      "flos": 64150068648960.0,
      "grad_norm": 0.8118699690218061,
      "language_loss": 0.54505324,
      "learning_rate": 5.965680372661e-07,
      "loss": 0.56549847,
      "num_input_tokens_seen": 135028060,
      "step": 6280,
      "time_per_iteration": 3.249664068222046
    },
    {
      "auxiliary_loss_clip": 0.01123644,
      "auxiliary_loss_mlp": 0.01023512,
      "balance_loss_clip": 1.04437554,
      "balance_loss_mlp": 1.01646686,
      "epoch": 0.7552455961041303,
      "flos": 26067986968320.0,
      "grad_norm": 2.4900633110182486,
      "language_loss": 0.56493199,
      "learning_rate": 5.960131600884266e-07,
      "loss": 0.58640349,
      "num_input_tokens_seen": 135047330,
      "step": 6281,
      "time_per_iteration": 2.7677085399627686
    },
    {
      "auxiliary_loss_clip": 0.01097632,
      "auxiliary_loss_mlp": 0.01024318,
      "balance_loss_clip": 1.03487408,
      "balance_loss_mlp": 1.01757336,
      "epoch": 0.7553658389947694,
      "flos": 24498223822080.0,
      "grad_norm": 2.010232373835343,
      "language_loss": 0.76076376,
      "learning_rate": 5.954584958944413e-07,
      "loss": 0.78198326,
      "num_input_tokens_seen": 135065995,
      "step": 6282,
      "time_per_iteration": 2.783870220184326
    },
    {
      "auxiliary_loss_clip": 0.01147751,
      "auxiliary_loss_mlp": 0.01022592,
      "balance_loss_clip": 1.04443526,
      "balance_loss_mlp": 1.01484358,
      "epoch": 0.7554860818854086,
      "flos": 21799465320960.0,
      "grad_norm": 2.235299876844357,
      "language_loss": 0.8174063,
      "learning_rate": 5.949040447682854e-07,
      "loss": 0.83910978,
      "num_input_tokens_seen": 135085820,
      "step": 6283,
      "time_per_iteration": 3.6682677268981934
    },
    {
      "auxiliary_loss_clip": 0.01055926,
      "auxiliary_loss_mlp": 0.0102686,
      "balance_loss_clip": 1.03467345,
      "balance_loss_mlp": 1.01869416,
      "epoch": 0.7556063247760476,
      "flos": 16362123114240.0,
      "grad_norm": 1.996204290211943,
      "language_loss": 0.68701392,
      "learning_rate": 5.943498067940686e-07,
      "loss": 0.70784175,
      "num_input_tokens_seen": 135102845,
      "step": 6284,
      "time_per_iteration": 3.074252128601074
    },
    {
      "auxiliary_loss_clip": 0.01137578,
      "auxiliary_loss_mlp": 0.01029563,
      "balance_loss_clip": 1.04583597,
      "balance_loss_mlp": 1.02252388,
      "epoch": 0.7557265676666867,
      "flos": 27235155502080.0,
      "grad_norm": 2.0154570272925416,
      "language_loss": 0.81572104,
      "learning_rate": 5.937957820558686e-07,
      "loss": 0.83739245,
      "num_input_tokens_seen": 135122190,
      "step": 6285,
      "time_per_iteration": 2.9289934635162354
    },
    {
      "auxiliary_loss_clip": 0.01062654,
      "auxiliary_loss_mlp": 0.01002262,
      "balance_loss_clip": 1.01272392,
      "balance_loss_mlp": 1.0009861,
      "epoch": 0.7558468105573258,
      "flos": 62189131415040.0,
      "grad_norm": 0.8465773511972587,
      "language_loss": 0.65352082,
      "learning_rate": 5.932419706377296e-07,
      "loss": 0.67417002,
      "num_input_tokens_seen": 135180495,
      "step": 6286,
      "time_per_iteration": 3.224156379699707
    },
    {
      "auxiliary_loss_clip": 0.01146528,
      "auxiliary_loss_mlp": 0.01024507,
      "balance_loss_clip": 1.04369891,
      "balance_loss_mlp": 1.01686609,
      "epoch": 0.7559670534479649,
      "flos": 33249078823680.0,
      "grad_norm": 3.959756522913556,
      "language_loss": 0.74614745,
      "learning_rate": 5.92688372623666e-07,
      "loss": 0.76785779,
      "num_input_tokens_seen": 135199200,
      "step": 6287,
      "time_per_iteration": 2.800272226333618
    },
    {
      "auxiliary_loss_clip": 0.01124251,
      "auxiliary_loss_mlp": 0.01021965,
      "balance_loss_clip": 1.04295039,
      "balance_loss_mlp": 1.01406741,
      "epoch": 0.7560872963386039,
      "flos": 14064379027200.0,
      "grad_norm": 4.335206983589056,
      "language_loss": 0.73931855,
      "learning_rate": 5.921349880976574e-07,
      "loss": 0.76078075,
      "num_input_tokens_seen": 135217035,
      "step": 6288,
      "time_per_iteration": 2.7249457836151123
    },
    {
      "auxiliary_loss_clip": 0.01136879,
      "auxiliary_loss_mlp": 0.01026952,
      "balance_loss_clip": 1.04313099,
      "balance_loss_mlp": 1.01910245,
      "epoch": 0.7562075392292431,
      "flos": 20412307941120.0,
      "grad_norm": 1.7362068426619044,
      "language_loss": 0.81981671,
      "learning_rate": 5.915818171436515e-07,
      "loss": 0.84145498,
      "num_input_tokens_seen": 135236370,
      "step": 6289,
      "time_per_iteration": 2.704519271850586
    },
    {
      "auxiliary_loss_clip": 0.0115903,
      "auxiliary_loss_mlp": 0.010257,
      "balance_loss_clip": 1.04658175,
      "balance_loss_mlp": 1.01844668,
      "epoch": 0.7563277821198822,
      "flos": 20376792368640.0,
      "grad_norm": 1.7340265651991786,
      "language_loss": 0.74890685,
      "learning_rate": 5.910288598455642e-07,
      "loss": 0.7707541,
      "num_input_tokens_seen": 135255720,
      "step": 6290,
      "time_per_iteration": 2.6452412605285645
    },
    {
      "auxiliary_loss_clip": 0.01114213,
      "auxiliary_loss_mlp": 0.01030586,
      "balance_loss_clip": 1.04369152,
      "balance_loss_mlp": 1.02216399,
      "epoch": 0.7564480250105212,
      "flos": 18588261438720.0,
      "grad_norm": 2.9226016681200724,
      "language_loss": 0.74840057,
      "learning_rate": 5.90476116287278e-07,
      "loss": 0.76984859,
      "num_input_tokens_seen": 135273320,
      "step": 6291,
      "time_per_iteration": 2.7229554653167725
    },
    {
      "auxiliary_loss_clip": 0.01095057,
      "auxiliary_loss_mlp": 0.01025,
      "balance_loss_clip": 1.0386095,
      "balance_loss_mlp": 1.01714396,
      "epoch": 0.7565682679011604,
      "flos": 21215521918080.0,
      "grad_norm": 2.360653919499069,
      "language_loss": 0.67916906,
      "learning_rate": 5.899235865526456e-07,
      "loss": 0.7003696,
      "num_input_tokens_seen": 135292615,
      "step": 6292,
      "time_per_iteration": 2.7729313373565674
    },
    {
      "auxiliary_loss_clip": 0.01082833,
      "auxiliary_loss_mlp": 0.01023289,
      "balance_loss_clip": 1.04170847,
      "balance_loss_mlp": 1.01586843,
      "epoch": 0.7566885107917994,
      "flos": 20449008662400.0,
      "grad_norm": 1.813315645559114,
      "language_loss": 0.8217274,
      "learning_rate": 5.893712707254825e-07,
      "loss": 0.84278864,
      "num_input_tokens_seen": 135310075,
      "step": 6293,
      "time_per_iteration": 5.5127928256988525
    },
    {
      "auxiliary_loss_clip": 0.01097718,
      "auxiliary_loss_mlp": 0.0102459,
      "balance_loss_clip": 1.03778934,
      "balance_loss_mlp": 1.01609683,
      "epoch": 0.7568087536824385,
      "flos": 19025832919680.0,
      "grad_norm": 2.6372502419995723,
      "language_loss": 0.66232747,
      "learning_rate": 5.888191688895769e-07,
      "loss": 0.68355054,
      "num_input_tokens_seen": 135327335,
      "step": 6294,
      "time_per_iteration": 2.775542736053467
    },
    {
      "auxiliary_loss_clip": 0.01158504,
      "auxiliary_loss_mlp": 0.0102683,
      "balance_loss_clip": 1.04614031,
      "balance_loss_mlp": 1.01800823,
      "epoch": 0.7569289965730777,
      "flos": 15225442248960.0,
      "grad_norm": 2.462840207427306,
      "language_loss": 0.61963207,
      "learning_rate": 5.882672811286813e-07,
      "loss": 0.64148545,
      "num_input_tokens_seen": 135343615,
      "step": 6295,
      "time_per_iteration": 3.5210139751434326
    },
    {
      "auxiliary_loss_clip": 0.01095057,
      "auxiliary_loss_mlp": 0.01026259,
      "balance_loss_clip": 1.03856862,
      "balance_loss_mlp": 1.01797414,
      "epoch": 0.7570492394637167,
      "flos": 20769367086720.0,
      "grad_norm": 2.2248594069899377,
      "language_loss": 0.6967746,
      "learning_rate": 5.877156075265166e-07,
      "loss": 0.71798778,
      "num_input_tokens_seen": 135359880,
      "step": 6296,
      "time_per_iteration": 2.7823593616485596
    },
    {
      "auxiliary_loss_clip": 0.01141502,
      "auxiliary_loss_mlp": 0.0102693,
      "balance_loss_clip": 1.04420614,
      "balance_loss_mlp": 1.01917541,
      "epoch": 0.7571694823543558,
      "flos": 15664091137920.0,
      "grad_norm": 2.8617306265084586,
      "language_loss": 0.69969881,
      "learning_rate": 5.871641481667715e-07,
      "loss": 0.72138309,
      "num_input_tokens_seen": 135374325,
      "step": 6297,
      "time_per_iteration": 2.668478488922119
    },
    {
      "auxiliary_loss_clip": 0.01107299,
      "auxiliary_loss_mlp": 0.01026687,
      "balance_loss_clip": 1.03917897,
      "balance_loss_mlp": 1.01893222,
      "epoch": 0.7572897252449949,
      "flos": 25409241492480.0,
      "grad_norm": 1.7126044989869575,
      "language_loss": 0.84140408,
      "learning_rate": 5.866129031331011e-07,
      "loss": 0.86274391,
      "num_input_tokens_seen": 135393980,
      "step": 6298,
      "time_per_iteration": 2.7473762035369873
    },
    {
      "auxiliary_loss_clip": 0.01139801,
      "auxiliary_loss_mlp": 0.01026119,
      "balance_loss_clip": 1.04506624,
      "balance_loss_mlp": 1.01768517,
      "epoch": 0.757409968135634,
      "flos": 24279348297600.0,
      "grad_norm": 2.167793981778553,
      "language_loss": 0.82991159,
      "learning_rate": 5.8606187250913e-07,
      "loss": 0.85157084,
      "num_input_tokens_seen": 135412030,
      "step": 6299,
      "time_per_iteration": 2.724100351333618
    },
    {
      "auxiliary_loss_clip": 0.01151297,
      "auxiliary_loss_mlp": 0.01024717,
      "balance_loss_clip": 1.04667711,
      "balance_loss_mlp": 1.01665878,
      "epoch": 0.757530211026273,
      "flos": 24133766474880.0,
      "grad_norm": 1.8573876978016326,
      "language_loss": 0.84098023,
      "learning_rate": 5.855110563784482e-07,
      "loss": 0.86274034,
      "num_input_tokens_seen": 135430565,
      "step": 6300,
      "time_per_iteration": 2.7396645545959473
    },
    {
      "auxiliary_loss_clip": 0.01126747,
      "auxiliary_loss_mlp": 0.01021837,
      "balance_loss_clip": 1.0438745,
      "balance_loss_mlp": 1.01442838,
      "epoch": 0.7576504539169122,
      "flos": 23951807153280.0,
      "grad_norm": 1.8697296186014989,
      "language_loss": 0.64115876,
      "learning_rate": 5.849604548246156e-07,
      "loss": 0.66264462,
      "num_input_tokens_seen": 135451675,
      "step": 6301,
      "time_per_iteration": 2.7918457984924316
    },
    {
      "auxiliary_loss_clip": 0.01135967,
      "auxiliary_loss_mlp": 0.01023736,
      "balance_loss_clip": 1.04257083,
      "balance_loss_mlp": 1.01569867,
      "epoch": 0.7577706968075513,
      "flos": 21251360712960.0,
      "grad_norm": 3.237701675971194,
      "language_loss": 0.80173802,
      "learning_rate": 5.844100679311565e-07,
      "loss": 0.82333505,
      "num_input_tokens_seen": 135470635,
      "step": 6302,
      "time_per_iteration": 2.6686277389526367
    },
    {
      "auxiliary_loss_clip": 0.01138394,
      "auxiliary_loss_mlp": 0.01026559,
      "balance_loss_clip": 1.04578745,
      "balance_loss_mlp": 1.01857209,
      "epoch": 0.7578909396981903,
      "flos": 18296595002880.0,
      "grad_norm": 2.1647322440805583,
      "language_loss": 0.76168728,
      "learning_rate": 5.838598957815637e-07,
      "loss": 0.78333688,
      "num_input_tokens_seen": 135487865,
      "step": 6303,
      "time_per_iteration": 2.748811960220337
    },
    {
      "auxiliary_loss_clip": 0.01125152,
      "auxiliary_loss_mlp": 0.01026265,
      "balance_loss_clip": 1.04242563,
      "balance_loss_mlp": 1.01880813,
      "epoch": 0.7580111825888295,
      "flos": 25373869574400.0,
      "grad_norm": 1.6772836034100664,
      "language_loss": 0.85661173,
      "learning_rate": 5.833099384592996e-07,
      "loss": 0.87812591,
      "num_input_tokens_seen": 135508440,
      "step": 6304,
      "time_per_iteration": 2.764899492263794
    },
    {
      "auxiliary_loss_clip": 0.01158855,
      "auxiliary_loss_mlp": 0.01023474,
      "balance_loss_clip": 1.04707956,
      "balance_loss_mlp": 1.01603508,
      "epoch": 0.7581314254794685,
      "flos": 23768662682880.0,
      "grad_norm": 2.4575213261581976,
      "language_loss": 0.71561897,
      "learning_rate": 5.827601960477913e-07,
      "loss": 0.73744226,
      "num_input_tokens_seen": 135526365,
      "step": 6305,
      "time_per_iteration": 2.6850948333740234
    },
    {
      "auxiliary_loss_clip": 0.01070852,
      "auxiliary_loss_mlp": 0.01022072,
      "balance_loss_clip": 1.03813732,
      "balance_loss_mlp": 1.01450789,
      "epoch": 0.7582516683701076,
      "flos": 22054610603520.0,
      "grad_norm": 1.9223660450212825,
      "language_loss": 0.70632446,
      "learning_rate": 5.822106686304344e-07,
      "loss": 0.72725368,
      "num_input_tokens_seen": 135545655,
      "step": 6306,
      "time_per_iteration": 2.8663642406463623
    },
    {
      "auxiliary_loss_clip": 0.01118442,
      "auxiliary_loss_mlp": 0.01027278,
      "balance_loss_clip": 1.03836942,
      "balance_loss_mlp": 1.01978004,
      "epoch": 0.7583719112607467,
      "flos": 31649725848960.0,
      "grad_norm": 2.696969116895551,
      "language_loss": 0.58109826,
      "learning_rate": 5.816613562905919e-07,
      "loss": 0.60255551,
      "num_input_tokens_seen": 135566840,
      "step": 6307,
      "time_per_iteration": 3.042877435684204
    },
    {
      "auxiliary_loss_clip": 0.01140691,
      "auxiliary_loss_mlp": 0.01025982,
      "balance_loss_clip": 1.04198205,
      "balance_loss_mlp": 1.01905572,
      "epoch": 0.7584921541513858,
      "flos": 33068376478080.0,
      "grad_norm": 1.750878901329676,
      "language_loss": 0.70229757,
      "learning_rate": 5.811122591115933e-07,
      "loss": 0.72396433,
      "num_input_tokens_seen": 135587825,
      "step": 6308,
      "time_per_iteration": 2.843101739883423
    },
    {
      "auxiliary_loss_clip": 0.01102741,
      "auxiliary_loss_mlp": 0.01028586,
      "balance_loss_clip": 1.03874302,
      "balance_loss_mlp": 1.02019966,
      "epoch": 0.7586123970420249,
      "flos": 23326350606720.0,
      "grad_norm": 2.7711319491676893,
      "language_loss": 0.71136373,
      "learning_rate": 5.805633771767376e-07,
      "loss": 0.73267698,
      "num_input_tokens_seen": 135605220,
      "step": 6309,
      "time_per_iteration": 3.8085591793060303
    },
    {
      "auxiliary_loss_clip": 0.01139831,
      "auxiliary_loss_mlp": 0.01024298,
      "balance_loss_clip": 1.04509735,
      "balance_loss_mlp": 1.01679945,
      "epoch": 0.7587326399326639,
      "flos": 18334229477760.0,
      "grad_norm": 2.2981215295032005,
      "language_loss": 0.7773751,
      "learning_rate": 5.800147105692888e-07,
      "loss": 0.79901636,
      "num_input_tokens_seen": 135624795,
      "step": 6310,
      "time_per_iteration": 2.698272228240967
    },
    {
      "auxiliary_loss_clip": 0.01131655,
      "auxiliary_loss_mlp": 0.00762143,
      "balance_loss_clip": 1.04335475,
      "balance_loss_mlp": 1.00052595,
      "epoch": 0.7588528828233031,
      "flos": 17275080119040.0,
      "grad_norm": 1.8194101230231032,
      "language_loss": 0.79457819,
      "learning_rate": 5.794662593724795e-07,
      "loss": 0.81351614,
      "num_input_tokens_seen": 135643800,
      "step": 6311,
      "time_per_iteration": 2.7199857234954834
    },
    {
      "auxiliary_loss_clip": 0.01144503,
      "auxiliary_loss_mlp": 0.01025861,
      "balance_loss_clip": 1.04766226,
      "balance_loss_mlp": 1.01776695,
      "epoch": 0.7589731257139422,
      "flos": 17713621267200.0,
      "grad_norm": 1.9354543569257765,
      "language_loss": 0.75134921,
      "learning_rate": 5.789180236695091e-07,
      "loss": 0.77305281,
      "num_input_tokens_seen": 135660655,
      "step": 6312,
      "time_per_iteration": 2.6251513957977295
    },
    {
      "auxiliary_loss_clip": 0.01114145,
      "auxiliary_loss_mlp": 0.01021958,
      "balance_loss_clip": 1.04152107,
      "balance_loss_mlp": 1.01484382,
      "epoch": 0.7590933686045812,
      "flos": 15961072786560.0,
      "grad_norm": 1.8491517522099,
      "language_loss": 0.85345054,
      "learning_rate": 5.78370003543544e-07,
      "loss": 0.87481159,
      "num_input_tokens_seen": 135679410,
      "step": 6313,
      "time_per_iteration": 2.764653205871582
    },
    {
      "auxiliary_loss_clip": 0.01079128,
      "auxiliary_loss_mlp": 0.01023048,
      "balance_loss_clip": 1.03814054,
      "balance_loss_mlp": 1.01530194,
      "epoch": 0.7592136114952204,
      "flos": 21068072588160.0,
      "grad_norm": 2.872861750679581,
      "language_loss": 0.83976012,
      "learning_rate": 5.778221990777203e-07,
      "loss": 0.86078191,
      "num_input_tokens_seen": 135697150,
      "step": 6314,
      "time_per_iteration": 3.004908561706543
    },
    {
      "auxiliary_loss_clip": 0.01161533,
      "auxiliary_loss_mlp": 0.01026501,
      "balance_loss_clip": 1.04897904,
      "balance_loss_mlp": 1.018502,
      "epoch": 0.7593338543858594,
      "flos": 25297666871040.0,
      "grad_norm": 7.7868884672762615,
      "language_loss": 0.83109641,
      "learning_rate": 5.772746103551372e-07,
      "loss": 0.85297674,
      "num_input_tokens_seen": 135712545,
      "step": 6315,
      "time_per_iteration": 3.424131393432617
    },
    {
      "auxiliary_loss_clip": 0.01139752,
      "auxiliary_loss_mlp": 0.00761958,
      "balance_loss_clip": 1.04405117,
      "balance_loss_mlp": 1.00048149,
      "epoch": 0.7594540972764985,
      "flos": 31832367528960.0,
      "grad_norm": 2.4806883336104226,
      "language_loss": 0.71809417,
      "learning_rate": 5.767272374588648e-07,
      "loss": 0.73711121,
      "num_input_tokens_seen": 135733950,
      "step": 6316,
      "time_per_iteration": 2.8502821922302246
    },
    {
      "auxiliary_loss_clip": 0.01106178,
      "auxiliary_loss_mlp": 0.0102583,
      "balance_loss_clip": 1.04038024,
      "balance_loss_mlp": 1.01806962,
      "epoch": 0.7595743401671377,
      "flos": 37597250880000.0,
      "grad_norm": 2.0224908669733277,
      "language_loss": 0.78274167,
      "learning_rate": 5.76180080471939e-07,
      "loss": 0.80406177,
      "num_input_tokens_seen": 135757120,
      "step": 6317,
      "time_per_iteration": 2.9264087677001953
    },
    {
      "auxiliary_loss_clip": 0.01175286,
      "auxiliary_loss_mlp": 0.01027892,
      "balance_loss_clip": 1.04997373,
      "balance_loss_mlp": 1.01863527,
      "epoch": 0.7596945830577767,
      "flos": 18287724343680.0,
      "grad_norm": 2.3394375051420346,
      "language_loss": 0.72042203,
      "learning_rate": 5.756331394773631e-07,
      "loss": 0.74245381,
      "num_input_tokens_seen": 135773335,
      "step": 6318,
      "time_per_iteration": 3.5379161834716797
    },
    {
      "auxiliary_loss_clip": 0.01111402,
      "auxiliary_loss_mlp": 0.01026446,
      "balance_loss_clip": 1.04194927,
      "balance_loss_mlp": 1.0186975,
      "epoch": 0.7598148259484158,
      "flos": 22233122219520.0,
      "grad_norm": 1.8180901906545037,
      "language_loss": 0.75935805,
      "learning_rate": 5.750864145581071e-07,
      "loss": 0.78073657,
      "num_input_tokens_seen": 135792555,
      "step": 6319,
      "time_per_iteration": 3.6946847438812256
    },
    {
      "auxiliary_loss_clip": 0.01129533,
      "auxiliary_loss_mlp": 0.01021482,
      "balance_loss_clip": 1.04597449,
      "balance_loss_mlp": 1.01454115,
      "epoch": 0.7599350688390549,
      "flos": 27161718145920.0,
      "grad_norm": 1.946215569541602,
      "language_loss": 0.86258686,
      "learning_rate": 5.745399057971085e-07,
      "loss": 0.88409704,
      "num_input_tokens_seen": 135813690,
      "step": 6320,
      "time_per_iteration": 3.6711606979370117
    },
    {
      "auxiliary_loss_clip": 0.01153398,
      "auxiliary_loss_mlp": 0.01023782,
      "balance_loss_clip": 1.0452621,
      "balance_loss_mlp": 1.01568747,
      "epoch": 0.760055311729694,
      "flos": 15560704817280.0,
      "grad_norm": 2.0685314691868637,
      "language_loss": 0.75378275,
      "learning_rate": 5.739936132772738e-07,
      "loss": 0.77555454,
      "num_input_tokens_seen": 135832255,
      "step": 6321,
      "time_per_iteration": 2.59848690032959
    },
    {
      "auxiliary_loss_clip": 0.01106411,
      "auxiliary_loss_mlp": 0.00763583,
      "balance_loss_clip": 1.04190683,
      "balance_loss_mlp": 1.00045788,
      "epoch": 0.760175554620333,
      "flos": 25155496840320.0,
      "grad_norm": 2.7349814673605013,
      "language_loss": 0.74442887,
      "learning_rate": 5.734475370814733e-07,
      "loss": 0.76312882,
      "num_input_tokens_seen": 135851935,
      "step": 6322,
      "time_per_iteration": 2.7575113773345947
    },
    {
      "auxiliary_loss_clip": 0.01120527,
      "auxiliary_loss_mlp": 0.01028653,
      "balance_loss_clip": 1.04253399,
      "balance_loss_mlp": 1.02127707,
      "epoch": 0.7602957975109722,
      "flos": 24353791234560.0,
      "grad_norm": 2.1168634828517097,
      "language_loss": 0.78657967,
      "learning_rate": 5.729016772925483e-07,
      "loss": 0.80807143,
      "num_input_tokens_seen": 135873510,
      "step": 6323,
      "time_per_iteration": 2.768563985824585
    },
    {
      "auxiliary_loss_clip": 0.01118595,
      "auxiliary_loss_mlp": 0.0102574,
      "balance_loss_clip": 1.0403291,
      "balance_loss_mlp": 1.01747894,
      "epoch": 0.7604160404016113,
      "flos": 25192664438400.0,
      "grad_norm": 1.674375431624466,
      "language_loss": 0.70506108,
      "learning_rate": 5.723560339933038e-07,
      "loss": 0.72650445,
      "num_input_tokens_seen": 135893845,
      "step": 6324,
      "time_per_iteration": 2.7268898487091064
    },
    {
      "auxiliary_loss_clip": 0.01155701,
      "auxiliary_loss_mlp": 0.01026943,
      "balance_loss_clip": 1.04764509,
      "balance_loss_mlp": 1.01909912,
      "epoch": 0.7605362832922503,
      "flos": 29861841363840.0,
      "grad_norm": 2.253919149277009,
      "language_loss": 0.65508199,
      "learning_rate": 5.71810607266513e-07,
      "loss": 0.67690837,
      "num_input_tokens_seen": 135912430,
      "step": 6325,
      "time_per_iteration": 2.692378044128418
    },
    {
      "auxiliary_loss_clip": 0.0112731,
      "auxiliary_loss_mlp": 0.01023998,
      "balance_loss_clip": 1.0429337,
      "balance_loss_mlp": 1.01593947,
      "epoch": 0.7606565261828895,
      "flos": 13917935278080.0,
      "grad_norm": 6.149997761212732,
      "language_loss": 0.60521275,
      "learning_rate": 5.712653971949184e-07,
      "loss": 0.62672579,
      "num_input_tokens_seen": 135930550,
      "step": 6326,
      "time_per_iteration": 2.6744649410247803
    },
    {
      "auxiliary_loss_clip": 0.01138497,
      "auxiliary_loss_mlp": 0.01024827,
      "balance_loss_clip": 1.04366636,
      "balance_loss_mlp": 1.01667857,
      "epoch": 0.7607767690735285,
      "flos": 18551273408640.0,
      "grad_norm": 3.153275247484097,
      "language_loss": 0.75599068,
      "learning_rate": 5.707204038612268e-07,
      "loss": 0.77762389,
      "num_input_tokens_seen": 135947980,
      "step": 6327,
      "time_per_iteration": 2.660759687423706
    },
    {
      "auxiliary_loss_clip": 0.01140599,
      "auxiliary_loss_mlp": 0.00762607,
      "balance_loss_clip": 1.04676557,
      "balance_loss_mlp": 1.00051284,
      "epoch": 0.7608970119641676,
      "flos": 20922993555840.0,
      "grad_norm": 2.268380833342977,
      "language_loss": 0.7394973,
      "learning_rate": 5.701756273481138e-07,
      "loss": 0.75852931,
      "num_input_tokens_seen": 135965400,
      "step": 6328,
      "time_per_iteration": 2.71389102935791
    },
    {
      "auxiliary_loss_clip": 0.01149854,
      "auxiliary_loss_mlp": 0.01023415,
      "balance_loss_clip": 1.04544592,
      "balance_loss_mlp": 1.01591706,
      "epoch": 0.7610172548548068,
      "flos": 23807302738560.0,
      "grad_norm": 1.5867209888519962,
      "language_loss": 0.74047214,
      "learning_rate": 5.696310677382212e-07,
      "loss": 0.76220483,
      "num_input_tokens_seen": 135986795,
      "step": 6329,
      "time_per_iteration": 2.698094367980957
    },
    {
      "auxiliary_loss_clip": 0.01062886,
      "auxiliary_loss_mlp": 0.00753753,
      "balance_loss_clip": 1.01258695,
      "balance_loss_mlp": 1.00103819,
      "epoch": 0.7611374977454458,
      "flos": 66496580426880.0,
      "grad_norm": 0.857285149467813,
      "language_loss": 0.61742145,
      "learning_rate": 5.690867251141576e-07,
      "loss": 0.63558787,
      "num_input_tokens_seen": 136053450,
      "step": 6330,
      "time_per_iteration": 3.3433942794799805
    },
    {
      "auxiliary_loss_clip": 0.01113762,
      "auxiliary_loss_mlp": 0.01025277,
      "balance_loss_clip": 1.03818321,
      "balance_loss_mlp": 1.01764178,
      "epoch": 0.7612577406360849,
      "flos": 15633136592640.0,
      "grad_norm": 2.687293494220995,
      "language_loss": 0.91846371,
      "learning_rate": 5.685425995585013e-07,
      "loss": 0.93985415,
      "num_input_tokens_seen": 136071375,
      "step": 6331,
      "time_per_iteration": 2.7212047576904297
    },
    {
      "auxiliary_loss_clip": 0.01015889,
      "auxiliary_loss_mlp": 0.010053,
      "balance_loss_clip": 1.01272321,
      "balance_loss_mlp": 1.00405979,
      "epoch": 0.761377983526724,
      "flos": 60526253237760.0,
      "grad_norm": 0.8342045227821716,
      "language_loss": 0.58965182,
      "learning_rate": 5.679986911537935e-07,
      "loss": 0.60986376,
      "num_input_tokens_seen": 136138905,
      "step": 6332,
      "time_per_iteration": 3.432213544845581
    },
    {
      "auxiliary_loss_clip": 0.01119548,
      "auxiliary_loss_mlp": 0.01023929,
      "balance_loss_clip": 1.04109764,
      "balance_loss_mlp": 1.0164963,
      "epoch": 0.7614982264173631,
      "flos": 35772522019200.0,
      "grad_norm": 1.7347038991463681,
      "language_loss": 0.67395747,
      "learning_rate": 5.674549999825462e-07,
      "loss": 0.69539219,
      "num_input_tokens_seen": 136161720,
      "step": 6333,
      "time_per_iteration": 2.8651504516601562
    },
    {
      "auxiliary_loss_clip": 0.01035794,
      "auxiliary_loss_mlp": 0.01002248,
      "balance_loss_clip": 1.03559124,
      "balance_loss_mlp": 1.00112152,
      "epoch": 0.7616184693080021,
      "flos": 67925502345600.0,
      "grad_norm": 0.9112713524505134,
      "language_loss": 0.71365201,
      "learning_rate": 5.669115261272363e-07,
      "loss": 0.73403245,
      "num_input_tokens_seen": 136222040,
      "step": 6334,
      "time_per_iteration": 3.4080936908721924
    },
    {
      "auxiliary_loss_clip": 0.01118106,
      "auxiliary_loss_mlp": 0.00763065,
      "balance_loss_clip": 1.04104662,
      "balance_loss_mlp": 1.00046992,
      "epoch": 0.7617387121986413,
      "flos": 20521979141760.0,
      "grad_norm": 2.7144475249886972,
      "language_loss": 0.72804064,
      "learning_rate": 5.663682696703081e-07,
      "loss": 0.74685234,
      "num_input_tokens_seen": 136240305,
      "step": 6335,
      "time_per_iteration": 3.959752321243286
    },
    {
      "auxiliary_loss_clip": 0.01080269,
      "auxiliary_loss_mlp": 0.01026256,
      "balance_loss_clip": 1.03660774,
      "balance_loss_mlp": 1.01911545,
      "epoch": 0.7618589550892804,
      "flos": 18624495283200.0,
      "grad_norm": 1.8316864807218132,
      "language_loss": 0.82257652,
      "learning_rate": 5.658252306941746e-07,
      "loss": 0.84364176,
      "num_input_tokens_seen": 136259625,
      "step": 6336,
      "time_per_iteration": 2.846055030822754
    },
    {
      "auxiliary_loss_clip": 0.01124235,
      "auxiliary_loss_mlp": 0.01026774,
      "balance_loss_clip": 1.04220891,
      "balance_loss_mlp": 1.01819706,
      "epoch": 0.7619791979799194,
      "flos": 17453735389440.0,
      "grad_norm": 2.838283552185259,
      "language_loss": 0.7568804,
      "learning_rate": 5.65282409281212e-07,
      "loss": 0.77839053,
      "num_input_tokens_seen": 136277090,
      "step": 6337,
      "time_per_iteration": 2.7479255199432373
    },
    {
      "auxiliary_loss_clip": 0.01113521,
      "auxiliary_loss_mlp": 0.01026126,
      "balance_loss_clip": 1.04271126,
      "balance_loss_mlp": 1.01849651,
      "epoch": 0.7620994408705585,
      "flos": 14137421333760.0,
      "grad_norm": 2.0040096717127285,
      "language_loss": 0.70138663,
      "learning_rate": 5.64739805513768e-07,
      "loss": 0.72278309,
      "num_input_tokens_seen": 136294635,
      "step": 6338,
      "time_per_iteration": 2.759239435195923
    },
    {
      "auxiliary_loss_clip": 0.01000531,
      "auxiliary_loss_mlp": 0.01001934,
      "balance_loss_clip": 1.01063776,
      "balance_loss_mlp": 1.00084925,
      "epoch": 0.7622196837611976,
      "flos": 70708792527360.0,
      "grad_norm": 0.7885126083583752,
      "language_loss": 0.55679262,
      "learning_rate": 5.641974194741541e-07,
      "loss": 0.57681727,
      "num_input_tokens_seen": 136350320,
      "step": 6339,
      "time_per_iteration": 3.522216320037842
    },
    {
      "auxiliary_loss_clip": 0.0101413,
      "auxiliary_loss_mlp": 0.01002031,
      "balance_loss_clip": 1.01205492,
      "balance_loss_mlp": 1.00082111,
      "epoch": 0.7623399266518367,
      "flos": 60684150447360.0,
      "grad_norm": 0.7759258400630665,
      "language_loss": 0.63687652,
      "learning_rate": 5.636552512446502e-07,
      "loss": 0.65703815,
      "num_input_tokens_seen": 136411375,
      "step": 6340,
      "time_per_iteration": 3.872860908508301
    },
    {
      "auxiliary_loss_clip": 0.01110979,
      "auxiliary_loss_mlp": 0.01031311,
      "balance_loss_clip": 1.04654169,
      "balance_loss_mlp": 1.02315664,
      "epoch": 0.7624601695424758,
      "flos": 26468893641600.0,
      "grad_norm": 1.6616548216136182,
      "language_loss": 0.78210235,
      "learning_rate": 5.631133009075027e-07,
      "loss": 0.80352527,
      "num_input_tokens_seen": 136430560,
      "step": 6341,
      "time_per_iteration": 2.845489501953125
    },
    {
      "auxiliary_loss_clip": 0.0108883,
      "auxiliary_loss_mlp": 0.01028858,
      "balance_loss_clip": 1.04318166,
      "balance_loss_mlp": 1.02060866,
      "epoch": 0.7625804124331149,
      "flos": 19135755515520.0,
      "grad_norm": 2.670009235046913,
      "language_loss": 0.68617713,
      "learning_rate": 5.625715685449242e-07,
      "loss": 0.70735395,
      "num_input_tokens_seen": 136448665,
      "step": 6342,
      "time_per_iteration": 2.803743600845337
    },
    {
      "auxiliary_loss_clip": 0.01129348,
      "auxiliary_loss_mlp": 0.01025298,
      "balance_loss_clip": 1.04477906,
      "balance_loss_mlp": 1.01740634,
      "epoch": 0.762700655323754,
      "flos": 26213101914240.0,
      "grad_norm": 2.106339426969571,
      "language_loss": 0.71448064,
      "learning_rate": 5.620300542390966e-07,
      "loss": 0.73602712,
      "num_input_tokens_seen": 136469710,
      "step": 6343,
      "time_per_iteration": 2.8301475048065186
    },
    {
      "auxiliary_loss_clip": 0.01155551,
      "auxiliary_loss_mlp": 0.01021522,
      "balance_loss_clip": 1.04458427,
      "balance_loss_mlp": 1.01464915,
      "epoch": 0.762820898214393,
      "flos": 22382582711040.0,
      "grad_norm": 1.710345039702798,
      "language_loss": 0.85167921,
      "learning_rate": 5.614887580721659e-07,
      "loss": 0.87344992,
      "num_input_tokens_seen": 136489855,
      "step": 6344,
      "time_per_iteration": 2.6277897357940674
    },
    {
      "auxiliary_loss_clip": 0.0112315,
      "auxiliary_loss_mlp": 0.01027829,
      "balance_loss_clip": 1.04375732,
      "balance_loss_mlp": 1.02012181,
      "epoch": 0.7629411411050322,
      "flos": 15700504550400.0,
      "grad_norm": 1.9403706964430878,
      "language_loss": 0.73845881,
      "learning_rate": 5.609476801262481e-07,
      "loss": 0.75996864,
      "num_input_tokens_seen": 136504715,
      "step": 6345,
      "time_per_iteration": 3.662846803665161
    },
    {
      "auxiliary_loss_clip": 0.0111518,
      "auxiliary_loss_mlp": 0.01025962,
      "balance_loss_clip": 1.04207122,
      "balance_loss_mlp": 1.01772499,
      "epoch": 0.7630613839956712,
      "flos": 13770342293760.0,
      "grad_norm": 2.3641711416750475,
      "language_loss": 0.64622068,
      "learning_rate": 5.604068204834223e-07,
      "loss": 0.6676321,
      "num_input_tokens_seen": 136521610,
      "step": 6346,
      "time_per_iteration": 4.85759711265564
    },
    {
      "auxiliary_loss_clip": 0.0113923,
      "auxiliary_loss_mlp": 0.00763559,
      "balance_loss_clip": 1.04614937,
      "balance_loss_mlp": 1.0004586,
      "epoch": 0.7631816268863103,
      "flos": 14569569861120.0,
      "grad_norm": 2.398421603769333,
      "language_loss": 0.76904106,
      "learning_rate": 5.598661792257367e-07,
      "loss": 0.78806889,
      "num_input_tokens_seen": 136538655,
      "step": 6347,
      "time_per_iteration": 2.734992027282715
    },
    {
      "auxiliary_loss_clip": 0.01168867,
      "auxiliary_loss_mlp": 0.01025922,
      "balance_loss_clip": 1.04682469,
      "balance_loss_mlp": 1.01858187,
      "epoch": 0.7633018697769495,
      "flos": 19062210418560.0,
      "grad_norm": 2.1673765260065943,
      "language_loss": 0.75949776,
      "learning_rate": 5.593257564352071e-07,
      "loss": 0.78144562,
      "num_input_tokens_seen": 136557095,
      "step": 6348,
      "time_per_iteration": 2.5565714836120605
    },
    {
      "auxiliary_loss_clip": 0.01076669,
      "auxiliary_loss_mlp": 0.0102629,
      "balance_loss_clip": 1.03648233,
      "balance_loss_mlp": 1.01879811,
      "epoch": 0.7634221126675885,
      "flos": 22052958577920.0,
      "grad_norm": 2.2725041035635165,
      "language_loss": 0.7567209,
      "learning_rate": 5.58785552193815e-07,
      "loss": 0.77775055,
      "num_input_tokens_seen": 136577340,
      "step": 6349,
      "time_per_iteration": 2.8624534606933594
    },
    {
      "auxiliary_loss_clip": 0.01143095,
      "auxiliary_loss_mlp": 0.01022158,
      "balance_loss_clip": 1.04547131,
      "balance_loss_mlp": 1.01484442,
      "epoch": 0.7635423555582276,
      "flos": 29382720825600.0,
      "grad_norm": 1.9450177459403397,
      "language_loss": 0.75758946,
      "learning_rate": 5.582455665835086e-07,
      "loss": 0.77924198,
      "num_input_tokens_seen": 136597635,
      "step": 6350,
      "time_per_iteration": 2.706873893737793
    },
    {
      "auxiliary_loss_clip": 0.01160443,
      "auxiliary_loss_mlp": 0.01026203,
      "balance_loss_clip": 1.04803765,
      "balance_loss_mlp": 1.01754224,
      "epoch": 0.7636625984488667,
      "flos": 17784903807360.0,
      "grad_norm": 10.644530128915383,
      "language_loss": 0.73006809,
      "learning_rate": 5.577057996862036e-07,
      "loss": 0.75193465,
      "num_input_tokens_seen": 136615260,
      "step": 6351,
      "time_per_iteration": 2.662644386291504
    },
    {
      "auxiliary_loss_clip": 0.01108563,
      "auxiliary_loss_mlp": 0.00761209,
      "balance_loss_clip": 1.04000688,
      "balance_loss_mlp": 1.00047624,
      "epoch": 0.7637828413395058,
      "flos": 23734583654400.0,
      "grad_norm": 1.649246594382124,
      "language_loss": 0.76164556,
      "learning_rate": 5.571662515837814e-07,
      "loss": 0.78034329,
      "num_input_tokens_seen": 136637220,
      "step": 6352,
      "time_per_iteration": 2.7354443073272705
    },
    {
      "auxiliary_loss_clip": 0.01107512,
      "auxiliary_loss_mlp": 0.01021361,
      "balance_loss_clip": 1.04341722,
      "balance_loss_mlp": 1.01417828,
      "epoch": 0.7639030842301449,
      "flos": 36283279461120.0,
      "grad_norm": 2.4075909940710827,
      "language_loss": 0.83915347,
      "learning_rate": 5.566269223580926e-07,
      "loss": 0.86044222,
      "num_input_tokens_seen": 136658930,
      "step": 6353,
      "time_per_iteration": 2.8181095123291016
    },
    {
      "auxiliary_loss_clip": 0.01107944,
      "auxiliary_loss_mlp": 0.01024515,
      "balance_loss_clip": 1.04161012,
      "balance_loss_mlp": 1.01748133,
      "epoch": 0.764023327120784,
      "flos": 28878104609280.0,
      "grad_norm": 1.5587685099369628,
      "language_loss": 0.75376421,
      "learning_rate": 5.560878120909511e-07,
      "loss": 0.77508879,
      "num_input_tokens_seen": 136681530,
      "step": 6354,
      "time_per_iteration": 2.7277212142944336
    },
    {
      "auxiliary_loss_clip": 0.01029499,
      "auxiliary_loss_mlp": 0.01000808,
      "balance_loss_clip": 1.0129714,
      "balance_loss_mlp": 0.99988455,
      "epoch": 0.7641435700114231,
      "flos": 64789711067520.0,
      "grad_norm": 0.8390577708784465,
      "language_loss": 0.58489227,
      "learning_rate": 5.55548920864141e-07,
      "loss": 0.60519534,
      "num_input_tokens_seen": 136742185,
      "step": 6355,
      "time_per_iteration": 3.371077299118042
    },
    {
      "auxiliary_loss_clip": 0.01155455,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.04657638,
      "balance_loss_mlp": 1.01838768,
      "epoch": 0.7642638129020621,
      "flos": 16835784785280.0,
      "grad_norm": 2.0477445218024672,
      "language_loss": 0.78088772,
      "learning_rate": 5.550102487594113e-07,
      "loss": 0.80270123,
      "num_input_tokens_seen": 136760855,
      "step": 6356,
      "time_per_iteration": 2.6455283164978027
    },
    {
      "auxiliary_loss_clip": 0.01129586,
      "auxiliary_loss_mlp": 0.01024107,
      "balance_loss_clip": 1.04073644,
      "balance_loss_mlp": 1.01648641,
      "epoch": 0.7643840557927013,
      "flos": 30408940391040.0,
      "grad_norm": 1.7805712431201428,
      "language_loss": 0.71924496,
      "learning_rate": 5.54471795858477e-07,
      "loss": 0.7407819,
      "num_input_tokens_seen": 136780925,
      "step": 6357,
      "time_per_iteration": 2.7853450775146484
    },
    {
      "auxiliary_loss_clip": 0.01105057,
      "auxiliary_loss_mlp": 0.01023079,
      "balance_loss_clip": 1.04085135,
      "balance_loss_mlp": 1.01555085,
      "epoch": 0.7645042986833404,
      "flos": 16983234115200.0,
      "grad_norm": 2.389303122695032,
      "language_loss": 0.8318063,
      "learning_rate": 5.539335622430235e-07,
      "loss": 0.85308766,
      "num_input_tokens_seen": 136799545,
      "step": 6358,
      "time_per_iteration": 2.753145456314087
    },
    {
      "auxiliary_loss_clip": 0.01141983,
      "auxiliary_loss_mlp": 0.01026533,
      "balance_loss_clip": 1.04251933,
      "balance_loss_mlp": 1.01922274,
      "epoch": 0.7646245415739794,
      "flos": 17311493531520.0,
      "grad_norm": 2.124275449627072,
      "language_loss": 0.74958384,
      "learning_rate": 5.533955479946975e-07,
      "loss": 0.77126896,
      "num_input_tokens_seen": 136818325,
      "step": 6359,
      "time_per_iteration": 2.6480369567871094
    },
    {
      "auxiliary_loss_clip": 0.01021079,
      "auxiliary_loss_mlp": 0.01000384,
      "balance_loss_clip": 1.01123524,
      "balance_loss_mlp": 0.99914402,
      "epoch": 0.7647447844646186,
      "flos": 70402332666240.0,
      "grad_norm": 0.8543330260214637,
      "language_loss": 0.65705609,
      "learning_rate": 5.528577531951173e-07,
      "loss": 0.67727077,
      "num_input_tokens_seen": 136878730,
      "step": 6360,
      "time_per_iteration": 3.470160484313965
    },
    {
      "auxiliary_loss_clip": 0.01167987,
      "auxiliary_loss_mlp": 0.01027652,
      "balance_loss_clip": 1.04825652,
      "balance_loss_mlp": 1.02023149,
      "epoch": 0.7648650273552576,
      "flos": 17675914965120.0,
      "grad_norm": 3.35086856700708,
      "language_loss": 0.74175751,
      "learning_rate": 5.523201779258653e-07,
      "loss": 0.76371396,
      "num_input_tokens_seen": 136897705,
      "step": 6361,
      "time_per_iteration": 4.18971848487854
    },
    {
      "auxiliary_loss_clip": 0.01137428,
      "auxiliary_loss_mlp": 0.01019418,
      "balance_loss_clip": 1.04035103,
      "balance_loss_mlp": 1.01239109,
      "epoch": 0.7649852702458967,
      "flos": 22162019247360.0,
      "grad_norm": 5.418249666203305,
      "language_loss": 0.84282935,
      "learning_rate": 5.517828222684912e-07,
      "loss": 0.86439776,
      "num_input_tokens_seen": 136918360,
      "step": 6362,
      "time_per_iteration": 2.72402024269104
    },
    {
      "auxiliary_loss_clip": 0.01047935,
      "auxiliary_loss_mlp": 0.01003098,
      "balance_loss_clip": 1.01018095,
      "balance_loss_mlp": 1.00205469,
      "epoch": 0.7651055131365359,
      "flos": 69848338227840.0,
      "grad_norm": 0.7746821821257146,
      "language_loss": 0.59023172,
      "learning_rate": 5.512456863045117e-07,
      "loss": 0.61074203,
      "num_input_tokens_seen": 136979050,
      "step": 6363,
      "time_per_iteration": 3.2835373878479004
    },
    {
      "auxiliary_loss_clip": 0.01116914,
      "auxiliary_loss_mlp": 0.01022565,
      "balance_loss_clip": 1.04149961,
      "balance_loss_mlp": 1.01422071,
      "epoch": 0.7652257560271749,
      "flos": 19464014931840.0,
      "grad_norm": 1.7905490833847872,
      "language_loss": 0.74222451,
      "learning_rate": 5.507087701154089e-07,
      "loss": 0.7636193,
      "num_input_tokens_seen": 136998970,
      "step": 6364,
      "time_per_iteration": 2.764063835144043
    },
    {
      "auxiliary_loss_clip": 0.01136962,
      "auxiliary_loss_mlp": 0.0102147,
      "balance_loss_clip": 1.04579866,
      "balance_loss_mlp": 1.01405561,
      "epoch": 0.765345998917814,
      "flos": 15961108700160.0,
      "grad_norm": 1.920554120468561,
      "language_loss": 0.75336754,
      "learning_rate": 5.50172073782634e-07,
      "loss": 0.77495188,
      "num_input_tokens_seen": 137016950,
      "step": 6365,
      "time_per_iteration": 2.712585926055908
    },
    {
      "auxiliary_loss_clip": 0.0111127,
      "auxiliary_loss_mlp": 0.01025485,
      "balance_loss_clip": 1.0469625,
      "balance_loss_mlp": 1.01758695,
      "epoch": 0.7654662418084531,
      "flos": 23659853408640.0,
      "grad_norm": 1.866587906472965,
      "language_loss": 0.87881994,
      "learning_rate": 5.496355973876023e-07,
      "loss": 0.90018749,
      "num_input_tokens_seen": 137036205,
      "step": 6366,
      "time_per_iteration": 2.8046252727508545
    },
    {
      "auxiliary_loss_clip": 0.01145818,
      "auxiliary_loss_mlp": 0.01022165,
      "balance_loss_clip": 1.04557109,
      "balance_loss_mlp": 1.01302171,
      "epoch": 0.7655864846990922,
      "flos": 41463608878080.0,
      "grad_norm": 1.6206601486584025,
      "language_loss": 0.71149713,
      "learning_rate": 5.490993410116984e-07,
      "loss": 0.73317695,
      "num_input_tokens_seen": 137059195,
      "step": 6367,
      "time_per_iteration": 2.9012818336486816
    },
    {
      "auxiliary_loss_clip": 0.0113961,
      "auxiliary_loss_mlp": 0.01023818,
      "balance_loss_clip": 1.04506493,
      "balance_loss_mlp": 1.01617706,
      "epoch": 0.7657067275897312,
      "flos": 43142684088960.0,
      "grad_norm": 3.1492468776581823,
      "language_loss": 0.6954689,
      "learning_rate": 5.485633047362704e-07,
      "loss": 0.71710312,
      "num_input_tokens_seen": 137081200,
      "step": 6368,
      "time_per_iteration": 2.8529739379882812
    },
    {
      "auxiliary_loss_clip": 0.01133343,
      "auxiliary_loss_mlp": 0.01025251,
      "balance_loss_clip": 1.04485261,
      "balance_loss_mlp": 1.01724589,
      "epoch": 0.7658269704803703,
      "flos": 17311780840320.0,
      "grad_norm": 3.2300247292497772,
      "language_loss": 0.78598684,
      "learning_rate": 5.480274886426341e-07,
      "loss": 0.80757278,
      "num_input_tokens_seen": 137097840,
      "step": 6369,
      "time_per_iteration": 2.6736066341400146
    },
    {
      "auxiliary_loss_clip": 0.01109996,
      "auxiliary_loss_mlp": 0.01024256,
      "balance_loss_clip": 1.04127789,
      "balance_loss_mlp": 1.01700807,
      "epoch": 0.7659472133710095,
      "flos": 12568160977920.0,
      "grad_norm": 2.198087856191799,
      "language_loss": 0.78112352,
      "learning_rate": 5.474918928120744e-07,
      "loss": 0.80246609,
      "num_input_tokens_seen": 137114335,
      "step": 6370,
      "time_per_iteration": 3.6712899208068848
    },
    {
      "auxiliary_loss_clip": 0.01124771,
      "auxiliary_loss_mlp": 0.01027215,
      "balance_loss_clip": 1.03930998,
      "balance_loss_mlp": 1.01990759,
      "epoch": 0.7660674562616485,
      "flos": 22707430335360.0,
      "grad_norm": 1.8996836500538854,
      "language_loss": 0.87680233,
      "learning_rate": 5.469565173258392e-07,
      "loss": 0.89832211,
      "num_input_tokens_seen": 137132850,
      "step": 6371,
      "time_per_iteration": 3.7648963928222656
    },
    {
      "auxiliary_loss_clip": 0.0112311,
      "auxiliary_loss_mlp": 0.01022234,
      "balance_loss_clip": 1.04411364,
      "balance_loss_mlp": 1.01411557,
      "epoch": 0.7661876991522876,
      "flos": 17056455989760.0,
      "grad_norm": 1.7804786074328498,
      "language_loss": 0.63951802,
      "learning_rate": 5.464213622651454e-07,
      "loss": 0.66097146,
      "num_input_tokens_seen": 137150665,
      "step": 6372,
      "time_per_iteration": 3.7170722484588623
    },
    {
      "auxiliary_loss_clip": 0.01131533,
      "auxiliary_loss_mlp": 0.01025272,
      "balance_loss_clip": 1.04011226,
      "balance_loss_mlp": 1.0176332,
      "epoch": 0.7663079420429267,
      "flos": 20084228092800.0,
      "grad_norm": 1.654461996167296,
      "language_loss": 0.84544152,
      "learning_rate": 5.458864277111753e-07,
      "loss": 0.86700958,
      "num_input_tokens_seen": 137168500,
      "step": 6373,
      "time_per_iteration": 2.6782138347625732
    },
    {
      "auxiliary_loss_clip": 0.01113474,
      "auxiliary_loss_mlp": 0.01028271,
      "balance_loss_clip": 1.03955209,
      "balance_loss_mlp": 1.0210824,
      "epoch": 0.7664281849335658,
      "flos": 12677473042560.0,
      "grad_norm": 2.7170162700013263,
      "language_loss": 0.69030517,
      "learning_rate": 5.453517137450769e-07,
      "loss": 0.71172261,
      "num_input_tokens_seen": 137185075,
      "step": 6374,
      "time_per_iteration": 2.7883987426757812
    },
    {
      "auxiliary_loss_clip": 0.01147088,
      "auxiliary_loss_mlp": 0.01026819,
      "balance_loss_clip": 1.05062604,
      "balance_loss_mlp": 1.01884341,
      "epoch": 0.7665484278242048,
      "flos": 22345271458560.0,
      "grad_norm": 1.8005318025245909,
      "language_loss": 0.75890195,
      "learning_rate": 5.448172204479684e-07,
      "loss": 0.78064108,
      "num_input_tokens_seen": 137204355,
      "step": 6375,
      "time_per_iteration": 2.6369032859802246
    },
    {
      "auxiliary_loss_clip": 0.01105856,
      "auxiliary_loss_mlp": 0.0102244,
      "balance_loss_clip": 1.03811586,
      "balance_loss_mlp": 1.01467979,
      "epoch": 0.766668670714844,
      "flos": 23617909301760.0,
      "grad_norm": 2.6081143148867865,
      "language_loss": 0.74880517,
      "learning_rate": 5.442829479009294e-07,
      "loss": 0.7700882,
      "num_input_tokens_seen": 137223135,
      "step": 6376,
      "time_per_iteration": 2.919816017150879
    },
    {
      "auxiliary_loss_clip": 0.0109534,
      "auxiliary_loss_mlp": 0.01026724,
      "balance_loss_clip": 1.03928268,
      "balance_loss_mlp": 1.01810527,
      "epoch": 0.7667889136054831,
      "flos": 19427134642560.0,
      "grad_norm": 2.8860918145128176,
      "language_loss": 0.71752048,
      "learning_rate": 5.437488961850103e-07,
      "loss": 0.73874116,
      "num_input_tokens_seen": 137242935,
      "step": 6377,
      "time_per_iteration": 2.73866605758667
    },
    {
      "auxiliary_loss_clip": 0.01132306,
      "auxiliary_loss_mlp": 0.01027216,
      "balance_loss_clip": 1.04313254,
      "balance_loss_mlp": 1.0200367,
      "epoch": 0.7669091564961221,
      "flos": 26866352609280.0,
      "grad_norm": 2.362274597291829,
      "language_loss": 0.7575469,
      "learning_rate": 5.432150653812258e-07,
      "loss": 0.77914214,
      "num_input_tokens_seen": 137262970,
      "step": 6378,
      "time_per_iteration": 2.7431252002716064
    },
    {
      "auxiliary_loss_clip": 0.01134547,
      "auxiliary_loss_mlp": 0.01019014,
      "balance_loss_clip": 1.04386258,
      "balance_loss_mlp": 1.01111948,
      "epoch": 0.7670293993867613,
      "flos": 12385303816320.0,
      "grad_norm": 2.3177002289068653,
      "language_loss": 0.82560086,
      "learning_rate": 5.42681455570557e-07,
      "loss": 0.8471365,
      "num_input_tokens_seen": 137279500,
      "step": 6379,
      "time_per_iteration": 2.6336824893951416
    },
    {
      "auxiliary_loss_clip": 0.01095639,
      "auxiliary_loss_mlp": 0.01021586,
      "balance_loss_clip": 1.03805482,
      "balance_loss_mlp": 1.01401067,
      "epoch": 0.7671496422774003,
      "flos": 21762944167680.0,
      "grad_norm": 2.4321880302939074,
      "language_loss": 0.64814842,
      "learning_rate": 5.42148066833954e-07,
      "loss": 0.6693207,
      "num_input_tokens_seen": 137298745,
      "step": 6380,
      "time_per_iteration": 2.8291687965393066
    },
    {
      "auxiliary_loss_clip": 0.01131763,
      "auxiliary_loss_mlp": 0.01023235,
      "balance_loss_clip": 1.04276419,
      "balance_loss_mlp": 1.01533747,
      "epoch": 0.7672698851680394,
      "flos": 21069221823360.0,
      "grad_norm": 2.2287363317039883,
      "language_loss": 0.75494063,
      "learning_rate": 5.416148992523289e-07,
      "loss": 0.77649063,
      "num_input_tokens_seen": 137317320,
      "step": 6381,
      "time_per_iteration": 2.7945966720581055
    },
    {
      "auxiliary_loss_clip": 0.01156152,
      "auxiliary_loss_mlp": 0.01020916,
      "balance_loss_clip": 1.04820812,
      "balance_loss_mlp": 1.01344168,
      "epoch": 0.7673901280586786,
      "flos": 16976697840000.0,
      "grad_norm": 2.466890431187124,
      "language_loss": 0.78473246,
      "learning_rate": 5.410819529065644e-07,
      "loss": 0.80650306,
      "num_input_tokens_seen": 137335275,
      "step": 6382,
      "time_per_iteration": 2.6422502994537354
    },
    {
      "auxiliary_loss_clip": 0.01142022,
      "auxiliary_loss_mlp": 0.01027649,
      "balance_loss_clip": 1.04549921,
      "balance_loss_mlp": 1.02035666,
      "epoch": 0.7675103709493176,
      "flos": 29242669697280.0,
      "grad_norm": 4.351692580549945,
      "language_loss": 0.65524077,
      "learning_rate": 5.405492278775079e-07,
      "loss": 0.67693746,
      "num_input_tokens_seen": 137355055,
      "step": 6383,
      "time_per_iteration": 2.7469024658203125
    },
    {
      "auxiliary_loss_clip": 0.01151185,
      "auxiliary_loss_mlp": 0.01023713,
      "balance_loss_clip": 1.04522836,
      "balance_loss_mlp": 1.01551163,
      "epoch": 0.7676306138399567,
      "flos": 29023004073600.0,
      "grad_norm": 2.125386886740223,
      "language_loss": 0.80294561,
      "learning_rate": 5.400167242459732e-07,
      "loss": 0.82469457,
      "num_input_tokens_seen": 137374015,
      "step": 6384,
      "time_per_iteration": 2.738474130630493
    },
    {
      "auxiliary_loss_clip": 0.01119654,
      "auxiliary_loss_mlp": 0.01021902,
      "balance_loss_clip": 1.04134023,
      "balance_loss_mlp": 1.01426077,
      "epoch": 0.7677508567305958,
      "flos": 22565116650240.0,
      "grad_norm": 3.0950868269708076,
      "language_loss": 0.80764002,
      "learning_rate": 5.394844420927405e-07,
      "loss": 0.82905555,
      "num_input_tokens_seen": 137393625,
      "step": 6385,
      "time_per_iteration": 2.7306864261627197
    },
    {
      "auxiliary_loss_clip": 0.0110392,
      "auxiliary_loss_mlp": 0.01023168,
      "balance_loss_clip": 1.04026461,
      "balance_loss_mlp": 1.01561904,
      "epoch": 0.7678710996212349,
      "flos": 25411432222080.0,
      "grad_norm": 9.343068497812176,
      "language_loss": 0.73713773,
      "learning_rate": 5.389523814985562e-07,
      "loss": 0.75840861,
      "num_input_tokens_seen": 137413045,
      "step": 6386,
      "time_per_iteration": 3.739349842071533
    },
    {
      "auxiliary_loss_clip": 0.01128274,
      "auxiliary_loss_mlp": 0.01027025,
      "balance_loss_clip": 1.04301894,
      "balance_loss_mlp": 1.01902008,
      "epoch": 0.767991342511874,
      "flos": 26756825063040.0,
      "grad_norm": 13.970201936849339,
      "language_loss": 0.76581991,
      "learning_rate": 5.384205425441344e-07,
      "loss": 0.78737289,
      "num_input_tokens_seen": 137433955,
      "step": 6387,
      "time_per_iteration": 2.751523017883301
    },
    {
      "auxiliary_loss_clip": 0.01122778,
      "auxiliary_loss_mlp": 0.0102914,
      "balance_loss_clip": 1.04248011,
      "balance_loss_mlp": 1.0216502,
      "epoch": 0.7681115854025131,
      "flos": 26359509749760.0,
      "grad_norm": 4.394126401857287,
      "language_loss": 0.84526837,
      "learning_rate": 5.378889253101537e-07,
      "loss": 0.86678755,
      "num_input_tokens_seen": 137454510,
      "step": 6388,
      "time_per_iteration": 2.7968122959136963
    },
    {
      "auxiliary_loss_clip": 0.01142017,
      "auxiliary_loss_mlp": 0.0102582,
      "balance_loss_clip": 1.04164696,
      "balance_loss_mlp": 1.01857257,
      "epoch": 0.7682318282931522,
      "flos": 23257043314560.0,
      "grad_norm": 1.59050417634339,
      "language_loss": 0.81018072,
      "learning_rate": 5.373575298772617e-07,
      "loss": 0.83185911,
      "num_input_tokens_seen": 137473630,
      "step": 6389,
      "time_per_iteration": 2.810410976409912
    },
    {
      "auxiliary_loss_clip": 0.01037436,
      "auxiliary_loss_mlp": 0.01000204,
      "balance_loss_clip": 1.03285503,
      "balance_loss_mlp": 0.99913114,
      "epoch": 0.7683520711837912,
      "flos": 70072457137920.0,
      "grad_norm": 0.7743789726776639,
      "language_loss": 0.61281449,
      "learning_rate": 5.368263563260689e-07,
      "loss": 0.63319087,
      "num_input_tokens_seen": 137538765,
      "step": 6390,
      "time_per_iteration": 3.432400703430176
    },
    {
      "auxiliary_loss_clip": 0.01129793,
      "auxiliary_loss_mlp": 0.01028277,
      "balance_loss_clip": 1.04327726,
      "balance_loss_mlp": 1.02020025,
      "epoch": 0.7684723140744304,
      "flos": 18624890332800.0,
      "grad_norm": 2.533117766620987,
      "language_loss": 0.64436007,
      "learning_rate": 5.362954047371537e-07,
      "loss": 0.66594076,
      "num_input_tokens_seen": 137557875,
      "step": 6391,
      "time_per_iteration": 2.634176015853882
    },
    {
      "auxiliary_loss_clip": 0.0112669,
      "auxiliary_loss_mlp": 0.01022559,
      "balance_loss_clip": 1.04648018,
      "balance_loss_mlp": 1.01440525,
      "epoch": 0.7685925569650695,
      "flos": 27452989532160.0,
      "grad_norm": 2.0193892578478643,
      "language_loss": 0.71925098,
      "learning_rate": 5.357646751910627e-07,
      "loss": 0.7407434,
      "num_input_tokens_seen": 137579055,
      "step": 6392,
      "time_per_iteration": 2.7751681804656982
    },
    {
      "auxiliary_loss_clip": 0.01174515,
      "auxiliary_loss_mlp": 0.0102772,
      "balance_loss_clip": 1.04884386,
      "balance_loss_mlp": 1.01957774,
      "epoch": 0.7687127998557085,
      "flos": 24535714642560.0,
      "grad_norm": 2.949654269675715,
      "language_loss": 0.79552555,
      "learning_rate": 5.352341677683061e-07,
      "loss": 0.81754786,
      "num_input_tokens_seen": 137600355,
      "step": 6393,
      "time_per_iteration": 2.5922656059265137
    },
    {
      "auxiliary_loss_clip": 0.01127385,
      "auxiliary_loss_mlp": 0.01021384,
      "balance_loss_clip": 1.04173446,
      "balance_loss_mlp": 1.01395714,
      "epoch": 0.7688330427463477,
      "flos": 25155963717120.0,
      "grad_norm": 1.8684323886416179,
      "language_loss": 0.79216659,
      "learning_rate": 5.347038825493617e-07,
      "loss": 0.81365418,
      "num_input_tokens_seen": 137621885,
      "step": 6394,
      "time_per_iteration": 2.734922170639038
    },
    {
      "auxiliary_loss_clip": 0.01118224,
      "auxiliary_loss_mlp": 0.01027108,
      "balance_loss_clip": 1.04525685,
      "balance_loss_mlp": 1.01962185,
      "epoch": 0.7689532856369867,
      "flos": 21211284113280.0,
      "grad_norm": 2.1611901959273676,
      "language_loss": 0.6841923,
      "learning_rate": 5.341738196146732e-07,
      "loss": 0.70564562,
      "num_input_tokens_seen": 137640230,
      "step": 6395,
      "time_per_iteration": 2.7305469512939453
    },
    {
      "auxiliary_loss_clip": 0.01115495,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.03983271,
      "balance_loss_mlp": 1.01626611,
      "epoch": 0.7690735285276258,
      "flos": 25119083427840.0,
      "grad_norm": 2.823075001549934,
      "language_loss": 0.73566294,
      "learning_rate": 5.336439790446503e-07,
      "loss": 0.757061,
      "num_input_tokens_seen": 137659330,
      "step": 6396,
      "time_per_iteration": 3.622738838195801
    },
    {
      "auxiliary_loss_clip": 0.01140127,
      "auxiliary_loss_mlp": 0.01023363,
      "balance_loss_clip": 1.04640579,
      "balance_loss_mlp": 1.01575184,
      "epoch": 0.769193771418265,
      "flos": 54744020640000.0,
      "grad_norm": 1.768048864240742,
      "language_loss": 0.62689996,
      "learning_rate": 5.331143609196711e-07,
      "loss": 0.64853483,
      "num_input_tokens_seen": 137683145,
      "step": 6397,
      "time_per_iteration": 3.92366623878479
    },
    {
      "auxiliary_loss_clip": 0.01152721,
      "auxiliary_loss_mlp": 0.0102477,
      "balance_loss_clip": 1.04529023,
      "balance_loss_mlp": 1.01637793,
      "epoch": 0.769314014308904,
      "flos": 37341890115840.0,
      "grad_norm": 3.4483582572849247,
      "language_loss": 0.771891,
      "learning_rate": 5.325849653200758e-07,
      "loss": 0.79366589,
      "num_input_tokens_seen": 137707095,
      "step": 6398,
      "time_per_iteration": 3.7127621173858643
    },
    {
      "auxiliary_loss_clip": 0.01107759,
      "auxiliary_loss_mlp": 0.01021303,
      "balance_loss_clip": 1.04211819,
      "balance_loss_mlp": 1.0131433,
      "epoch": 0.7694342571995431,
      "flos": 20631686256000.0,
      "grad_norm": 1.7122105374928653,
      "language_loss": 0.76378012,
      "learning_rate": 5.32055792326175e-07,
      "loss": 0.78507078,
      "num_input_tokens_seen": 137725520,
      "step": 6399,
      "time_per_iteration": 2.8518970012664795
    },
    {
      "auxiliary_loss_clip": 0.01168699,
      "auxiliary_loss_mlp": 0.01023562,
      "balance_loss_clip": 1.04659867,
      "balance_loss_mlp": 1.0152235,
      "epoch": 0.7695545000901821,
      "flos": 24207706621440.0,
      "grad_norm": 1.9442446552842418,
      "language_loss": 0.72764122,
      "learning_rate": 5.315268420182437e-07,
      "loss": 0.74956387,
      "num_input_tokens_seen": 137744195,
      "step": 6400,
      "time_per_iteration": 2.6445586681365967
    },
    {
      "auxiliary_loss_clip": 0.01108056,
      "auxiliary_loss_mlp": 0.01022818,
      "balance_loss_clip": 1.04286802,
      "balance_loss_mlp": 1.0144614,
      "epoch": 0.7696747429808213,
      "flos": 28001273708160.0,
      "grad_norm": 2.674545402041332,
      "language_loss": 0.76650071,
      "learning_rate": 5.309981144765221e-07,
      "loss": 0.78780949,
      "num_input_tokens_seen": 137764340,
      "step": 6401,
      "time_per_iteration": 2.7942140102386475
    },
    {
      "auxiliary_loss_clip": 0.01159966,
      "auxiliary_loss_mlp": 0.01026375,
      "balance_loss_clip": 1.04661608,
      "balance_loss_mlp": 1.01872218,
      "epoch": 0.7697949858714603,
      "flos": 11509550323200.0,
      "grad_norm": 2.7218129243884346,
      "language_loss": 0.75752366,
      "learning_rate": 5.304696097812196e-07,
      "loss": 0.77938712,
      "num_input_tokens_seen": 137780940,
      "step": 6402,
      "time_per_iteration": 2.6037960052490234
    },
    {
      "auxiliary_loss_clip": 0.01138891,
      "auxiliary_loss_mlp": 0.01025333,
      "balance_loss_clip": 1.04205108,
      "balance_loss_mlp": 1.01622009,
      "epoch": 0.7699152287620994,
      "flos": 26688271956480.0,
      "grad_norm": 2.7032107009192705,
      "language_loss": 0.60309142,
      "learning_rate": 5.299413280125078e-07,
      "loss": 0.62473363,
      "num_input_tokens_seen": 137799250,
      "step": 6403,
      "time_per_iteration": 2.6812899112701416
    },
    {
      "auxiliary_loss_clip": 0.0112175,
      "auxiliary_loss_mlp": 0.01025945,
      "balance_loss_clip": 1.04376626,
      "balance_loss_mlp": 1.01808286,
      "epoch": 0.7700354716527386,
      "flos": 16544944362240.0,
      "grad_norm": 2.5494627748609964,
      "language_loss": 0.72892869,
      "learning_rate": 5.294132692505284e-07,
      "loss": 0.75040567,
      "num_input_tokens_seen": 137817660,
      "step": 6404,
      "time_per_iteration": 2.588594436645508
    },
    {
      "auxiliary_loss_clip": 0.01123434,
      "auxiliary_loss_mlp": 0.01023559,
      "balance_loss_clip": 1.03819978,
      "balance_loss_mlp": 1.01561713,
      "epoch": 0.7701557145433776,
      "flos": 19242733196160.0,
      "grad_norm": 2.0625562683306016,
      "language_loss": 0.79434061,
      "learning_rate": 5.288854335753861e-07,
      "loss": 0.81581056,
      "num_input_tokens_seen": 137835920,
      "step": 6405,
      "time_per_iteration": 2.567150831222534
    },
    {
      "auxiliary_loss_clip": 0.01135493,
      "auxiliary_loss_mlp": 0.01024593,
      "balance_loss_clip": 1.04318357,
      "balance_loss_mlp": 1.01630211,
      "epoch": 0.7702759574340167,
      "flos": 31685744211840.0,
      "grad_norm": 1.6048176640069283,
      "language_loss": 0.75575697,
      "learning_rate": 5.283578210671551e-07,
      "loss": 0.77735788,
      "num_input_tokens_seen": 137858160,
      "step": 6406,
      "time_per_iteration": 2.5963735580444336
    },
    {
      "auxiliary_loss_clip": 0.011239,
      "auxiliary_loss_mlp": 0.00761999,
      "balance_loss_clip": 1.04030001,
      "balance_loss_mlp": 1.00054073,
      "epoch": 0.7703962003246558,
      "flos": 16800089644800.0,
      "grad_norm": 2.2139495537348237,
      "language_loss": 0.76572835,
      "learning_rate": 5.278304318058719e-07,
      "loss": 0.78458738,
      "num_input_tokens_seen": 137876015,
      "step": 6407,
      "time_per_iteration": 2.6979525089263916
    },
    {
      "auxiliary_loss_clip": 0.01128648,
      "auxiliary_loss_mlp": 0.01029435,
      "balance_loss_clip": 1.04077435,
      "balance_loss_mlp": 1.02164519,
      "epoch": 0.7705164432152949,
      "flos": 35736072693120.0,
      "grad_norm": 2.227479393092731,
      "language_loss": 0.79109073,
      "learning_rate": 5.273032658715411e-07,
      "loss": 0.8126716,
      "num_input_tokens_seen": 137898825,
      "step": 6408,
      "time_per_iteration": 2.785487651824951
    },
    {
      "auxiliary_loss_clip": 0.01079722,
      "auxiliary_loss_mlp": 0.01025395,
      "balance_loss_clip": 1.03514016,
      "balance_loss_mlp": 1.01801836,
      "epoch": 0.7706366861059339,
      "flos": 23365960329600.0,
      "grad_norm": 3.0270898545548293,
      "language_loss": 0.76789629,
      "learning_rate": 5.267763233441347e-07,
      "loss": 0.78894746,
      "num_input_tokens_seen": 137919455,
      "step": 6409,
      "time_per_iteration": 2.829953670501709
    },
    {
      "auxiliary_loss_clip": 0.01129656,
      "auxiliary_loss_mlp": 0.01030004,
      "balance_loss_clip": 1.04172945,
      "balance_loss_mlp": 1.02178419,
      "epoch": 0.7707569289965731,
      "flos": 22929897219840.0,
      "grad_norm": 2.3864029473114665,
      "language_loss": 0.69679165,
      "learning_rate": 5.26249604303588e-07,
      "loss": 0.7183882,
      "num_input_tokens_seen": 137937960,
      "step": 6410,
      "time_per_iteration": 2.7613606452941895
    },
    {
      "auxiliary_loss_clip": 0.01170751,
      "auxiliary_loss_mlp": 0.0102621,
      "balance_loss_clip": 1.04871297,
      "balance_loss_mlp": 1.01878643,
      "epoch": 0.7708771718872122,
      "flos": 17420661941760.0,
      "grad_norm": 2.1306036261805565,
      "language_loss": 0.78808004,
      "learning_rate": 5.257231088298057e-07,
      "loss": 0.81004965,
      "num_input_tokens_seen": 137956370,
      "step": 6411,
      "time_per_iteration": 2.6444778442382812
    },
    {
      "auxiliary_loss_clip": 0.01027846,
      "auxiliary_loss_mlp": 0.01001377,
      "balance_loss_clip": 1.01425898,
      "balance_loss_mlp": 1.00035787,
      "epoch": 0.7709974147778512,
      "flos": 72241316248320.0,
      "grad_norm": 0.7931195727615057,
      "language_loss": 0.53966302,
      "learning_rate": 5.25196837002655e-07,
      "loss": 0.55995524,
      "num_input_tokens_seen": 138016080,
      "step": 6412,
      "time_per_iteration": 4.460240125656128
    },
    {
      "auxiliary_loss_clip": 0.01107644,
      "auxiliary_loss_mlp": 0.01032582,
      "balance_loss_clip": 1.04703307,
      "balance_loss_mlp": 1.02424335,
      "epoch": 0.7711176576684904,
      "flos": 39859694876160.0,
      "grad_norm": 2.8287624860350427,
      "language_loss": 0.68535995,
      "learning_rate": 5.24670788901971e-07,
      "loss": 0.70676225,
      "num_input_tokens_seen": 138039170,
      "step": 6413,
      "time_per_iteration": 2.8546104431152344
    },
    {
      "auxiliary_loss_clip": 0.01120987,
      "auxiliary_loss_mlp": 0.01023506,
      "balance_loss_clip": 1.04446208,
      "balance_loss_mlp": 1.0142014,
      "epoch": 0.7712379005591294,
      "flos": 36976391274240.0,
      "grad_norm": 3.373005504421935,
      "language_loss": 0.68778574,
      "learning_rate": 5.241449646075557e-07,
      "loss": 0.70923072,
      "num_input_tokens_seen": 138062395,
      "step": 6414,
      "time_per_iteration": 2.8782718181610107
    },
    {
      "auxiliary_loss_clip": 0.01159974,
      "auxiliary_loss_mlp": 0.01028553,
      "balance_loss_clip": 1.04563725,
      "balance_loss_mlp": 1.02033401,
      "epoch": 0.7713581434497685,
      "flos": 22776773541120.0,
      "grad_norm": 3.22962847047607,
      "language_loss": 0.73204201,
      "learning_rate": 5.236193641991762e-07,
      "loss": 0.75392723,
      "num_input_tokens_seen": 138080325,
      "step": 6415,
      "time_per_iteration": 2.5807981491088867
    },
    {
      "auxiliary_loss_clip": 0.01139698,
      "auxiliary_loss_mlp": 0.01025543,
      "balance_loss_clip": 1.04456151,
      "balance_loss_mlp": 1.0178895,
      "epoch": 0.7714783863404077,
      "flos": 24097460803200.0,
      "grad_norm": 2.297760194314681,
      "language_loss": 0.69997382,
      "learning_rate": 5.23093987756565e-07,
      "loss": 0.72162616,
      "num_input_tokens_seen": 138099020,
      "step": 6416,
      "time_per_iteration": 2.642779588699341
    },
    {
      "auxiliary_loss_clip": 0.01160921,
      "auxiliary_loss_mlp": 0.01028663,
      "balance_loss_clip": 1.04797411,
      "balance_loss_mlp": 1.02029419,
      "epoch": 0.7715986292310467,
      "flos": 21063655215360.0,
      "grad_norm": 1.9392256013739408,
      "language_loss": 0.75500131,
      "learning_rate": 5.225688353594217e-07,
      "loss": 0.77689719,
      "num_input_tokens_seen": 138118650,
      "step": 6417,
      "time_per_iteration": 2.6256942749023438
    },
    {
      "auxiliary_loss_clip": 0.01074447,
      "auxiliary_loss_mlp": 0.01028417,
      "balance_loss_clip": 1.03492928,
      "balance_loss_mlp": 1.0207963,
      "epoch": 0.7717188721216858,
      "flos": 20594877793920.0,
      "grad_norm": 2.200846687676542,
      "language_loss": 0.77799642,
      "learning_rate": 5.220439070874108e-07,
      "loss": 0.79902506,
      "num_input_tokens_seen": 138137890,
      "step": 6418,
      "time_per_iteration": 2.783684730529785
    },
    {
      "auxiliary_loss_clip": 0.01138512,
      "auxiliary_loss_mlp": 0.01023302,
      "balance_loss_clip": 1.04476345,
      "balance_loss_mlp": 1.01558328,
      "epoch": 0.7718391150123249,
      "flos": 26250951870720.0,
      "grad_norm": 2.3557076842904743,
      "language_loss": 0.71051395,
      "learning_rate": 5.215192030201652e-07,
      "loss": 0.73213208,
      "num_input_tokens_seen": 138158880,
      "step": 6419,
      "time_per_iteration": 2.7863879203796387
    },
    {
      "auxiliary_loss_clip": 0.01134517,
      "auxiliary_loss_mlp": 0.01024308,
      "balance_loss_clip": 1.04225934,
      "balance_loss_mlp": 1.01704562,
      "epoch": 0.771959357902964,
      "flos": 22049762267520.0,
      "grad_norm": 1.7143207381319814,
      "language_loss": 0.86251765,
      "learning_rate": 5.209947232372798e-07,
      "loss": 0.88410592,
      "num_input_tokens_seen": 138176370,
      "step": 6420,
      "time_per_iteration": 2.6544394493103027
    },
    {
      "auxiliary_loss_clip": 0.01139463,
      "auxiliary_loss_mlp": 0.01030586,
      "balance_loss_clip": 1.04624343,
      "balance_loss_mlp": 1.02296233,
      "epoch": 0.772079600793603,
      "flos": 30446000248320.0,
      "grad_norm": 2.4336924398540147,
      "language_loss": 0.81018907,
      "learning_rate": 5.204704678183196e-07,
      "loss": 0.83188957,
      "num_input_tokens_seen": 138195105,
      "step": 6421,
      "time_per_iteration": 2.786574125289917
    },
    {
      "auxiliary_loss_clip": 0.01084645,
      "auxiliary_loss_mlp": 0.01027556,
      "balance_loss_clip": 1.03607261,
      "balance_loss_mlp": 1.01938081,
      "epoch": 0.7721998436842422,
      "flos": 12969857750400.0,
      "grad_norm": 1.8572947686068615,
      "language_loss": 0.85456997,
      "learning_rate": 5.19946436842813e-07,
      "loss": 0.87569195,
      "num_input_tokens_seen": 138212235,
      "step": 6422,
      "time_per_iteration": 3.568347215652466
    },
    {
      "auxiliary_loss_clip": 0.01099357,
      "auxiliary_loss_mlp": 0.01025768,
      "balance_loss_clip": 1.03707647,
      "balance_loss_mlp": 1.01850545,
      "epoch": 0.7723200865748813,
      "flos": 32635509678720.0,
      "grad_norm": 2.0433729911150174,
      "language_loss": 0.68524188,
      "learning_rate": 5.194226303902546e-07,
      "loss": 0.70649314,
      "num_input_tokens_seen": 138231970,
      "step": 6423,
      "time_per_iteration": 3.7480006217956543
    },
    {
      "auxiliary_loss_clip": 0.01131054,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.04614282,
      "balance_loss_mlp": 1.0160017,
      "epoch": 0.7724403294655203,
      "flos": 21105707063040.0,
      "grad_norm": 1.7168000856504255,
      "language_loss": 0.71211076,
      "learning_rate": 5.188990485401072e-07,
      "loss": 0.73366559,
      "num_input_tokens_seen": 138251175,
      "step": 6424,
      "time_per_iteration": 3.6366167068481445
    },
    {
      "auxiliary_loss_clip": 0.01059622,
      "auxiliary_loss_mlp": 0.01025551,
      "balance_loss_clip": 1.0379647,
      "balance_loss_mlp": 1.0170399,
      "epoch": 0.7725605723561595,
      "flos": 22090736707200.0,
      "grad_norm": 1.9533863127158788,
      "language_loss": 0.86511445,
      "learning_rate": 5.183756913717954e-07,
      "loss": 0.88596618,
      "num_input_tokens_seen": 138270950,
      "step": 6425,
      "time_per_iteration": 2.9897162914276123
    },
    {
      "auxiliary_loss_clip": 0.01138736,
      "auxiliary_loss_mlp": 0.01028169,
      "balance_loss_clip": 1.04379392,
      "balance_loss_mlp": 1.02058125,
      "epoch": 0.7726808152467985,
      "flos": 34495610457600.0,
      "grad_norm": 2.4190680687538424,
      "language_loss": 0.73483419,
      "learning_rate": 5.178525589647136e-07,
      "loss": 0.75650322,
      "num_input_tokens_seen": 138292590,
      "step": 6426,
      "time_per_iteration": 3.182819366455078
    },
    {
      "auxiliary_loss_clip": 0.01131518,
      "auxiliary_loss_mlp": 0.01033526,
      "balance_loss_clip": 1.04994702,
      "balance_loss_mlp": 1.02575898,
      "epoch": 0.7728010581374376,
      "flos": 22306344094080.0,
      "grad_norm": 1.7837294376069763,
      "language_loss": 0.79213572,
      "learning_rate": 5.173296513982197e-07,
      "loss": 0.81378615,
      "num_input_tokens_seen": 138311115,
      "step": 6427,
      "time_per_iteration": 2.7326178550720215
    },
    {
      "auxiliary_loss_clip": 0.01115506,
      "auxiliary_loss_mlp": 0.00763363,
      "balance_loss_clip": 1.04181111,
      "balance_loss_mlp": 1.00043917,
      "epoch": 0.7729213010280768,
      "flos": 27126453968640.0,
      "grad_norm": 2.235423030279093,
      "language_loss": 0.64904141,
      "learning_rate": 5.168069687516398e-07,
      "loss": 0.66783011,
      "num_input_tokens_seen": 138330885,
      "step": 6428,
      "time_per_iteration": 2.7634782791137695
    },
    {
      "auxiliary_loss_clip": 0.01099275,
      "auxiliary_loss_mlp": 0.01022908,
      "balance_loss_clip": 1.04510117,
      "balance_loss_mlp": 1.01576173,
      "epoch": 0.7730415439187158,
      "flos": 18150223080960.0,
      "grad_norm": 5.843432625434206,
      "language_loss": 0.71352506,
      "learning_rate": 5.16284511104263e-07,
      "loss": 0.73474687,
      "num_input_tokens_seen": 138350020,
      "step": 6429,
      "time_per_iteration": 2.814375400543213
    },
    {
      "auxiliary_loss_clip": 0.01099108,
      "auxiliary_loss_mlp": 0.01024815,
      "balance_loss_clip": 1.0410769,
      "balance_loss_mlp": 1.01680088,
      "epoch": 0.7731617868093549,
      "flos": 11947480940160.0,
      "grad_norm": 2.4899941081255434,
      "language_loss": 0.8064115,
      "learning_rate": 5.157622785353457e-07,
      "loss": 0.82765073,
      "num_input_tokens_seen": 138368135,
      "step": 6430,
      "time_per_iteration": 2.7714390754699707
    },
    {
      "auxiliary_loss_clip": 0.01045423,
      "auxiliary_loss_mlp": 0.01005432,
      "balance_loss_clip": 1.01145852,
      "balance_loss_mlp": 1.00416875,
      "epoch": 0.7732820296999939,
      "flos": 64201027069440.0,
      "grad_norm": 0.6436954291570365,
      "language_loss": 0.60369253,
      "learning_rate": 5.152402711241113e-07,
      "loss": 0.62420118,
      "num_input_tokens_seen": 138436040,
      "step": 6431,
      "time_per_iteration": 3.3363234996795654
    },
    {
      "auxiliary_loss_clip": 0.011352,
      "auxiliary_loss_mlp": 0.01021764,
      "balance_loss_clip": 1.04260254,
      "balance_loss_mlp": 1.01441789,
      "epoch": 0.7734022725906331,
      "flos": 25302191984640.0,
      "grad_norm": 1.9627439669160547,
      "language_loss": 0.83161575,
      "learning_rate": 5.147184889497465e-07,
      "loss": 0.85318536,
      "num_input_tokens_seen": 138455510,
      "step": 6432,
      "time_per_iteration": 2.8767309188842773
    },
    {
      "auxiliary_loss_clip": 0.01107401,
      "auxiliary_loss_mlp": 0.01024585,
      "balance_loss_clip": 1.04258382,
      "balance_loss_mlp": 1.01635313,
      "epoch": 0.7735225154812722,
      "flos": 17347440067200.0,
      "grad_norm": 2.169959911902001,
      "language_loss": 0.79802012,
      "learning_rate": 5.141969320914072e-07,
      "loss": 0.81933999,
      "num_input_tokens_seen": 138473015,
      "step": 6433,
      "time_per_iteration": 2.6987497806549072
    },
    {
      "auxiliary_loss_clip": 0.01105388,
      "auxiliary_loss_mlp": 0.01034002,
      "balance_loss_clip": 1.04482925,
      "balance_loss_mlp": 1.02493072,
      "epoch": 0.7736427583719112,
      "flos": 32630086725120.0,
      "grad_norm": 2.4796687259624197,
      "language_loss": 0.62514675,
      "learning_rate": 5.136756006282113e-07,
      "loss": 0.64654064,
      "num_input_tokens_seen": 138491680,
      "step": 6434,
      "time_per_iteration": 2.870701789855957
    },
    {
      "auxiliary_loss_clip": 0.01161514,
      "auxiliary_loss_mlp": 0.01029007,
      "balance_loss_clip": 1.04705656,
      "balance_loss_mlp": 1.0215621,
      "epoch": 0.7737630012625504,
      "flos": 19860073269120.0,
      "grad_norm": 17.95124883894829,
      "language_loss": 0.85229516,
      "learning_rate": 5.131544946392446e-07,
      "loss": 0.87420034,
      "num_input_tokens_seen": 138506960,
      "step": 6435,
      "time_per_iteration": 2.599607467651367
    },
    {
      "auxiliary_loss_clip": 0.01122782,
      "auxiliary_loss_mlp": 0.01025358,
      "balance_loss_clip": 1.04360867,
      "balance_loss_mlp": 1.01746678,
      "epoch": 0.7738832441531894,
      "flos": 36022639397760.0,
      "grad_norm": 2.0782771983521613,
      "language_loss": 0.63947493,
      "learning_rate": 5.126336142035592e-07,
      "loss": 0.66095632,
      "num_input_tokens_seen": 138526995,
      "step": 6436,
      "time_per_iteration": 2.892214298248291
    },
    {
      "auxiliary_loss_clip": 0.01153141,
      "auxiliary_loss_mlp": 0.01020311,
      "balance_loss_clip": 1.04529226,
      "balance_loss_mlp": 1.01244271,
      "epoch": 0.7740034870438285,
      "flos": 13405274415360.0,
      "grad_norm": 2.3993729745470955,
      "language_loss": 0.7257421,
      "learning_rate": 5.121129594001721e-07,
      "loss": 0.74747658,
      "num_input_tokens_seen": 138541260,
      "step": 6437,
      "time_per_iteration": 4.198429584503174
    },
    {
      "auxiliary_loss_clip": 0.01144898,
      "auxiliary_loss_mlp": 0.01028326,
      "balance_loss_clip": 1.04500651,
      "balance_loss_mlp": 1.02041674,
      "epoch": 0.7741237299344677,
      "flos": 22086714384000.0,
      "grad_norm": 1.6738099015134982,
      "language_loss": 0.81344622,
      "learning_rate": 5.115925303080661e-07,
      "loss": 0.83517849,
      "num_input_tokens_seen": 138560970,
      "step": 6438,
      "time_per_iteration": 2.7367637157440186
    },
    {
      "auxiliary_loss_clip": 0.01140727,
      "auxiliary_loss_mlp": 0.01024484,
      "balance_loss_clip": 1.04456663,
      "balance_loss_mlp": 1.01691449,
      "epoch": 0.7742439728251067,
      "flos": 19864777950720.0,
      "grad_norm": 2.2330709681741214,
      "language_loss": 0.79460323,
      "learning_rate": 5.110723270061899e-07,
      "loss": 0.81625533,
      "num_input_tokens_seen": 138577460,
      "step": 6439,
      "time_per_iteration": 2.7195944786071777
    },
    {
      "auxiliary_loss_clip": 0.01139118,
      "auxiliary_loss_mlp": 0.01028283,
      "balance_loss_clip": 1.04299963,
      "balance_loss_mlp": 1.02054989,
      "epoch": 0.7743642157157458,
      "flos": 16690167048960.0,
      "grad_norm": 2.2881119450492022,
      "language_loss": 0.79556173,
      "learning_rate": 5.105523495734572e-07,
      "loss": 0.81723571,
      "num_input_tokens_seen": 138594860,
      "step": 6440,
      "time_per_iteration": 2.6711313724517822
    },
    {
      "auxiliary_loss_clip": 0.01135182,
      "auxiliary_loss_mlp": 0.01022923,
      "balance_loss_clip": 1.04202425,
      "balance_loss_mlp": 1.01521575,
      "epoch": 0.7744844586063849,
      "flos": 20304360593280.0,
      "grad_norm": 1.620297219718392,
      "language_loss": 0.75428939,
      "learning_rate": 5.100325980887499e-07,
      "loss": 0.77587044,
      "num_input_tokens_seen": 138614785,
      "step": 6441,
      "time_per_iteration": 2.7073311805725098
    },
    {
      "auxiliary_loss_clip": 0.01092738,
      "auxiliary_loss_mlp": 0.0102767,
      "balance_loss_clip": 1.04425991,
      "balance_loss_mlp": 1.02018654,
      "epoch": 0.774604701497024,
      "flos": 22966705681920.0,
      "grad_norm": 1.7637722914532352,
      "language_loss": 0.82967365,
      "learning_rate": 5.095130726309116e-07,
      "loss": 0.8508777,
      "num_input_tokens_seen": 138634960,
      "step": 6442,
      "time_per_iteration": 2.760009765625
    },
    {
      "auxiliary_loss_clip": 0.01051812,
      "auxiliary_loss_mlp": 0.00753831,
      "balance_loss_clip": 1.01081419,
      "balance_loss_mlp": 1.00093019,
      "epoch": 0.774724944387663,
      "flos": 60288523073280.0,
      "grad_norm": 0.7902977596048895,
      "language_loss": 0.58988297,
      "learning_rate": 5.089937732787559e-07,
      "loss": 0.60793936,
      "num_input_tokens_seen": 138699520,
      "step": 6443,
      "time_per_iteration": 3.3469061851501465
    },
    {
      "auxiliary_loss_clip": 0.01156658,
      "auxiliary_loss_mlp": 0.01031285,
      "balance_loss_clip": 1.05040693,
      "balance_loss_mlp": 1.02202809,
      "epoch": 0.7748451872783022,
      "flos": 26761026954240.0,
      "grad_norm": 2.2316078316408903,
      "language_loss": 0.66677195,
      "learning_rate": 5.084747001110592e-07,
      "loss": 0.68865138,
      "num_input_tokens_seen": 138719145,
      "step": 6444,
      "time_per_iteration": 2.6818747520446777
    },
    {
      "auxiliary_loss_clip": 0.01167503,
      "auxiliary_loss_mlp": 0.00762353,
      "balance_loss_clip": 1.04723132,
      "balance_loss_mlp": 1.00042725,
      "epoch": 0.7749654301689413,
      "flos": 30338627518080.0,
      "grad_norm": 1.8430454372314002,
      "language_loss": 0.70352066,
      "learning_rate": 5.07955853206564e-07,
      "loss": 0.72281915,
      "num_input_tokens_seen": 138743850,
      "step": 6445,
      "time_per_iteration": 2.6746227741241455
    },
    {
      "auxiliary_loss_clip": 0.01145545,
      "auxiliary_loss_mlp": 0.01022018,
      "balance_loss_clip": 1.04474771,
      "balance_loss_mlp": 1.0138818,
      "epoch": 0.7750856730595803,
      "flos": 43179851687040.0,
      "grad_norm": 1.8126621097046736,
      "language_loss": 0.70921421,
      "learning_rate": 5.074372326439807e-07,
      "loss": 0.73088986,
      "num_input_tokens_seen": 138766860,
      "step": 6446,
      "time_per_iteration": 2.8531250953674316
    },
    {
      "auxiliary_loss_clip": 0.01140075,
      "auxiliary_loss_mlp": 0.01031024,
      "balance_loss_clip": 1.04401815,
      "balance_loss_mlp": 1.02313828,
      "epoch": 0.7752059159502195,
      "flos": 17640040256640.0,
      "grad_norm": 2.406753720524923,
      "language_loss": 0.73794603,
      "learning_rate": 5.069188385019814e-07,
      "loss": 0.75965703,
      "num_input_tokens_seen": 138784560,
      "step": 6447,
      "time_per_iteration": 2.6376748085021973
    },
    {
      "auxiliary_loss_clip": 0.01157409,
      "auxiliary_loss_mlp": 0.01025357,
      "balance_loss_clip": 1.04744816,
      "balance_loss_mlp": 1.01653516,
      "epoch": 0.7753261588408585,
      "flos": 12677688524160.0,
      "grad_norm": 2.7313075878857447,
      "language_loss": 0.60972196,
      "learning_rate": 5.064006708592077e-07,
      "loss": 0.6315496,
      "num_input_tokens_seen": 138800805,
      "step": 6448,
      "time_per_iteration": 4.4366984367370605
    },
    {
      "auxiliary_loss_clip": 0.01097523,
      "auxiliary_loss_mlp": 0.0102185,
      "balance_loss_clip": 1.04286051,
      "balance_loss_mlp": 1.01491213,
      "epoch": 0.7754464017314976,
      "flos": 16690741666560.0,
      "grad_norm": 2.4516633742572145,
      "language_loss": 0.75920153,
      "learning_rate": 5.058827297942641e-07,
      "loss": 0.78039527,
      "num_input_tokens_seen": 138815910,
      "step": 6449,
      "time_per_iteration": 3.7217249870300293
    },
    {
      "auxiliary_loss_clip": 0.01103853,
      "auxiliary_loss_mlp": 0.01020758,
      "balance_loss_clip": 1.03949642,
      "balance_loss_mlp": 1.01349807,
      "epoch": 0.7755666446221368,
      "flos": 19718944732800.0,
      "grad_norm": 2.300598286239985,
      "language_loss": 0.75152481,
      "learning_rate": 5.053650153857237e-07,
      "loss": 0.772771,
      "num_input_tokens_seen": 138834920,
      "step": 6450,
      "time_per_iteration": 2.7482872009277344
    },
    {
      "auxiliary_loss_clip": 0.01157647,
      "auxiliary_loss_mlp": 0.01023324,
      "balance_loss_clip": 1.04591668,
      "balance_loss_mlp": 1.01626134,
      "epoch": 0.7756868875127758,
      "flos": 18693623007360.0,
      "grad_norm": 3.608164430434069,
      "language_loss": 0.69925404,
      "learning_rate": 5.048475277121214e-07,
      "loss": 0.72106373,
      "num_input_tokens_seen": 138852135,
      "step": 6451,
      "time_per_iteration": 2.7748801708221436
    },
    {
      "auxiliary_loss_clip": 0.01132466,
      "auxiliary_loss_mlp": 0.01026504,
      "balance_loss_clip": 1.04512191,
      "balance_loss_mlp": 1.0187552,
      "epoch": 0.7758071304034149,
      "flos": 28404191543040.0,
      "grad_norm": 2.2530849738681,
      "language_loss": 0.77119148,
      "learning_rate": 5.043302668519598e-07,
      "loss": 0.79278123,
      "num_input_tokens_seen": 138871470,
      "step": 6452,
      "time_per_iteration": 2.9755210876464844
    },
    {
      "auxiliary_loss_clip": 0.01104661,
      "auxiliary_loss_mlp": 0.01021796,
      "balance_loss_clip": 1.04173708,
      "balance_loss_mlp": 1.0137856,
      "epoch": 0.775927373294054,
      "flos": 20595344670720.0,
      "grad_norm": 2.03013835446101,
      "language_loss": 0.72174084,
      "learning_rate": 5.038132328837079e-07,
      "loss": 0.74300545,
      "num_input_tokens_seen": 138889860,
      "step": 6453,
      "time_per_iteration": 2.7716224193573
    },
    {
      "auxiliary_loss_clip": 0.01126049,
      "auxiliary_loss_mlp": 0.0102563,
      "balance_loss_clip": 1.0458504,
      "balance_loss_mlp": 1.01789951,
      "epoch": 0.7760476161846931,
      "flos": 22526368853760.0,
      "grad_norm": 2.277343152097135,
      "language_loss": 0.74071252,
      "learning_rate": 5.032964258857993e-07,
      "loss": 0.76222932,
      "num_input_tokens_seen": 138909955,
      "step": 6454,
      "time_per_iteration": 2.7774739265441895
    },
    {
      "auxiliary_loss_clip": 0.01130272,
      "auxiliary_loss_mlp": 0.01025416,
      "balance_loss_clip": 1.04326868,
      "balance_loss_mlp": 1.01752985,
      "epoch": 0.7761678590753321,
      "flos": 48651488403840.0,
      "grad_norm": 1.7998563538070935,
      "language_loss": 0.68483192,
      "learning_rate": 5.027798459366329e-07,
      "loss": 0.70638883,
      "num_input_tokens_seen": 138935320,
      "step": 6455,
      "time_per_iteration": 2.921630382537842
    },
    {
      "auxiliary_loss_clip": 0.01156994,
      "auxiliary_loss_mlp": 0.01024746,
      "balance_loss_clip": 1.04459214,
      "balance_loss_mlp": 1.01661015,
      "epoch": 0.7762881019659713,
      "flos": 26177047637760.0,
      "grad_norm": 1.4854553602188607,
      "language_loss": 0.635625,
      "learning_rate": 5.02263493114573e-07,
      "loss": 0.65744233,
      "num_input_tokens_seen": 138957115,
      "step": 6456,
      "time_per_iteration": 2.7497074604034424
    },
    {
      "auxiliary_loss_clip": 0.01114741,
      "auxiliary_loss_mlp": 0.01027462,
      "balance_loss_clip": 1.03887129,
      "balance_loss_mlp": 1.01887345,
      "epoch": 0.7764083448566104,
      "flos": 20588341518720.0,
      "grad_norm": 27.4980776119751,
      "language_loss": 0.77471536,
      "learning_rate": 5.017473674979502e-07,
      "loss": 0.79613739,
      "num_input_tokens_seen": 138973140,
      "step": 6457,
      "time_per_iteration": 2.689283609390259
    },
    {
      "auxiliary_loss_clip": 0.01034061,
      "auxiliary_loss_mlp": 0.01001643,
      "balance_loss_clip": 1.0140698,
      "balance_loss_mlp": 1.00012898,
      "epoch": 0.7765285877472494,
      "flos": 67293078560640.0,
      "grad_norm": 0.7436530998016256,
      "language_loss": 0.58331221,
      "learning_rate": 5.01231469165061e-07,
      "loss": 0.60366929,
      "num_input_tokens_seen": 139028965,
      "step": 6458,
      "time_per_iteration": 3.189532995223999
    },
    {
      "auxiliary_loss_clip": 0.01009367,
      "auxiliary_loss_mlp": 0.01003021,
      "balance_loss_clip": 1.0253973,
      "balance_loss_mlp": 1.00186431,
      "epoch": 0.7766488306378886,
      "flos": 61344476121600.0,
      "grad_norm": 0.8244244029304664,
      "language_loss": 0.56797814,
      "learning_rate": 5.007157981941663e-07,
      "loss": 0.58810198,
      "num_input_tokens_seen": 139094325,
      "step": 6459,
      "time_per_iteration": 3.5998611450195312
    },
    {
      "auxiliary_loss_clip": 0.01041358,
      "auxiliary_loss_mlp": 0.01003111,
      "balance_loss_clip": 1.01267624,
      "balance_loss_mlp": 1.00181723,
      "epoch": 0.7767690735285276,
      "flos": 62946199393920.0,
      "grad_norm": 0.8856158113617459,
      "language_loss": 0.67451131,
      "learning_rate": 5.002003546634928e-07,
      "loss": 0.69495595,
      "num_input_tokens_seen": 139150425,
      "step": 6460,
      "time_per_iteration": 3.5406534671783447
    },
    {
      "auxiliary_loss_clip": 0.01129991,
      "auxiliary_loss_mlp": 0.01027621,
      "balance_loss_clip": 1.04463065,
      "balance_loss_mlp": 1.02033734,
      "epoch": 0.7768893164191667,
      "flos": 20886400575360.0,
      "grad_norm": 1.8885130846977674,
      "language_loss": 0.76341462,
      "learning_rate": 4.996851386512331e-07,
      "loss": 0.78499067,
      "num_input_tokens_seen": 139169130,
      "step": 6461,
      "time_per_iteration": 2.784761667251587
    },
    {
      "auxiliary_loss_clip": 0.0115263,
      "auxiliary_loss_mlp": 0.0102198,
      "balance_loss_clip": 1.04475188,
      "balance_loss_mlp": 1.01407075,
      "epoch": 0.7770095593098058,
      "flos": 20704584908160.0,
      "grad_norm": 2.532936139999323,
      "language_loss": 0.82920659,
      "learning_rate": 4.991701502355444e-07,
      "loss": 0.85095263,
      "num_input_tokens_seen": 139189595,
      "step": 6462,
      "time_per_iteration": 2.6404201984405518
    },
    {
      "auxiliary_loss_clip": 0.01132679,
      "auxiliary_loss_mlp": 0.01023802,
      "balance_loss_clip": 1.04142261,
      "balance_loss_mlp": 1.01682782,
      "epoch": 0.7771298022004449,
      "flos": 24717709877760.0,
      "grad_norm": 1.724384987439989,
      "language_loss": 0.76149398,
      "learning_rate": 4.986553894945518e-07,
      "loss": 0.78305876,
      "num_input_tokens_seen": 139210805,
      "step": 6463,
      "time_per_iteration": 3.9524333477020264
    },
    {
      "auxiliary_loss_clip": 0.01115544,
      "auxiliary_loss_mlp": 0.01026151,
      "balance_loss_clip": 1.03846884,
      "balance_loss_mlp": 1.01888525,
      "epoch": 0.777250045091084,
      "flos": 25009232659200.0,
      "grad_norm": 2.002942161124168,
      "language_loss": 0.86031634,
      "learning_rate": 4.981408565063416e-07,
      "loss": 0.8817333,
      "num_input_tokens_seen": 139230750,
      "step": 6464,
      "time_per_iteration": 2.882211446762085
    },
    {
      "auxiliary_loss_clip": 0.0109669,
      "auxiliary_loss_mlp": 0.01024815,
      "balance_loss_clip": 1.04249287,
      "balance_loss_mlp": 1.01746571,
      "epoch": 0.777370287981723,
      "flos": 20119887319680.0,
      "grad_norm": 3.9450314485746087,
      "language_loss": 0.76022029,
      "learning_rate": 4.976265513489701e-07,
      "loss": 0.78143531,
      "num_input_tokens_seen": 139250720,
      "step": 6465,
      "time_per_iteration": 2.8159565925598145
    },
    {
      "auxiliary_loss_clip": 0.01041271,
      "auxiliary_loss_mlp": 0.01025569,
      "balance_loss_clip": 1.03201246,
      "balance_loss_mlp": 1.01755798,
      "epoch": 0.7774905308723622,
      "flos": 21718809331200.0,
      "grad_norm": 2.086788146490928,
      "language_loss": 0.80811471,
      "learning_rate": 4.971124741004562e-07,
      "loss": 0.82878309,
      "num_input_tokens_seen": 139269720,
      "step": 6466,
      "time_per_iteration": 2.8727645874023438
    },
    {
      "auxiliary_loss_clip": 0.0115439,
      "auxiliary_loss_mlp": 0.01023517,
      "balance_loss_clip": 1.0429939,
      "balance_loss_mlp": 1.0159297,
      "epoch": 0.7776107737630013,
      "flos": 16034115093120.0,
      "grad_norm": 1.9745641758054984,
      "language_loss": 0.76132303,
      "learning_rate": 4.965986248387846e-07,
      "loss": 0.7831021,
      "num_input_tokens_seen": 139288035,
      "step": 6467,
      "time_per_iteration": 2.666079521179199
    },
    {
      "auxiliary_loss_clip": 0.01112411,
      "auxiliary_loss_mlp": 0.01022614,
      "balance_loss_clip": 1.04201388,
      "balance_loss_mlp": 1.01493144,
      "epoch": 0.7777310166536403,
      "flos": 24790895838720.0,
      "grad_norm": 1.6733572552375038,
      "language_loss": 0.77455902,
      "learning_rate": 4.960850036419073e-07,
      "loss": 0.79590923,
      "num_input_tokens_seen": 139307135,
      "step": 6468,
      "time_per_iteration": 2.777951717376709
    },
    {
      "auxiliary_loss_clip": 0.01153551,
      "auxiliary_loss_mlp": 0.01024855,
      "balance_loss_clip": 1.04346764,
      "balance_loss_mlp": 1.01732445,
      "epoch": 0.7778512595442795,
      "flos": 17272530253440.0,
      "grad_norm": 2.304167864249859,
      "language_loss": 0.78990448,
      "learning_rate": 4.955716105877378e-07,
      "loss": 0.81168854,
      "num_input_tokens_seen": 139325905,
      "step": 6469,
      "time_per_iteration": 2.6766517162323
    },
    {
      "auxiliary_loss_clip": 0.01124086,
      "auxiliary_loss_mlp": 0.00762114,
      "balance_loss_clip": 1.04231763,
      "balance_loss_mlp": 1.00044143,
      "epoch": 0.7779715024349185,
      "flos": 17748418567680.0,
      "grad_norm": 1.8622896719666586,
      "language_loss": 0.83193624,
      "learning_rate": 4.950584457541598e-07,
      "loss": 0.85079825,
      "num_input_tokens_seen": 139344370,
      "step": 6470,
      "time_per_iteration": 2.7301688194274902
    },
    {
      "auxiliary_loss_clip": 0.01094065,
      "auxiliary_loss_mlp": 0.01027527,
      "balance_loss_clip": 1.03899133,
      "balance_loss_mlp": 1.01908362,
      "epoch": 0.7780917453255576,
      "flos": 24316875031680.0,
      "grad_norm": 2.4513483969553365,
      "language_loss": 0.82130444,
      "learning_rate": 4.945455092190183e-07,
      "loss": 0.84252036,
      "num_input_tokens_seen": 139365625,
      "step": 6471,
      "time_per_iteration": 2.889512538909912
    },
    {
      "auxiliary_loss_clip": 0.01048448,
      "auxiliary_loss_mlp": 0.01001656,
      "balance_loss_clip": 1.01461577,
      "balance_loss_mlp": 1.00011861,
      "epoch": 0.7782119882161967,
      "flos": 56364601530240.0,
      "grad_norm": 0.6999909855906273,
      "language_loss": 0.55974078,
      "learning_rate": 4.940328010601271e-07,
      "loss": 0.5802418,
      "num_input_tokens_seen": 139430540,
      "step": 6472,
      "time_per_iteration": 3.3854074478149414
    },
    {
      "auxiliary_loss_clip": 0.01151557,
      "auxiliary_loss_mlp": 0.01030975,
      "balance_loss_clip": 1.05023837,
      "balance_loss_mlp": 1.02242219,
      "epoch": 0.7783322311068358,
      "flos": 46789986994560.0,
      "grad_norm": 2.110160255847476,
      "language_loss": 0.76926547,
      "learning_rate": 4.935203213552621e-07,
      "loss": 0.79109085,
      "num_input_tokens_seen": 139454280,
      "step": 6473,
      "time_per_iteration": 4.152997732162476
    },
    {
      "auxiliary_loss_clip": 0.01112574,
      "auxiliary_loss_mlp": 0.01025688,
      "balance_loss_clip": 1.04193819,
      "balance_loss_mlp": 1.01763558,
      "epoch": 0.7784524739974749,
      "flos": 19057864872960.0,
      "grad_norm": 2.424686524630094,
      "language_loss": 0.668805,
      "learning_rate": 4.930080701821662e-07,
      "loss": 0.69018763,
      "num_input_tokens_seen": 139471745,
      "step": 6474,
      "time_per_iteration": 3.701076030731201
    },
    {
      "auxiliary_loss_clip": 0.01145893,
      "auxiliary_loss_mlp": 0.01027328,
      "balance_loss_clip": 1.04320621,
      "balance_loss_mlp": 1.01897764,
      "epoch": 0.778572716888114,
      "flos": 24791111320320.0,
      "grad_norm": 3.340400793623708,
      "language_loss": 0.7752744,
      "learning_rate": 4.92496047618548e-07,
      "loss": 0.79700661,
      "num_input_tokens_seen": 139491505,
      "step": 6475,
      "time_per_iteration": 3.927767753601074
    },
    {
      "auxiliary_loss_clip": 0.01171959,
      "auxiliary_loss_mlp": 0.01022698,
      "balance_loss_clip": 1.04870558,
      "balance_loss_mlp": 1.01491916,
      "epoch": 0.7786929597787531,
      "flos": 20078086867200.0,
      "grad_norm": 2.6289807190693786,
      "language_loss": 0.77547014,
      "learning_rate": 4.919842537420811e-07,
      "loss": 0.79741669,
      "num_input_tokens_seen": 139508620,
      "step": 6476,
      "time_per_iteration": 2.5928261280059814
    },
    {
      "auxiliary_loss_clip": 0.01134351,
      "auxiliary_loss_mlp": 0.01023258,
      "balance_loss_clip": 1.04852366,
      "balance_loss_mlp": 1.01584625,
      "epoch": 0.7788132026693921,
      "flos": 21872220318720.0,
      "grad_norm": 1.6333933880040161,
      "language_loss": 0.79406321,
      "learning_rate": 4.91472688630404e-07,
      "loss": 0.81563926,
      "num_input_tokens_seen": 139529360,
      "step": 6477,
      "time_per_iteration": 2.7841060161590576
    },
    {
      "auxiliary_loss_clip": 0.01117846,
      "auxiliary_loss_mlp": 0.01026725,
      "balance_loss_clip": 1.03941321,
      "balance_loss_mlp": 1.01962614,
      "epoch": 0.7789334455600313,
      "flos": 11181937351680.0,
      "grad_norm": 2.081396232899467,
      "language_loss": 0.74250233,
      "learning_rate": 4.909613523611202e-07,
      "loss": 0.76394808,
      "num_input_tokens_seen": 139546240,
      "step": 6478,
      "time_per_iteration": 2.6526498794555664
    },
    {
      "auxiliary_loss_clip": 0.01131616,
      "auxiliary_loss_mlp": 0.01024795,
      "balance_loss_clip": 1.04187,
      "balance_loss_mlp": 1.01718938,
      "epoch": 0.7790536884506704,
      "flos": 28695427015680.0,
      "grad_norm": 2.632095885084147,
      "language_loss": 0.74789101,
      "learning_rate": 4.904502450117991e-07,
      "loss": 0.76945508,
      "num_input_tokens_seen": 139567200,
      "step": 6479,
      "time_per_iteration": 2.762993812561035
    },
    {
      "auxiliary_loss_clip": 0.01125645,
      "auxiliary_loss_mlp": 0.01023555,
      "balance_loss_clip": 1.04427719,
      "balance_loss_mlp": 1.01609278,
      "epoch": 0.7791739313413094,
      "flos": 11072302064640.0,
      "grad_norm": 3.059665978345944,
      "language_loss": 0.72500527,
      "learning_rate": 4.899393666599762e-07,
      "loss": 0.74649727,
      "num_input_tokens_seen": 139583775,
      "step": 6480,
      "time_per_iteration": 2.7937018871307373
    },
    {
      "auxiliary_loss_clip": 0.0110164,
      "auxiliary_loss_mlp": 0.01020954,
      "balance_loss_clip": 1.03856015,
      "balance_loss_mlp": 1.01358151,
      "epoch": 0.7792941742319486,
      "flos": 14679276975360.0,
      "grad_norm": 7.2645918687229925,
      "language_loss": 0.72742069,
      "learning_rate": 4.894287173831506e-07,
      "loss": 0.74864662,
      "num_input_tokens_seen": 139599735,
      "step": 6481,
      "time_per_iteration": 2.7225499153137207
    },
    {
      "auxiliary_loss_clip": 0.01105305,
      "auxiliary_loss_mlp": 0.01026535,
      "balance_loss_clip": 1.04053259,
      "balance_loss_mlp": 1.01814294,
      "epoch": 0.7794144171225876,
      "flos": 23258874908160.0,
      "grad_norm": 2.1721498670408086,
      "language_loss": 0.84559983,
      "learning_rate": 4.889182972587877e-07,
      "loss": 0.86691821,
      "num_input_tokens_seen": 139619030,
      "step": 6482,
      "time_per_iteration": 2.8259851932525635
    },
    {
      "auxiliary_loss_clip": 0.01091102,
      "auxiliary_loss_mlp": 0.01029592,
      "balance_loss_clip": 1.04836237,
      "balance_loss_mlp": 1.02174771,
      "epoch": 0.7795346600132267,
      "flos": 21507080613120.0,
      "grad_norm": 1.9208358972563666,
      "language_loss": 0.6639353,
      "learning_rate": 4.884081063643177e-07,
      "loss": 0.68514222,
      "num_input_tokens_seen": 139637690,
      "step": 6483,
      "time_per_iteration": 2.787053108215332
    },
    {
      "auxiliary_loss_clip": 0.01013224,
      "auxiliary_loss_mlp": 0.01001521,
      "balance_loss_clip": 1.01771212,
      "balance_loss_mlp": 0.99962586,
      "epoch": 0.7796549029038659,
      "flos": 70052273694720.0,
      "grad_norm": 0.8525849136998287,
      "language_loss": 0.52441227,
      "learning_rate": 4.878981447771353e-07,
      "loss": 0.54455972,
      "num_input_tokens_seen": 139692070,
      "step": 6484,
      "time_per_iteration": 3.4350955486297607
    },
    {
      "auxiliary_loss_clip": 0.01146145,
      "auxiliary_loss_mlp": 0.01023282,
      "balance_loss_clip": 1.04551744,
      "balance_loss_mlp": 1.01531005,
      "epoch": 0.7797751457945049,
      "flos": 23989405714560.0,
      "grad_norm": 2.129181844941264,
      "language_loss": 0.73085511,
      "learning_rate": 4.873884125746035e-07,
      "loss": 0.75254941,
      "num_input_tokens_seen": 139713745,
      "step": 6485,
      "time_per_iteration": 3.247326374053955
    },
    {
      "auxiliary_loss_clip": 0.01112111,
      "auxiliary_loss_mlp": 0.01023482,
      "balance_loss_clip": 1.04301643,
      "balance_loss_mlp": 1.01619196,
      "epoch": 0.779895388685144,
      "flos": 22674751937280.0,
      "grad_norm": 2.182476189481371,
      "language_loss": 0.71991533,
      "learning_rate": 4.868789098340456e-07,
      "loss": 0.74127126,
      "num_input_tokens_seen": 139731650,
      "step": 6486,
      "time_per_iteration": 2.789104700088501
    },
    {
      "auxiliary_loss_clip": 0.0115533,
      "auxiliary_loss_mlp": 0.01024809,
      "balance_loss_clip": 1.04691589,
      "balance_loss_mlp": 1.01681626,
      "epoch": 0.7800156315757831,
      "flos": 23768698596480.0,
      "grad_norm": 2.2626448823267262,
      "language_loss": 0.73638523,
      "learning_rate": 4.863696366327543e-07,
      "loss": 0.75818658,
      "num_input_tokens_seen": 139750820,
      "step": 6487,
      "time_per_iteration": 2.6815755367279053
    },
    {
      "auxiliary_loss_clip": 0.01144089,
      "auxiliary_loss_mlp": 0.01024323,
      "balance_loss_clip": 1.04385805,
      "balance_loss_mlp": 1.01675296,
      "epoch": 0.7801358744664222,
      "flos": 26429714881920.0,
      "grad_norm": 1.8312145807327072,
      "language_loss": 0.77873456,
      "learning_rate": 4.85860593047986e-07,
      "loss": 0.80041867,
      "num_input_tokens_seen": 139770885,
      "step": 6488,
      "time_per_iteration": 2.712266206741333
    },
    {
      "auxiliary_loss_clip": 0.01079959,
      "auxiliary_loss_mlp": 0.0102568,
      "balance_loss_clip": 1.03540516,
      "balance_loss_mlp": 1.01825356,
      "epoch": 0.7802561173570612,
      "flos": 26322162583680.0,
      "grad_norm": 1.6679497108765897,
      "language_loss": 0.74818647,
      "learning_rate": 4.853517791569613e-07,
      "loss": 0.76924282,
      "num_input_tokens_seen": 139793065,
      "step": 6489,
      "time_per_iteration": 3.7390947341918945
    },
    {
      "auxiliary_loss_clip": 0.01106504,
      "auxiliary_loss_mlp": 0.01024433,
      "balance_loss_clip": 1.03827548,
      "balance_loss_mlp": 1.01634479,
      "epoch": 0.7803763602477004,
      "flos": 40333751596800.0,
      "grad_norm": 1.951981443251654,
      "language_loss": 0.65806562,
      "learning_rate": 4.848431950368684e-07,
      "loss": 0.67937499,
      "num_input_tokens_seen": 139815625,
      "step": 6490,
      "time_per_iteration": 2.9453818798065186
    },
    {
      "auxiliary_loss_clip": 0.0104625,
      "auxiliary_loss_mlp": 0.01002593,
      "balance_loss_clip": 1.01776803,
      "balance_loss_mlp": 1.00082898,
      "epoch": 0.7804966031383395,
      "flos": 67001448038400.0,
      "grad_norm": 0.7138557558141608,
      "language_loss": 0.55748355,
      "learning_rate": 4.843348407648569e-07,
      "loss": 0.57797199,
      "num_input_tokens_seen": 139876905,
      "step": 6491,
      "time_per_iteration": 3.262885093688965
    },
    {
      "auxiliary_loss_clip": 0.01159522,
      "auxiliary_loss_mlp": 0.01027818,
      "balance_loss_clip": 1.04411459,
      "balance_loss_mlp": 1.02037346,
      "epoch": 0.7806168460289785,
      "flos": 17740733057280.0,
      "grad_norm": 5.419833591182873,
      "language_loss": 0.83200443,
      "learning_rate": 4.838267164180457e-07,
      "loss": 0.85387778,
      "num_input_tokens_seen": 139892575,
      "step": 6492,
      "time_per_iteration": 2.6948940753936768
    },
    {
      "auxiliary_loss_clip": 0.01147187,
      "auxiliary_loss_mlp": 0.01027726,
      "balance_loss_clip": 1.04652762,
      "balance_loss_mlp": 1.019912,
      "epoch": 0.7807370889196176,
      "flos": 23946240545280.0,
      "grad_norm": 2.185801955669936,
      "language_loss": 0.83858025,
      "learning_rate": 4.833188220735156e-07,
      "loss": 0.86032939,
      "num_input_tokens_seen": 139912245,
      "step": 6493,
      "time_per_iteration": 2.8032009601593018
    },
    {
      "auxiliary_loss_clip": 0.01119901,
      "auxiliary_loss_mlp": 0.0076206,
      "balance_loss_clip": 1.04275084,
      "balance_loss_mlp": 1.00040889,
      "epoch": 0.7808573318102567,
      "flos": 18989024457600.0,
      "grad_norm": 2.0258948743071703,
      "language_loss": 0.74666446,
      "learning_rate": 4.828111578083152e-07,
      "loss": 0.76548409,
      "num_input_tokens_seen": 139929150,
      "step": 6494,
      "time_per_iteration": 2.7746191024780273
    },
    {
      "auxiliary_loss_clip": 0.011374,
      "auxiliary_loss_mlp": 0.01028997,
      "balance_loss_clip": 1.04591298,
      "balance_loss_mlp": 1.02105153,
      "epoch": 0.7809775747008958,
      "flos": 23980750536960.0,
      "grad_norm": 1.975505341232043,
      "language_loss": 0.81745392,
      "learning_rate": 4.823037236994556e-07,
      "loss": 0.83911788,
      "num_input_tokens_seen": 139947315,
      "step": 6495,
      "time_per_iteration": 2.787749767303467
    },
    {
      "auxiliary_loss_clip": 0.01021815,
      "auxiliary_loss_mlp": 0.01002834,
      "balance_loss_clip": 1.01542163,
      "balance_loss_mlp": 1.00146866,
      "epoch": 0.7810978175915348,
      "flos": 68535875180160.0,
      "grad_norm": 0.7176951448560099,
      "language_loss": 0.56291109,
      "learning_rate": 4.817965198239136e-07,
      "loss": 0.5831576,
      "num_input_tokens_seen": 140013775,
      "step": 6496,
      "time_per_iteration": 3.4326069355010986
    },
    {
      "auxiliary_loss_clip": 0.01126517,
      "auxiliary_loss_mlp": 0.01024984,
      "balance_loss_clip": 1.04093122,
      "balance_loss_mlp": 1.01763511,
      "epoch": 0.781218060482174,
      "flos": 19642131498240.0,
      "grad_norm": 7.426187332935483,
      "language_loss": 0.74379718,
      "learning_rate": 4.812895462586331e-07,
      "loss": 0.76531219,
      "num_input_tokens_seen": 140031600,
      "step": 6497,
      "time_per_iteration": 2.689439296722412
    },
    {
      "auxiliary_loss_clip": 0.01126703,
      "auxiliary_loss_mlp": 0.00762047,
      "balance_loss_clip": 1.04192543,
      "balance_loss_mlp": 1.00047195,
      "epoch": 0.7813383033728131,
      "flos": 25627865621760.0,
      "grad_norm": 1.7142783823427805,
      "language_loss": 0.82220852,
      "learning_rate": 4.807828030805207e-07,
      "loss": 0.84109598,
      "num_input_tokens_seen": 140050590,
      "step": 6498,
      "time_per_iteration": 3.663648843765259
    },
    {
      "auxiliary_loss_clip": 0.01140436,
      "auxiliary_loss_mlp": 0.01027766,
      "balance_loss_clip": 1.04378033,
      "balance_loss_mlp": 1.02027953,
      "epoch": 0.7814585462634521,
      "flos": 20485924865280.0,
      "grad_norm": 2.092457156149274,
      "language_loss": 0.6842978,
      "learning_rate": 4.802762903664495e-07,
      "loss": 0.70597982,
      "num_input_tokens_seen": 140069770,
      "step": 6499,
      "time_per_iteration": 2.709014654159546
    },
    {
      "auxiliary_loss_clip": 0.01100322,
      "auxiliary_loss_mlp": 0.01029327,
      "balance_loss_clip": 1.0411576,
      "balance_loss_mlp": 1.02107215,
      "epoch": 0.7815787891540913,
      "flos": 22304297018880.0,
      "grad_norm": 4.3685063629193195,
      "language_loss": 0.74564987,
      "learning_rate": 4.797700081932565e-07,
      "loss": 0.76694632,
      "num_input_tokens_seen": 140087635,
      "step": 6500,
      "time_per_iteration": 3.735821008682251
    },
    {
      "auxiliary_loss_clip": 0.01105515,
      "auxiliary_loss_mlp": 0.0102596,
      "balance_loss_clip": 1.04288268,
      "balance_loss_mlp": 1.01834226,
      "epoch": 0.7816990320447303,
      "flos": 22600668136320.0,
      "grad_norm": 2.6804364109769248,
      "language_loss": 0.81986684,
      "learning_rate": 4.792639566377442e-07,
      "loss": 0.84118158,
      "num_input_tokens_seen": 140105045,
      "step": 6501,
      "time_per_iteration": 3.6193690299987793
    },
    {
      "auxiliary_loss_clip": 0.01091617,
      "auxiliary_loss_mlp": 0.01022182,
      "balance_loss_clip": 1.04299128,
      "balance_loss_mlp": 1.01442194,
      "epoch": 0.7818192749353694,
      "flos": 24935974871040.0,
      "grad_norm": 2.158872963901261,
      "language_loss": 0.7739222,
      "learning_rate": 4.78758135776681e-07,
      "loss": 0.79506016,
      "num_input_tokens_seen": 140124900,
      "step": 6502,
      "time_per_iteration": 2.890331983566284
    },
    {
      "auxiliary_loss_clip": 0.01136722,
      "auxiliary_loss_mlp": 0.01025054,
      "balance_loss_clip": 1.04379177,
      "balance_loss_mlp": 1.0171448,
      "epoch": 0.7819395178260086,
      "flos": 23733039369600.0,
      "grad_norm": 1.9879082172234315,
      "language_loss": 0.79265833,
      "learning_rate": 4.782525456867989e-07,
      "loss": 0.81427616,
      "num_input_tokens_seen": 140143755,
      "step": 6503,
      "time_per_iteration": 2.9467198848724365
    },
    {
      "auxiliary_loss_clip": 0.01111566,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.04133987,
      "balance_loss_mlp": 1.01696587,
      "epoch": 0.7820597607166476,
      "flos": 23221671396480.0,
      "grad_norm": 1.6793818544244554,
      "language_loss": 0.83580315,
      "learning_rate": 4.777471864447959e-07,
      "loss": 0.8571673,
      "num_input_tokens_seen": 140164495,
      "step": 6504,
      "time_per_iteration": 2.767273426055908
    },
    {
      "auxiliary_loss_clip": 0.01162089,
      "auxiliary_loss_mlp": 0.0102129,
      "balance_loss_clip": 1.0479461,
      "balance_loss_mlp": 1.01392889,
      "epoch": 0.7821800036072867,
      "flos": 22309540404480.0,
      "grad_norm": 2.0608430145318826,
      "language_loss": 0.80826604,
      "learning_rate": 4.772420581273344e-07,
      "loss": 0.83009982,
      "num_input_tokens_seen": 140181980,
      "step": 6505,
      "time_per_iteration": 2.770864963531494
    },
    {
      "auxiliary_loss_clip": 0.01145299,
      "auxiliary_loss_mlp": 0.00761874,
      "balance_loss_clip": 1.04484439,
      "balance_loss_mlp": 1.00048232,
      "epoch": 0.7823002464979258,
      "flos": 21544176384000.0,
      "grad_norm": 2.0102064158166986,
      "language_loss": 0.76266086,
      "learning_rate": 4.7673716081104134e-07,
      "loss": 0.78173256,
      "num_input_tokens_seen": 140202155,
      "step": 6506,
      "time_per_iteration": 2.6800360679626465
    },
    {
      "auxiliary_loss_clip": 0.01171611,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.04815602,
      "balance_loss_mlp": 1.01738882,
      "epoch": 0.7824204893885649,
      "flos": 24535642815360.0,
      "grad_norm": 2.038246747974726,
      "language_loss": 0.84593856,
      "learning_rate": 4.762324945725109e-07,
      "loss": 0.86790311,
      "num_input_tokens_seen": 140221600,
      "step": 6507,
      "time_per_iteration": 2.6476387977600098
    },
    {
      "auxiliary_loss_clip": 0.01171715,
      "auxiliary_loss_mlp": 0.00762065,
      "balance_loss_clip": 1.04954362,
      "balance_loss_mlp": 1.00055039,
      "epoch": 0.782540732279204,
      "flos": 27415211402880.0,
      "grad_norm": 1.8033351093642378,
      "language_loss": 0.75905669,
      "learning_rate": 4.7572805948829844e-07,
      "loss": 0.77839446,
      "num_input_tokens_seen": 140241860,
      "step": 6508,
      "time_per_iteration": 2.6580827236175537
    },
    {
      "auxiliary_loss_clip": 0.01132071,
      "auxiliary_loss_mlp": 0.01024223,
      "balance_loss_clip": 1.04101074,
      "balance_loss_mlp": 1.0164094,
      "epoch": 0.7826609751698431,
      "flos": 24353216616960.0,
      "grad_norm": 1.6102931902214344,
      "language_loss": 0.70820469,
      "learning_rate": 4.7522385563492795e-07,
      "loss": 0.72976768,
      "num_input_tokens_seen": 140262160,
      "step": 6509,
      "time_per_iteration": 2.735670328140259
    },
    {
      "auxiliary_loss_clip": 0.01159424,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 1.04875827,
      "balance_loss_mlp": 1.01743913,
      "epoch": 0.7827812180604822,
      "flos": 23988543788160.0,
      "grad_norm": 1.9966374267689597,
      "language_loss": 0.70430553,
      "learning_rate": 4.747198830888863e-07,
      "loss": 0.72615159,
      "num_input_tokens_seen": 140282030,
      "step": 6510,
      "time_per_iteration": 2.6564793586730957
    },
    {
      "auxiliary_loss_clip": 0.01158065,
      "auxiliary_loss_mlp": 0.00761717,
      "balance_loss_clip": 1.04639769,
      "balance_loss_mlp": 1.00046277,
      "epoch": 0.7829014609511212,
      "flos": 27454318335360.0,
      "grad_norm": 2.1015587302480547,
      "language_loss": 0.68396008,
      "learning_rate": 4.742161419266251e-07,
      "loss": 0.70315784,
      "num_input_tokens_seen": 140301190,
      "step": 6511,
      "time_per_iteration": 2.6981043815612793
    },
    {
      "auxiliary_loss_clip": 0.01114706,
      "auxiliary_loss_mlp": 0.00763049,
      "balance_loss_clip": 1.04354858,
      "balance_loss_mlp": 1.00060761,
      "epoch": 0.7830217038417604,
      "flos": 29204532432000.0,
      "grad_norm": 2.5389735232633868,
      "language_loss": 0.65489352,
      "learning_rate": 4.7371263222456304e-07,
      "loss": 0.67367113,
      "num_input_tokens_seen": 140318510,
      "step": 6512,
      "time_per_iteration": 2.722058057785034
    },
    {
      "auxiliary_loss_clip": 0.0107546,
      "auxiliary_loss_mlp": 0.0100522,
      "balance_loss_clip": 1.02304745,
      "balance_loss_mlp": 1.00314605,
      "epoch": 0.7831419467323995,
      "flos": 60950895822720.0,
      "grad_norm": 0.7974284812360275,
      "language_loss": 0.61374372,
      "learning_rate": 4.7320935405908004e-07,
      "loss": 0.63455057,
      "num_input_tokens_seen": 140379380,
      "step": 6513,
      "time_per_iteration": 3.1787941455841064
    },
    {
      "auxiliary_loss_clip": 0.01129362,
      "auxiliary_loss_mlp": 0.01028133,
      "balance_loss_clip": 1.04465294,
      "balance_loss_mlp": 1.02030134,
      "epoch": 0.7832621896230385,
      "flos": 19682531320320.0,
      "grad_norm": 2.6842788114194973,
      "language_loss": 0.84054011,
      "learning_rate": 4.7270630750652475e-07,
      "loss": 0.86211509,
      "num_input_tokens_seen": 140395335,
      "step": 6514,
      "time_per_iteration": 2.7258312702178955
    },
    {
      "auxiliary_loss_clip": 0.01121031,
      "auxiliary_loss_mlp": 0.01024795,
      "balance_loss_clip": 1.04259503,
      "balance_loss_mlp": 1.01731443,
      "epoch": 0.7833824325136777,
      "flos": 25009232659200.0,
      "grad_norm": 1.9967542464948236,
      "language_loss": 0.80576217,
      "learning_rate": 4.7220349264320746e-07,
      "loss": 0.82722044,
      "num_input_tokens_seen": 140414420,
      "step": 6515,
      "time_per_iteration": 3.656421422958374
    },
    {
      "auxiliary_loss_clip": 0.01076402,
      "auxiliary_loss_mlp": 0.01001427,
      "balance_loss_clip": 1.03317475,
      "balance_loss_mlp": 1.00045586,
      "epoch": 0.7835026754043167,
      "flos": 68800142517120.0,
      "grad_norm": 0.7372144084539117,
      "language_loss": 0.54855239,
      "learning_rate": 4.71700909545407e-07,
      "loss": 0.56933069,
      "num_input_tokens_seen": 140477365,
      "step": 6516,
      "time_per_iteration": 3.252194881439209
    },
    {
      "auxiliary_loss_clip": 0.01117382,
      "auxiliary_loss_mlp": 0.00762233,
      "balance_loss_clip": 1.03942132,
      "balance_loss_mlp": 1.00056124,
      "epoch": 0.7836229182949558,
      "flos": 19864598382720.0,
      "grad_norm": 2.0224982926764836,
      "language_loss": 0.77212065,
      "learning_rate": 4.711985582893627e-07,
      "loss": 0.7909168,
      "num_input_tokens_seen": 140495885,
      "step": 6517,
      "time_per_iteration": 2.7101778984069824
    },
    {
      "auxiliary_loss_clip": 0.01162178,
      "auxiliary_loss_mlp": 0.01027635,
      "balance_loss_clip": 1.04676068,
      "balance_loss_mlp": 1.01989245,
      "epoch": 0.783743161185595,
      "flos": 22965843755520.0,
      "grad_norm": 1.7220472166999135,
      "language_loss": 0.71751463,
      "learning_rate": 4.706964389512811e-07,
      "loss": 0.73941278,
      "num_input_tokens_seen": 140515920,
      "step": 6518,
      "time_per_iteration": 2.6879804134368896
    },
    {
      "auxiliary_loss_clip": 0.01131925,
      "auxiliary_loss_mlp": 0.00762087,
      "balance_loss_clip": 1.04521036,
      "balance_loss_mlp": 1.00054348,
      "epoch": 0.783863404076234,
      "flos": 12458489777280.0,
      "grad_norm": 1.8318330500964564,
      "language_loss": 0.8747046,
      "learning_rate": 4.701945516073345e-07,
      "loss": 0.89364469,
      "num_input_tokens_seen": 140533395,
      "step": 6519,
      "time_per_iteration": 2.6425914764404297
    },
    {
      "auxiliary_loss_clip": 0.01115404,
      "auxiliary_loss_mlp": 0.010293,
      "balance_loss_clip": 1.04124868,
      "balance_loss_mlp": 1.02179313,
      "epoch": 0.7839836469668731,
      "flos": 24243940465920.0,
      "grad_norm": 2.0588820988163596,
      "language_loss": 0.75482583,
      "learning_rate": 4.696928963336577e-07,
      "loss": 0.77627289,
      "num_input_tokens_seen": 140552825,
      "step": 6520,
      "time_per_iteration": 2.807187080383301
    },
    {
      "auxiliary_loss_clip": 0.01067952,
      "auxiliary_loss_mlp": 0.00999467,
      "balance_loss_clip": 1.03398037,
      "balance_loss_mlp": 0.99841785,
      "epoch": 0.7841038898575122,
      "flos": 62121978938880.0,
      "grad_norm": 0.8523561909262458,
      "language_loss": 0.60960948,
      "learning_rate": 4.6919147320635224e-07,
      "loss": 0.63028371,
      "num_input_tokens_seen": 140615535,
      "step": 6521,
      "time_per_iteration": 3.205608606338501
    },
    {
      "auxiliary_loss_clip": 0.01157711,
      "auxiliary_loss_mlp": 0.01029888,
      "balance_loss_clip": 1.04661036,
      "balance_loss_mlp": 1.02286613,
      "epoch": 0.7842241327481513,
      "flos": 20193899293440.0,
      "grad_norm": 2.2875167787874884,
      "language_loss": 0.73191851,
      "learning_rate": 4.6869028230148286e-07,
      "loss": 0.75379449,
      "num_input_tokens_seen": 140633330,
      "step": 6522,
      "time_per_iteration": 2.622668504714966
    },
    {
      "auxiliary_loss_clip": 0.01106917,
      "auxiliary_loss_mlp": 0.01026656,
      "balance_loss_clip": 1.03987575,
      "balance_loss_mlp": 1.0188446,
      "epoch": 0.7843443756387903,
      "flos": 28074531496320.0,
      "grad_norm": 2.721294044905143,
      "language_loss": 0.6008544,
      "learning_rate": 4.6818932369507957e-07,
      "loss": 0.62219012,
      "num_input_tokens_seen": 140652830,
      "step": 6523,
      "time_per_iteration": 2.802543878555298
    },
    {
      "auxiliary_loss_clip": 0.01065477,
      "auxiliary_loss_mlp": 0.01032599,
      "balance_loss_clip": 1.0361228,
      "balance_loss_mlp": 1.02528286,
      "epoch": 0.7844646185294295,
      "flos": 21323397438720.0,
      "grad_norm": 2.1762316212786033,
      "language_loss": 0.89108133,
      "learning_rate": 4.676885974631386e-07,
      "loss": 0.91206211,
      "num_input_tokens_seen": 140671190,
      "step": 6524,
      "time_per_iteration": 3.830735683441162
    },
    {
      "auxiliary_loss_clip": 0.01129995,
      "auxiliary_loss_mlp": 0.00762207,
      "balance_loss_clip": 1.04488146,
      "balance_loss_mlp": 1.00048757,
      "epoch": 0.7845848614200686,
      "flos": 23656585271040.0,
      "grad_norm": 2.0818192810774314,
      "language_loss": 0.811427,
      "learning_rate": 4.67188103681619e-07,
      "loss": 0.83034903,
      "num_input_tokens_seen": 140690975,
      "step": 6525,
      "time_per_iteration": 3.0058846473693848
    },
    {
      "auxiliary_loss_clip": 0.01156376,
      "auxiliary_loss_mlp": 0.01023095,
      "balance_loss_clip": 1.04493701,
      "balance_loss_mlp": 1.01589465,
      "epoch": 0.7847051043107076,
      "flos": 23402194174080.0,
      "grad_norm": 2.5552769925511507,
      "language_loss": 0.69367963,
      "learning_rate": 4.666878424264453e-07,
      "loss": 0.71547437,
      "num_input_tokens_seen": 140710930,
      "step": 6526,
      "time_per_iteration": 3.5655479431152344
    },
    {
      "auxiliary_loss_clip": 0.0109987,
      "auxiliary_loss_mlp": 0.01021025,
      "balance_loss_clip": 1.04250669,
      "balance_loss_mlp": 1.01408744,
      "epoch": 0.7848253472013467,
      "flos": 19022277473280.0,
      "grad_norm": 1.7659032919729105,
      "language_loss": 0.73987019,
      "learning_rate": 4.661878137735069e-07,
      "loss": 0.76107913,
      "num_input_tokens_seen": 140729120,
      "step": 6527,
      "time_per_iteration": 3.6597728729248047
    },
    {
      "auxiliary_loss_clip": 0.01123886,
      "auxiliary_loss_mlp": 0.01025918,
      "balance_loss_clip": 1.04199743,
      "balance_loss_mlp": 1.01810348,
      "epoch": 0.7849455900919858,
      "flos": 21179180332800.0,
      "grad_norm": 1.917557504420922,
      "language_loss": 0.74963379,
      "learning_rate": 4.656880177986571e-07,
      "loss": 0.77113187,
      "num_input_tokens_seen": 140747665,
      "step": 6528,
      "time_per_iteration": 2.7284367084503174
    },
    {
      "auxiliary_loss_clip": 0.01147353,
      "auxiliary_loss_mlp": 0.01026293,
      "balance_loss_clip": 1.04320574,
      "balance_loss_mlp": 1.01849413,
      "epoch": 0.7850658329826249,
      "flos": 19536482620800.0,
      "grad_norm": 2.1914796286743794,
      "language_loss": 0.81562954,
      "learning_rate": 4.6518845457771607e-07,
      "loss": 0.83736598,
      "num_input_tokens_seen": 140766525,
      "step": 6529,
      "time_per_iteration": 2.721792459487915
    },
    {
      "auxiliary_loss_clip": 0.01131821,
      "auxiliary_loss_mlp": 0.01025939,
      "balance_loss_clip": 1.04281497,
      "balance_loss_mlp": 1.018345,
      "epoch": 0.7851860758732639,
      "flos": 12495334152960.0,
      "grad_norm": 1.896094906144328,
      "language_loss": 0.79133058,
      "learning_rate": 4.646891241864652e-07,
      "loss": 0.81290817,
      "num_input_tokens_seen": 140785090,
      "step": 6530,
      "time_per_iteration": 2.672797679901123
    },
    {
      "auxiliary_loss_clip": 0.01098579,
      "auxiliary_loss_mlp": 0.010302,
      "balance_loss_clip": 1.03973031,
      "balance_loss_mlp": 1.02166498,
      "epoch": 0.7853063187639031,
      "flos": 22960959505920.0,
      "grad_norm": 4.579681194693032,
      "language_loss": 0.73416382,
      "learning_rate": 4.6419002670065397e-07,
      "loss": 0.75545156,
      "num_input_tokens_seen": 140804670,
      "step": 6531,
      "time_per_iteration": 2.859881639480591
    },
    {
      "auxiliary_loss_clip": 0.01077525,
      "auxiliary_loss_mlp": 0.0102659,
      "balance_loss_clip": 1.04043388,
      "balance_loss_mlp": 1.0182631,
      "epoch": 0.7854265616545422,
      "flos": 17347260499200.0,
      "grad_norm": 2.2296765950402886,
      "language_loss": 0.86493653,
      "learning_rate": 4.6369116219599445e-07,
      "loss": 0.88597769,
      "num_input_tokens_seen": 140820655,
      "step": 6532,
      "time_per_iteration": 2.9505257606506348
    },
    {
      "auxiliary_loss_clip": 0.01118838,
      "auxiliary_loss_mlp": 0.01026999,
      "balance_loss_clip": 1.03874326,
      "balance_loss_mlp": 1.01949763,
      "epoch": 0.7855468045451812,
      "flos": 23838293197440.0,
      "grad_norm": 1.7291981577653146,
      "language_loss": 0.79555678,
      "learning_rate": 4.631925307481637e-07,
      "loss": 0.81701517,
      "num_input_tokens_seen": 140840470,
      "step": 6533,
      "time_per_iteration": 3.20285964012146
    },
    {
      "auxiliary_loss_clip": 0.01091194,
      "auxiliary_loss_mlp": 0.01025608,
      "balance_loss_clip": 1.03815734,
      "balance_loss_mlp": 1.01834846,
      "epoch": 0.7856670474358204,
      "flos": 25666792986240.0,
      "grad_norm": 2.5201319563880995,
      "language_loss": 0.75537938,
      "learning_rate": 4.6269413243280533e-07,
      "loss": 0.77654743,
      "num_input_tokens_seen": 140859890,
      "step": 6534,
      "time_per_iteration": 2.9087531566619873
    },
    {
      "auxiliary_loss_clip": 0.01101798,
      "auxiliary_loss_mlp": 0.0102435,
      "balance_loss_clip": 1.0398612,
      "balance_loss_mlp": 1.01597297,
      "epoch": 0.7857872903264594,
      "flos": 18144656472960.0,
      "grad_norm": 3.0280221313128464,
      "language_loss": 0.73836553,
      "learning_rate": 4.621959673255236e-07,
      "loss": 0.75962704,
      "num_input_tokens_seen": 140876190,
      "step": 6535,
      "time_per_iteration": 2.758784532546997
    },
    {
      "auxiliary_loss_clip": 0.01121824,
      "auxiliary_loss_mlp": 0.0103026,
      "balance_loss_clip": 1.0436002,
      "balance_loss_mlp": 1.02292228,
      "epoch": 0.7859075332170985,
      "flos": 14386138081920.0,
      "grad_norm": 1.9794195056625756,
      "language_loss": 0.90779114,
      "learning_rate": 4.6169803550189135e-07,
      "loss": 0.92931199,
      "num_input_tokens_seen": 140891885,
      "step": 6536,
      "time_per_iteration": 2.7111294269561768
    },
    {
      "auxiliary_loss_clip": 0.01097122,
      "auxiliary_loss_mlp": 0.0076312,
      "balance_loss_clip": 1.03975582,
      "balance_loss_mlp": 1.00053334,
      "epoch": 0.7860277761077377,
      "flos": 19864059678720.0,
      "grad_norm": 2.000418825236103,
      "language_loss": 0.77710092,
      "learning_rate": 4.6120033703744355e-07,
      "loss": 0.79570341,
      "num_input_tokens_seen": 140910780,
      "step": 6537,
      "time_per_iteration": 2.7575716972351074
    },
    {
      "auxiliary_loss_clip": 0.01111904,
      "auxiliary_loss_mlp": 0.01027399,
      "balance_loss_clip": 1.04263008,
      "balance_loss_mlp": 1.01955557,
      "epoch": 0.7861480189983767,
      "flos": 26396174557440.0,
      "grad_norm": 2.2943227996688624,
      "language_loss": 0.78625154,
      "learning_rate": 4.607028720076822e-07,
      "loss": 0.80764461,
      "num_input_tokens_seen": 140927460,
      "step": 6538,
      "time_per_iteration": 2.795926094055176
    },
    {
      "auxiliary_loss_clip": 0.01140976,
      "auxiliary_loss_mlp": 0.01027803,
      "balance_loss_clip": 1.04376316,
      "balance_loss_mlp": 1.02031016,
      "epoch": 0.7862682618890158,
      "flos": 24236578177920.0,
      "grad_norm": 1.957513080716449,
      "language_loss": 0.7347132,
      "learning_rate": 4.6020564048807074e-07,
      "loss": 0.75640094,
      "num_input_tokens_seen": 140945135,
      "step": 6539,
      "time_per_iteration": 2.7741143703460693
    },
    {
      "auxiliary_loss_clip": 0.01116193,
      "auxiliary_loss_mlp": 0.01023028,
      "balance_loss_clip": 1.04020667,
      "balance_loss_mlp": 1.01567864,
      "epoch": 0.7863885047796549,
      "flos": 47551508259840.0,
      "grad_norm": 1.95155194149786,
      "language_loss": 0.71576285,
      "learning_rate": 4.5970864255403883e-07,
      "loss": 0.73715508,
      "num_input_tokens_seen": 140966660,
      "step": 6540,
      "time_per_iteration": 3.972367286682129
    },
    {
      "auxiliary_loss_clip": 0.01116488,
      "auxiliary_loss_mlp": 0.01032237,
      "balance_loss_clip": 1.04398584,
      "balance_loss_mlp": 1.02450609,
      "epoch": 0.786508747670294,
      "flos": 24389234979840.0,
      "grad_norm": 2.4189762150968397,
      "language_loss": 0.82013547,
      "learning_rate": 4.59211878280982e-07,
      "loss": 0.84162271,
      "num_input_tokens_seen": 140986175,
      "step": 6541,
      "time_per_iteration": 2.7451364994049072
    },
    {
      "auxiliary_loss_clip": 0.01096563,
      "auxiliary_loss_mlp": 0.01022202,
      "balance_loss_clip": 1.04028773,
      "balance_loss_mlp": 1.01490617,
      "epoch": 0.786628990560933,
      "flos": 18041234238720.0,
      "grad_norm": 2.775931471608682,
      "language_loss": 0.70680928,
      "learning_rate": 4.587153477442578e-07,
      "loss": 0.72799689,
      "num_input_tokens_seen": 141002490,
      "step": 6542,
      "time_per_iteration": 2.7614834308624268
    },
    {
      "auxiliary_loss_clip": 0.01144929,
      "auxiliary_loss_mlp": 0.01024851,
      "balance_loss_clip": 1.04618859,
      "balance_loss_mlp": 1.01620877,
      "epoch": 0.7867492334515722,
      "flos": 25848860048640.0,
      "grad_norm": 2.1179512792080613,
      "language_loss": 0.81677246,
      "learning_rate": 4.582190510191899e-07,
      "loss": 0.83847034,
      "num_input_tokens_seen": 141021150,
      "step": 6543,
      "time_per_iteration": 2.797966241836548
    },
    {
      "auxiliary_loss_clip": 0.01116825,
      "auxiliary_loss_mlp": 0.01031486,
      "balance_loss_clip": 1.04221117,
      "balance_loss_mlp": 1.02423215,
      "epoch": 0.7868694763422113,
      "flos": 16580819070720.0,
      "grad_norm": 2.2009600175890043,
      "language_loss": 0.87338018,
      "learning_rate": 4.5772298818106625e-07,
      "loss": 0.89486337,
      "num_input_tokens_seen": 141036940,
      "step": 6544,
      "time_per_iteration": 2.6795284748077393
    },
    {
      "auxiliary_loss_clip": 0.01098428,
      "auxiliary_loss_mlp": 0.01023004,
      "balance_loss_clip": 1.04199839,
      "balance_loss_mlp": 1.01524329,
      "epoch": 0.7869897192328503,
      "flos": 29386276272000.0,
      "grad_norm": 3.0944169649877638,
      "language_loss": 0.72095442,
      "learning_rate": 4.572271593051384e-07,
      "loss": 0.74216872,
      "num_input_tokens_seen": 141054295,
      "step": 6545,
      "time_per_iteration": 2.8123466968536377
    },
    {
      "auxiliary_loss_clip": 0.01133946,
      "auxiliary_loss_mlp": 0.01027575,
      "balance_loss_clip": 1.04444671,
      "balance_loss_mlp": 1.01986241,
      "epoch": 0.7871099621234895,
      "flos": 17128923678720.0,
      "grad_norm": 2.0151166697352747,
      "language_loss": 0.77961969,
      "learning_rate": 4.567315644666245e-07,
      "loss": 0.80123484,
      "num_input_tokens_seen": 141073090,
      "step": 6546,
      "time_per_iteration": 2.849215507507324
    },
    {
      "auxiliary_loss_clip": 0.01115665,
      "auxiliary_loss_mlp": 0.01029376,
      "balance_loss_clip": 1.04293919,
      "balance_loss_mlp": 1.02224767,
      "epoch": 0.7872302050141285,
      "flos": 23440187784960.0,
      "grad_norm": 19.494761070988893,
      "language_loss": 0.8502022,
      "learning_rate": 4.5623620374070507e-07,
      "loss": 0.8716526,
      "num_input_tokens_seen": 141092405,
      "step": 6547,
      "time_per_iteration": 2.88266921043396
    },
    {
      "auxiliary_loss_clip": 0.01035036,
      "auxiliary_loss_mlp": 0.01001755,
      "balance_loss_clip": 1.01453066,
      "balance_loss_mlp": 1.00019312,
      "epoch": 0.7873504479047676,
      "flos": 65959752689280.0,
      "grad_norm": 0.7603160081589154,
      "language_loss": 0.58340788,
      "learning_rate": 4.557410772025263e-07,
      "loss": 0.6037758,
      "num_input_tokens_seen": 141154355,
      "step": 6548,
      "time_per_iteration": 3.3885293006896973
    },
    {
      "auxiliary_loss_clip": 0.01125687,
      "auxiliary_loss_mlp": 0.01026591,
      "balance_loss_clip": 1.04344261,
      "balance_loss_mlp": 1.01875854,
      "epoch": 0.7874706907954068,
      "flos": 23258336204160.0,
      "grad_norm": 2.2452309663862215,
      "language_loss": 0.66165972,
      "learning_rate": 4.5524618492719803e-07,
      "loss": 0.68318248,
      "num_input_tokens_seen": 141173575,
      "step": 6549,
      "time_per_iteration": 2.7672386169433594
    },
    {
      "auxiliary_loss_clip": 0.01122575,
      "auxiliary_loss_mlp": 0.01024597,
      "balance_loss_clip": 1.03982246,
      "balance_loss_mlp": 1.01690245,
      "epoch": 0.7875909336860458,
      "flos": 28767786963840.0,
      "grad_norm": 2.896063991594732,
      "language_loss": 0.79273087,
      "learning_rate": 4.54751526989795e-07,
      "loss": 0.81420261,
      "num_input_tokens_seen": 141195415,
      "step": 6550,
      "time_per_iteration": 3.6755549907684326
    },
    {
      "auxiliary_loss_clip": 0.01142331,
      "auxiliary_loss_mlp": 0.01024758,
      "balance_loss_clip": 1.04453015,
      "balance_loss_mlp": 1.01729274,
      "epoch": 0.7877111765766849,
      "flos": 18697286194560.0,
      "grad_norm": 1.882354724471456,
      "language_loss": 0.79216725,
      "learning_rate": 4.5425710346535775e-07,
      "loss": 0.81383812,
      "num_input_tokens_seen": 141213360,
      "step": 6551,
      "time_per_iteration": 2.694065809249878
    },
    {
      "auxiliary_loss_clip": 0.01135828,
      "auxiliary_loss_mlp": 0.01026129,
      "balance_loss_clip": 1.04222679,
      "balance_loss_mlp": 1.018345,
      "epoch": 0.787831419467324,
      "flos": 27592968833280.0,
      "grad_norm": 2.079848903511766,
      "language_loss": 0.81537676,
      "learning_rate": 4.537629144288877e-07,
      "loss": 0.83699638,
      "num_input_tokens_seen": 141230815,
      "step": 6552,
      "time_per_iteration": 3.752958059310913
    },
    {
      "auxiliary_loss_clip": 0.0115338,
      "auxiliary_loss_mlp": 0.01030419,
      "balance_loss_clip": 1.04563832,
      "balance_loss_mlp": 1.02285552,
      "epoch": 0.7879516623579631,
      "flos": 18150187167360.0,
      "grad_norm": 2.275433212717665,
      "language_loss": 0.75102752,
      "learning_rate": 4.5326895995535477e-07,
      "loss": 0.77286553,
      "num_input_tokens_seen": 141249715,
      "step": 6553,
      "time_per_iteration": 3.541685104370117
    },
    {
      "auxiliary_loss_clip": 0.01053998,
      "auxiliary_loss_mlp": 0.01018121,
      "balance_loss_clip": 1.03378606,
      "balance_loss_mlp": 1.01094413,
      "epoch": 0.7880719052486022,
      "flos": 20339193807360.0,
      "grad_norm": 2.648910894836471,
      "language_loss": 0.84381849,
      "learning_rate": 4.527752401196907e-07,
      "loss": 0.86453968,
      "num_input_tokens_seen": 141267730,
      "step": 6554,
      "time_per_iteration": 3.1136531829833984
    },
    {
      "auxiliary_loss_clip": 0.01159839,
      "auxiliary_loss_mlp": 0.01024847,
      "balance_loss_clip": 1.04568374,
      "balance_loss_mlp": 1.01728642,
      "epoch": 0.7881921481392413,
      "flos": 21653237053440.0,
      "grad_norm": 2.0292103269327946,
      "language_loss": 0.67065084,
      "learning_rate": 4.5228175499679254e-07,
      "loss": 0.69249773,
      "num_input_tokens_seen": 141287315,
      "step": 6555,
      "time_per_iteration": 3.0176801681518555
    },
    {
      "auxiliary_loss_clip": 0.01031526,
      "auxiliary_loss_mlp": 0.01002623,
      "balance_loss_clip": 1.01969838,
      "balance_loss_mlp": 1.00100172,
      "epoch": 0.7883123910298804,
      "flos": 68565860058240.0,
      "grad_norm": 0.832720140646836,
      "language_loss": 0.54509252,
      "learning_rate": 4.5178850466152174e-07,
      "loss": 0.56543404,
      "num_input_tokens_seen": 141346145,
      "step": 6556,
      "time_per_iteration": 3.4227774143218994
    },
    {
      "auxiliary_loss_clip": 0.0111498,
      "auxiliary_loss_mlp": 0.0076122,
      "balance_loss_clip": 1.04233813,
      "balance_loss_mlp": 1.00052261,
      "epoch": 0.7884326339205194,
      "flos": 19318217627520.0,
      "grad_norm": 2.113405758039426,
      "language_loss": 0.8195827,
      "learning_rate": 4.512954891887031e-07,
      "loss": 0.83834469,
      "num_input_tokens_seen": 141364445,
      "step": 6557,
      "time_per_iteration": 2.7164878845214844
    },
    {
      "auxiliary_loss_clip": 0.01154984,
      "auxiliary_loss_mlp": 0.01026636,
      "balance_loss_clip": 1.04512179,
      "balance_loss_mlp": 1.01877737,
      "epoch": 0.7885528768111585,
      "flos": 17784903807360.0,
      "grad_norm": 5.31690381351716,
      "language_loss": 0.8342703,
      "learning_rate": 4.5080270865312806e-07,
      "loss": 0.85608649,
      "num_input_tokens_seen": 141381640,
      "step": 6558,
      "time_per_iteration": 3.098055124282837
    },
    {
      "auxiliary_loss_clip": 0.01125227,
      "auxiliary_loss_mlp": 0.0102402,
      "balance_loss_clip": 1.04009271,
      "balance_loss_mlp": 1.01617634,
      "epoch": 0.7886731197017977,
      "flos": 18807639753600.0,
      "grad_norm": 1.901043658178147,
      "language_loss": 0.70955169,
      "learning_rate": 4.5031016312954985e-07,
      "loss": 0.73104417,
      "num_input_tokens_seen": 141399955,
      "step": 6559,
      "time_per_iteration": 2.741140365600586
    },
    {
      "auxiliary_loss_clip": 0.01138892,
      "auxiliary_loss_mlp": 0.01031527,
      "balance_loss_clip": 1.04574633,
      "balance_loss_mlp": 1.02346289,
      "epoch": 0.7887933625924367,
      "flos": 33365358126720.0,
      "grad_norm": 2.022267334268788,
      "language_loss": 0.7454744,
      "learning_rate": 4.498178526926886e-07,
      "loss": 0.7671786,
      "num_input_tokens_seen": 141420820,
      "step": 6560,
      "time_per_iteration": 2.781341075897217
    },
    {
      "auxiliary_loss_clip": 0.01117579,
      "auxiliary_loss_mlp": 0.0102635,
      "balance_loss_clip": 1.03995895,
      "balance_loss_mlp": 1.01879191,
      "epoch": 0.7889136054830758,
      "flos": 17019360218880.0,
      "grad_norm": 2.0225132835232884,
      "language_loss": 0.72216386,
      "learning_rate": 4.4932577741722635e-07,
      "loss": 0.74360311,
      "num_input_tokens_seen": 141439350,
      "step": 6561,
      "time_per_iteration": 2.795041799545288
    },
    {
      "auxiliary_loss_clip": 0.01142502,
      "auxiliary_loss_mlp": 0.0076205,
      "balance_loss_clip": 1.04602253,
      "balance_loss_mlp": 1.00043428,
      "epoch": 0.7890338483737149,
      "flos": 29424629018880.0,
      "grad_norm": 2.6317810822496446,
      "language_loss": 0.74292409,
      "learning_rate": 4.4883393737780985e-07,
      "loss": 0.76196963,
      "num_input_tokens_seen": 141460300,
      "step": 6562,
      "time_per_iteration": 2.720811367034912
    },
    {
      "auxiliary_loss_clip": 0.01101474,
      "auxiliary_loss_mlp": 0.01026099,
      "balance_loss_clip": 1.03599441,
      "balance_loss_mlp": 1.01860404,
      "epoch": 0.789154091264354,
      "flos": 19971576063360.0,
      "grad_norm": 1.9413319166924081,
      "language_loss": 0.78556377,
      "learning_rate": 4.4834233264905254e-07,
      "loss": 0.80683947,
      "num_input_tokens_seen": 141477315,
      "step": 6563,
      "time_per_iteration": 2.7855823040008545
    },
    {
      "auxiliary_loss_clip": 0.01113125,
      "auxiliary_loss_mlp": 0.01028202,
      "balance_loss_clip": 1.0451225,
      "balance_loss_mlp": 1.01933932,
      "epoch": 0.789274334154993,
      "flos": 14537825216640.0,
      "grad_norm": 2.571129888007294,
      "language_loss": 0.72033209,
      "learning_rate": 4.478509633055294e-07,
      "loss": 0.74174535,
      "num_input_tokens_seen": 141495025,
      "step": 6564,
      "time_per_iteration": 2.703831672668457
    },
    {
      "auxiliary_loss_clip": 0.01162191,
      "auxiliary_loss_mlp": 0.01031001,
      "balance_loss_clip": 1.04793012,
      "balance_loss_mlp": 1.02189398,
      "epoch": 0.7893945770456322,
      "flos": 21827403123840.0,
      "grad_norm": 2.7977627947657675,
      "language_loss": 0.80102426,
      "learning_rate": 4.473598294217813e-07,
      "loss": 0.82295614,
      "num_input_tokens_seen": 141510450,
      "step": 6565,
      "time_per_iteration": 2.6597187519073486
    },
    {
      "auxiliary_loss_clip": 0.01112201,
      "auxiliary_loss_mlp": 0.01026258,
      "balance_loss_clip": 1.04281402,
      "balance_loss_mlp": 1.01840818,
      "epoch": 0.7895148199362713,
      "flos": 20740639184640.0,
      "grad_norm": 2.1012728765637205,
      "language_loss": 0.71598732,
      "learning_rate": 4.468689310723124e-07,
      "loss": 0.73737186,
      "num_input_tokens_seen": 141528265,
      "step": 6566,
      "time_per_iteration": 3.663184881210327
    },
    {
      "auxiliary_loss_clip": 0.01170387,
      "auxiliary_loss_mlp": 0.01028568,
      "balance_loss_clip": 1.04844451,
      "balance_loss_mlp": 1.02058733,
      "epoch": 0.7896350628269103,
      "flos": 16690669839360.0,
      "grad_norm": 3.433259217483596,
      "language_loss": 0.7900275,
      "learning_rate": 4.463782683315913e-07,
      "loss": 0.81201708,
      "num_input_tokens_seen": 141547270,
      "step": 6567,
      "time_per_iteration": 2.6519277095794678
    },
    {
      "auxiliary_loss_clip": 0.01150987,
      "auxiliary_loss_mlp": 0.00761715,
      "balance_loss_clip": 1.04527831,
      "balance_loss_mlp": 1.00048089,
      "epoch": 0.7897553057175495,
      "flos": 22638374438400.0,
      "grad_norm": 2.2929949701878565,
      "language_loss": 0.73566222,
      "learning_rate": 4.458878412740523e-07,
      "loss": 0.75478917,
      "num_input_tokens_seen": 141566050,
      "step": 6568,
      "time_per_iteration": 2.7715096473693848
    },
    {
      "auxiliary_loss_clip": 0.01132349,
      "auxiliary_loss_mlp": 0.01023059,
      "balance_loss_clip": 1.04422188,
      "balance_loss_mlp": 1.01516187,
      "epoch": 0.7898755486081885,
      "flos": 14537573821440.0,
      "grad_norm": 2.53666573924089,
      "language_loss": 0.77733976,
      "learning_rate": 4.453976499740919e-07,
      "loss": 0.79889381,
      "num_input_tokens_seen": 141583695,
      "step": 6569,
      "time_per_iteration": 2.6901416778564453
    },
    {
      "auxiliary_loss_clip": 0.0113304,
      "auxiliary_loss_mlp": 0.01024705,
      "balance_loss_clip": 1.04631746,
      "balance_loss_mlp": 1.01709342,
      "epoch": 0.7899957914988276,
      "flos": 17238487138560.0,
      "grad_norm": 1.9135840313835442,
      "language_loss": 0.78585517,
      "learning_rate": 4.4490769450607215e-07,
      "loss": 0.80743259,
      "num_input_tokens_seen": 141601320,
      "step": 6570,
      "time_per_iteration": 2.707357406616211
    },
    {
      "auxiliary_loss_clip": 0.0109344,
      "auxiliary_loss_mlp": 0.01027499,
      "balance_loss_clip": 1.03772342,
      "balance_loss_mlp": 1.01948237,
      "epoch": 0.7901160343894668,
      "flos": 41279351086080.0,
      "grad_norm": 1.933232158146051,
      "language_loss": 0.73006546,
      "learning_rate": 4.4441797494431845e-07,
      "loss": 0.75127482,
      "num_input_tokens_seen": 141623125,
      "step": 6571,
      "time_per_iteration": 2.9223830699920654
    },
    {
      "auxiliary_loss_clip": 0.01143386,
      "auxiliary_loss_mlp": 0.01025052,
      "balance_loss_clip": 1.04516745,
      "balance_loss_mlp": 1.01738644,
      "epoch": 0.7902362772801058,
      "flos": 16837005847680.0,
      "grad_norm": 2.102762320145258,
      "language_loss": 0.77853531,
      "learning_rate": 4.439284913631207e-07,
      "loss": 0.80021966,
      "num_input_tokens_seen": 141640335,
      "step": 6572,
      "time_per_iteration": 2.680100679397583
    },
    {
      "auxiliary_loss_clip": 0.01173,
      "auxiliary_loss_mlp": 0.01024285,
      "balance_loss_clip": 1.04893708,
      "balance_loss_mlp": 1.01593113,
      "epoch": 0.7903565201707449,
      "flos": 27125987091840.0,
      "grad_norm": 2.0041418009894936,
      "language_loss": 0.83828849,
      "learning_rate": 4.434392438367347e-07,
      "loss": 0.86026132,
      "num_input_tokens_seen": 141659760,
      "step": 6573,
      "time_per_iteration": 2.660717248916626
    },
    {
      "auxiliary_loss_clip": 0.01101061,
      "auxiliary_loss_mlp": 0.0102471,
      "balance_loss_clip": 1.04019713,
      "balance_loss_mlp": 1.01649618,
      "epoch": 0.790476763061384,
      "flos": 31025167142400.0,
      "grad_norm": 2.069074507634518,
      "language_loss": 0.7421459,
      "learning_rate": 4.4295023243937677e-07,
      "loss": 0.76340365,
      "num_input_tokens_seen": 141679965,
      "step": 6574,
      "time_per_iteration": 2.851468324661255
    },
    {
      "auxiliary_loss_clip": 0.01150169,
      "auxiliary_loss_mlp": 0.01028982,
      "balance_loss_clip": 1.05027032,
      "balance_loss_mlp": 1.01989198,
      "epoch": 0.7905970059520231,
      "flos": 22089084681600.0,
      "grad_norm": 1.660772286811889,
      "language_loss": 0.80232972,
      "learning_rate": 4.4246145724523123e-07,
      "loss": 0.82412118,
      "num_input_tokens_seen": 141697710,
      "step": 6575,
      "time_per_iteration": 2.786025047302246
    },
    {
      "auxiliary_loss_clip": 0.01154407,
      "auxiliary_loss_mlp": 0.0102527,
      "balance_loss_clip": 1.04594946,
      "balance_loss_mlp": 1.01739073,
      "epoch": 0.7907172488426621,
      "flos": 20558141159040.0,
      "grad_norm": 2.2682075020076153,
      "language_loss": 0.77471292,
      "learning_rate": 4.41972918328444e-07,
      "loss": 0.79650968,
      "num_input_tokens_seen": 141715145,
      "step": 6576,
      "time_per_iteration": 3.544952392578125
    },
    {
      "auxiliary_loss_clip": 0.01102906,
      "auxiliary_loss_mlp": 0.01024534,
      "balance_loss_clip": 1.03899777,
      "balance_loss_mlp": 1.01630223,
      "epoch": 0.7908374917333013,
      "flos": 30081542901120.0,
      "grad_norm": 4.195082098856084,
      "language_loss": 0.77636051,
      "learning_rate": 4.4148461576312646e-07,
      "loss": 0.79763496,
      "num_input_tokens_seen": 141734810,
      "step": 6577,
      "time_per_iteration": 2.840134859085083
    },
    {
      "auxiliary_loss_clip": 0.01121596,
      "auxiliary_loss_mlp": 0.01026976,
      "balance_loss_clip": 1.0424453,
      "balance_loss_mlp": 1.01982069,
      "epoch": 0.7909577346239404,
      "flos": 20996359084800.0,
      "grad_norm": 1.564048228762802,
      "language_loss": 0.74806476,
      "learning_rate": 4.4099654962335343e-07,
      "loss": 0.7695505,
      "num_input_tokens_seen": 141755260,
      "step": 6578,
      "time_per_iteration": 3.6469554901123047
    },
    {
      "auxiliary_loss_clip": 0.01088595,
      "auxiliary_loss_mlp": 0.01028463,
      "balance_loss_clip": 1.04281437,
      "balance_loss_mlp": 1.02008247,
      "epoch": 0.7910779775145794,
      "flos": 26247935128320.0,
      "grad_norm": 2.1982799072369454,
      "language_loss": 0.7524755,
      "learning_rate": 4.405087199831636e-07,
      "loss": 0.77364612,
      "num_input_tokens_seen": 141775500,
      "step": 6579,
      "time_per_iteration": 4.039482593536377
    },
    {
      "auxiliary_loss_clip": 0.01099676,
      "auxiliary_loss_mlp": 0.0102604,
      "balance_loss_clip": 1.0429467,
      "balance_loss_mlp": 1.01828551,
      "epoch": 0.7911982204052186,
      "flos": 22564434291840.0,
      "grad_norm": 1.968541850268615,
      "language_loss": 0.67001754,
      "learning_rate": 4.400211269165619e-07,
      "loss": 0.6912747,
      "num_input_tokens_seen": 141791955,
      "step": 6580,
      "time_per_iteration": 3.323281764984131
    },
    {
      "auxiliary_loss_clip": 0.01123461,
      "auxiliary_loss_mlp": 0.00761655,
      "balance_loss_clip": 1.04276919,
      "balance_loss_mlp": 1.00043178,
      "epoch": 0.7913184632958576,
      "flos": 23112538899840.0,
      "grad_norm": 1.514761946908337,
      "language_loss": 0.77156222,
      "learning_rate": 4.3953377049751416e-07,
      "loss": 0.79041338,
      "num_input_tokens_seen": 141812380,
      "step": 6581,
      "time_per_iteration": 2.8166749477386475
    },
    {
      "auxiliary_loss_clip": 0.01134623,
      "auxiliary_loss_mlp": 0.01024749,
      "balance_loss_clip": 1.04439998,
      "balance_loss_mlp": 1.01729822,
      "epoch": 0.7914387061864967,
      "flos": 12311758719360.0,
      "grad_norm": 2.6185542122475174,
      "language_loss": 0.77821279,
      "learning_rate": 4.390466507999537e-07,
      "loss": 0.79980648,
      "num_input_tokens_seen": 141828130,
      "step": 6582,
      "time_per_iteration": 2.692070484161377
    },
    {
      "auxiliary_loss_clip": 0.01121605,
      "auxiliary_loss_mlp": 0.01019695,
      "balance_loss_clip": 1.04092586,
      "balance_loss_mlp": 1.01195252,
      "epoch": 0.7915589490771359,
      "flos": 17603267708160.0,
      "grad_norm": 2.119799048971303,
      "language_loss": 0.76106465,
      "learning_rate": 4.385597678977748e-07,
      "loss": 0.78247762,
      "num_input_tokens_seen": 141846965,
      "step": 6583,
      "time_per_iteration": 2.8108463287353516
    },
    {
      "auxiliary_loss_clip": 0.01172337,
      "auxiliary_loss_mlp": 0.01027283,
      "balance_loss_clip": 1.04619908,
      "balance_loss_mlp": 1.02006245,
      "epoch": 0.7916791919677749,
      "flos": 25591272641280.0,
      "grad_norm": 1.8124893735920564,
      "language_loss": 0.75607955,
      "learning_rate": 4.3807312186483726e-07,
      "loss": 0.77807575,
      "num_input_tokens_seen": 141867685,
      "step": 6584,
      "time_per_iteration": 2.6519267559051514
    },
    {
      "auxiliary_loss_clip": 0.01133555,
      "auxiliary_loss_mlp": 0.01023861,
      "balance_loss_clip": 1.04397833,
      "balance_loss_mlp": 1.0167768,
      "epoch": 0.791799434858414,
      "flos": 18844340474880.0,
      "grad_norm": 2.2879429872569683,
      "language_loss": 0.78752303,
      "learning_rate": 4.375867127749655e-07,
      "loss": 0.80909717,
      "num_input_tokens_seen": 141885960,
      "step": 6585,
      "time_per_iteration": 2.6872947216033936
    },
    {
      "auxiliary_loss_clip": 0.01110207,
      "auxiliary_loss_mlp": 0.01025356,
      "balance_loss_clip": 1.04362202,
      "balance_loss_mlp": 1.0174222,
      "epoch": 0.7919196777490531,
      "flos": 25812015672960.0,
      "grad_norm": 1.8830635194638463,
      "language_loss": 0.67361689,
      "learning_rate": 4.3710054070194744e-07,
      "loss": 0.69497252,
      "num_input_tokens_seen": 141905655,
      "step": 6586,
      "time_per_iteration": 2.7980048656463623
    },
    {
      "auxiliary_loss_clip": 0.01090567,
      "auxiliary_loss_mlp": 0.01029333,
      "balance_loss_clip": 1.039083,
      "balance_loss_mlp": 1.02135181,
      "epoch": 0.7920399206396922,
      "flos": 11947624594560.0,
      "grad_norm": 2.7529727649600955,
      "language_loss": 0.66675317,
      "learning_rate": 4.3661460571953455e-07,
      "loss": 0.68795216,
      "num_input_tokens_seen": 141922390,
      "step": 6587,
      "time_per_iteration": 2.7797231674194336
    },
    {
      "auxiliary_loss_clip": 0.01137254,
      "auxiliary_loss_mlp": 0.00761845,
      "balance_loss_clip": 1.04060447,
      "balance_loss_mlp": 1.00043464,
      "epoch": 0.7921601635303313,
      "flos": 21579907438080.0,
      "grad_norm": 2.9445105381225565,
      "language_loss": 0.68765223,
      "learning_rate": 4.36128907901443e-07,
      "loss": 0.70664328,
      "num_input_tokens_seen": 141941985,
      "step": 6588,
      "time_per_iteration": 2.703261613845825
    },
    {
      "auxiliary_loss_clip": 0.01125219,
      "auxiliary_loss_mlp": 0.01028311,
      "balance_loss_clip": 1.0442214,
      "balance_loss_mlp": 1.02064586,
      "epoch": 0.7922804064209703,
      "flos": 18113989236480.0,
      "grad_norm": 3.79159150306087,
      "language_loss": 0.72841525,
      "learning_rate": 4.356434473213519e-07,
      "loss": 0.74995053,
      "num_input_tokens_seen": 141959435,
      "step": 6589,
      "time_per_iteration": 2.85678768157959
    },
    {
      "auxiliary_loss_clip": 0.01090419,
      "auxiliary_loss_mlp": 0.01035548,
      "balance_loss_clip": 1.03999126,
      "balance_loss_mlp": 1.0276804,
      "epoch": 0.7924006493116095,
      "flos": 21652806090240.0,
      "grad_norm": 2.0519057000301757,
      "language_loss": 0.79603803,
      "learning_rate": 4.351582240529068e-07,
      "loss": 0.8172977,
      "num_input_tokens_seen": 141980265,
      "step": 6590,
      "time_per_iteration": 2.8836452960968018
    },
    {
      "auxiliary_loss_clip": 0.01044028,
      "auxiliary_loss_mlp": 0.01001386,
      "balance_loss_clip": 1.01491392,
      "balance_loss_mlp": 0.99978888,
      "epoch": 0.7925208922022485,
      "flos": 64242755694720.0,
      "grad_norm": 0.6841993018544582,
      "language_loss": 0.58171409,
      "learning_rate": 4.346732381697149e-07,
      "loss": 0.60216826,
      "num_input_tokens_seen": 142044395,
      "step": 6591,
      "time_per_iteration": 5.080576181411743
    },
    {
      "auxiliary_loss_clip": 0.01139505,
      "auxiliary_loss_mlp": 0.01021587,
      "balance_loss_clip": 1.04119706,
      "balance_loss_mlp": 1.0137794,
      "epoch": 0.7926411350928876,
      "flos": 16941541403520.0,
      "grad_norm": 1.9759244830918996,
      "language_loss": 0.81733978,
      "learning_rate": 4.3418848974534825e-07,
      "loss": 0.83895069,
      "num_input_tokens_seen": 142061335,
      "step": 6592,
      "time_per_iteration": 2.7012059688568115
    },
    {
      "auxiliary_loss_clip": 0.01117857,
      "auxiliary_loss_mlp": 0.00761811,
      "balance_loss_clip": 1.04287338,
      "balance_loss_mlp": 1.00045657,
      "epoch": 0.7927613779835267,
      "flos": 34459987144320.0,
      "grad_norm": 2.291423736872176,
      "language_loss": 0.68731022,
      "learning_rate": 4.3370397885334276e-07,
      "loss": 0.70610696,
      "num_input_tokens_seen": 142081965,
      "step": 6593,
      "time_per_iteration": 2.8745086193084717
    },
    {
      "auxiliary_loss_clip": 0.01123893,
      "auxiliary_loss_mlp": 0.01024701,
      "balance_loss_clip": 1.04435372,
      "balance_loss_mlp": 1.01603997,
      "epoch": 0.7928816208741658,
      "flos": 18951174501120.0,
      "grad_norm": 1.8682402853466995,
      "language_loss": 0.75886846,
      "learning_rate": 4.3321970556719777e-07,
      "loss": 0.78035438,
      "num_input_tokens_seen": 142100260,
      "step": 6594,
      "time_per_iteration": 2.7295308113098145
    },
    {
      "auxiliary_loss_clip": 0.01139893,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.04428792,
      "balance_loss_mlp": 1.02152479,
      "epoch": 0.7930018637648049,
      "flos": 18623022825600.0,
      "grad_norm": 2.2335468224038166,
      "language_loss": 0.72110057,
      "learning_rate": 4.3273566996037856e-07,
      "loss": 0.74278831,
      "num_input_tokens_seen": 142116955,
      "step": 6595,
      "time_per_iteration": 2.6740481853485107
    },
    {
      "auxiliary_loss_clip": 0.01120752,
      "auxiliary_loss_mlp": 0.01024045,
      "balance_loss_clip": 1.03864717,
      "balance_loss_mlp": 1.01612091,
      "epoch": 0.793122106655444,
      "flos": 24530650824960.0,
      "grad_norm": 15.761755527273099,
      "language_loss": 0.80894136,
      "learning_rate": 4.322518721063113e-07,
      "loss": 0.83038938,
      "num_input_tokens_seen": 142135505,
      "step": 6596,
      "time_per_iteration": 2.7176666259765625
    },
    {
      "auxiliary_loss_clip": 0.01147488,
      "auxiliary_loss_mlp": 0.0102648,
      "balance_loss_clip": 1.04375386,
      "balance_loss_mlp": 1.01813483,
      "epoch": 0.7932423495460831,
      "flos": 34421203434240.0,
      "grad_norm": 2.5124478000828416,
      "language_loss": 0.70592064,
      "learning_rate": 4.3176831207838906e-07,
      "loss": 0.7276603,
      "num_input_tokens_seen": 142158915,
      "step": 6597,
      "time_per_iteration": 2.8045806884765625
    },
    {
      "auxiliary_loss_clip": 0.01104082,
      "auxiliary_loss_mlp": 0.01027569,
      "balance_loss_clip": 1.04032767,
      "balance_loss_mlp": 1.01958156,
      "epoch": 0.7933625924367221,
      "flos": 26980333441920.0,
      "grad_norm": 1.799376219839804,
      "language_loss": 0.74960566,
      "learning_rate": 4.3128498994996685e-07,
      "loss": 0.77092212,
      "num_input_tokens_seen": 142178390,
      "step": 6598,
      "time_per_iteration": 2.753209352493286
    },
    {
      "auxiliary_loss_clip": 0.0110444,
      "auxiliary_loss_mlp": 0.0102745,
      "balance_loss_clip": 1.0386188,
      "balance_loss_mlp": 1.018682,
      "epoch": 0.7934828353273613,
      "flos": 29568630643200.0,
      "grad_norm": 2.36296529554938,
      "language_loss": 0.71843857,
      "learning_rate": 4.308019057943646e-07,
      "loss": 0.73975748,
      "num_input_tokens_seen": 142200115,
      "step": 6599,
      "time_per_iteration": 2.859708070755005
    },
    {
      "auxiliary_loss_clip": 0.01147094,
      "auxiliary_loss_mlp": 0.0102498,
      "balance_loss_clip": 1.04376543,
      "balance_loss_mlp": 1.01676059,
      "epoch": 0.7936030782180004,
      "flos": 28615381557120.0,
      "grad_norm": 1.5734407333021188,
      "language_loss": 0.74615258,
      "learning_rate": 4.3031905968486535e-07,
      "loss": 0.76787335,
      "num_input_tokens_seen": 142220945,
      "step": 6600,
      "time_per_iteration": 2.858826160430908
    },
    {
      "auxiliary_loss_clip": 0.01123958,
      "auxiliary_loss_mlp": 0.0102452,
      "balance_loss_clip": 1.04275227,
      "balance_loss_mlp": 1.01650882,
      "epoch": 0.7937233211086394,
      "flos": 16392574869120.0,
      "grad_norm": 2.114453064999599,
      "language_loss": 0.68821561,
      "learning_rate": 4.298364516947162e-07,
      "loss": 0.70970047,
      "num_input_tokens_seen": 142238175,
      "step": 6601,
      "time_per_iteration": 2.7183618545532227
    },
    {
      "auxiliary_loss_clip": 0.01104139,
      "auxiliary_loss_mlp": 0.01024802,
      "balance_loss_clip": 1.03995788,
      "balance_loss_mlp": 1.01757824,
      "epoch": 0.7938435639992786,
      "flos": 22013420682240.0,
      "grad_norm": 2.6490578962131885,
      "language_loss": 0.65438062,
      "learning_rate": 4.293540818971295e-07,
      "loss": 0.67567003,
      "num_input_tokens_seen": 142255980,
      "step": 6602,
      "time_per_iteration": 3.7059216499328613
    },
    {
      "auxiliary_loss_clip": 0.01093453,
      "auxiliary_loss_mlp": 0.0102291,
      "balance_loss_clip": 1.03835762,
      "balance_loss_mlp": 1.01523256,
      "epoch": 0.7939638068899176,
      "flos": 22197032029440.0,
      "grad_norm": 2.17799819628961,
      "language_loss": 0.76863867,
      "learning_rate": 4.2887195036527934e-07,
      "loss": 0.78980231,
      "num_input_tokens_seen": 142274785,
      "step": 6603,
      "time_per_iteration": 2.7672626972198486
    },
    {
      "auxiliary_loss_clip": 0.0113346,
      "auxiliary_loss_mlp": 0.01020215,
      "balance_loss_clip": 1.04371715,
      "balance_loss_mlp": 1.01223993,
      "epoch": 0.7940840497805567,
      "flos": 17745186343680.0,
      "grad_norm": 3.1076002202604904,
      "language_loss": 0.7365135,
      "learning_rate": 4.28390057172306e-07,
      "loss": 0.75805032,
      "num_input_tokens_seen": 142291290,
      "step": 6604,
      "time_per_iteration": 4.556440114974976
    },
    {
      "auxiliary_loss_clip": 0.01160703,
      "auxiliary_loss_mlp": 0.01033108,
      "balance_loss_clip": 1.04805946,
      "balance_loss_mlp": 1.0249722,
      "epoch": 0.7942042926711959,
      "flos": 23805435231360.0,
      "grad_norm": 2.1145113063762313,
      "language_loss": 0.72356904,
      "learning_rate": 4.279084023913111e-07,
      "loss": 0.74550712,
      "num_input_tokens_seen": 142309165,
      "step": 6605,
      "time_per_iteration": 2.699657440185547
    },
    {
      "auxiliary_loss_clip": 0.01119952,
      "auxiliary_loss_mlp": 0.01024486,
      "balance_loss_clip": 1.04087615,
      "balance_loss_mlp": 1.01687121,
      "epoch": 0.7943245355618349,
      "flos": 19244959839360.0,
      "grad_norm": 1.9464449994476094,
      "language_loss": 0.69480276,
      "learning_rate": 4.2742698609536096e-07,
      "loss": 0.71624708,
      "num_input_tokens_seen": 142327475,
      "step": 6606,
      "time_per_iteration": 2.720489501953125
    },
    {
      "auxiliary_loss_clip": 0.01084466,
      "auxiliary_loss_mlp": 0.01026402,
      "balance_loss_clip": 1.03580499,
      "balance_loss_mlp": 1.01840317,
      "epoch": 0.794444778452474,
      "flos": 25007616547200.0,
      "grad_norm": 2.518616499108518,
      "language_loss": 0.7885313,
      "learning_rate": 4.2694580835748706e-07,
      "loss": 0.80963999,
      "num_input_tokens_seen": 142347335,
      "step": 6607,
      "time_per_iteration": 2.805316209793091
    },
    {
      "auxiliary_loss_clip": 0.01142107,
      "auxiliary_loss_mlp": 0.01022271,
      "balance_loss_clip": 1.04602528,
      "balance_loss_mlp": 1.01450491,
      "epoch": 0.7945650213431131,
      "flos": 23221491828480.0,
      "grad_norm": 2.0241302277427753,
      "language_loss": 0.74328315,
      "learning_rate": 4.264648692506836e-07,
      "loss": 0.76492691,
      "num_input_tokens_seen": 142366125,
      "step": 6608,
      "time_per_iteration": 2.7283029556274414
    },
    {
      "auxiliary_loss_clip": 0.01130583,
      "auxiliary_loss_mlp": 0.01026653,
      "balance_loss_clip": 1.04197598,
      "balance_loss_mlp": 1.01893115,
      "epoch": 0.7946852642337522,
      "flos": 26062887237120.0,
      "grad_norm": 1.9271730484549225,
      "language_loss": 0.72195762,
      "learning_rate": 4.2598416884790824e-07,
      "loss": 0.74353004,
      "num_input_tokens_seen": 142385175,
      "step": 6609,
      "time_per_iteration": 2.7364776134490967
    },
    {
      "auxiliary_loss_clip": 0.0114019,
      "auxiliary_loss_mlp": 0.01028022,
      "balance_loss_clip": 1.04497361,
      "balance_loss_mlp": 1.01961803,
      "epoch": 0.7948055071243912,
      "flos": 23769704177280.0,
      "grad_norm": 6.8732719400868625,
      "language_loss": 0.81082773,
      "learning_rate": 4.255037072220828e-07,
      "loss": 0.83250988,
      "num_input_tokens_seen": 142406545,
      "step": 6610,
      "time_per_iteration": 2.75764799118042
    },
    {
      "auxiliary_loss_clip": 0.0111168,
      "auxiliary_loss_mlp": 0.01024838,
      "balance_loss_clip": 1.03955746,
      "balance_loss_mlp": 1.01780152,
      "epoch": 0.7949257500150304,
      "flos": 21980814111360.0,
      "grad_norm": 1.6896769606021222,
      "language_loss": 0.71736038,
      "learning_rate": 4.2502348444609293e-07,
      "loss": 0.73872554,
      "num_input_tokens_seen": 142426165,
      "step": 6611,
      "time_per_iteration": 2.732295036315918
    },
    {
      "auxiliary_loss_clip": 0.01133027,
      "auxiliary_loss_mlp": 0.01024335,
      "balance_loss_clip": 1.04375434,
      "balance_loss_mlp": 1.01692569,
      "epoch": 0.7950459929056695,
      "flos": 25774129802880.0,
      "grad_norm": 2.056298770703719,
      "language_loss": 0.69732141,
      "learning_rate": 4.2454350059278844e-07,
      "loss": 0.71889508,
      "num_input_tokens_seen": 142447225,
      "step": 6612,
      "time_per_iteration": 2.8677029609680176
    },
    {
      "auxiliary_loss_clip": 0.01118281,
      "auxiliary_loss_mlp": 0.01024248,
      "balance_loss_clip": 1.04354143,
      "balance_loss_mlp": 1.01574802,
      "epoch": 0.7951662357963085,
      "flos": 22158068751360.0,
      "grad_norm": 1.8900673171930367,
      "language_loss": 0.84367704,
      "learning_rate": 4.240637557349824e-07,
      "loss": 0.86510229,
      "num_input_tokens_seen": 142464440,
      "step": 6613,
      "time_per_iteration": 2.6941330432891846
    },
    {
      "auxiliary_loss_clip": 0.01138396,
      "auxiliary_loss_mlp": 0.01021919,
      "balance_loss_clip": 1.04504228,
      "balance_loss_mlp": 1.01465356,
      "epoch": 0.7952864786869477,
      "flos": 24641938137600.0,
      "grad_norm": 2.0119868154243195,
      "language_loss": 0.66881037,
      "learning_rate": 4.235842499454516e-07,
      "loss": 0.69041353,
      "num_input_tokens_seen": 142484355,
      "step": 6614,
      "time_per_iteration": 2.7928106784820557
    },
    {
      "auxiliary_loss_clip": 0.01111997,
      "auxiliary_loss_mlp": 0.00761571,
      "balance_loss_clip": 1.0444355,
      "balance_loss_mlp": 1.00054145,
      "epoch": 0.7954067215775867,
      "flos": 21830922656640.0,
      "grad_norm": 2.2012751038106204,
      "language_loss": 0.82751966,
      "learning_rate": 4.2310498329693687e-07,
      "loss": 0.8462553,
      "num_input_tokens_seen": 142505255,
      "step": 6615,
      "time_per_iteration": 2.806581735610962
    },
    {
      "auxiliary_loss_clip": 0.01149392,
      "auxiliary_loss_mlp": 0.01026605,
      "balance_loss_clip": 1.04557991,
      "balance_loss_mlp": 1.01840973,
      "epoch": 0.7955269644682258,
      "flos": 24060652341120.0,
      "grad_norm": 48.26875237562629,
      "language_loss": 0.80797482,
      "learning_rate": 4.2262595586214164e-07,
      "loss": 0.8297348,
      "num_input_tokens_seen": 142526350,
      "step": 6616,
      "time_per_iteration": 2.757478952407837
    },
    {
      "auxiliary_loss_clip": 0.01079528,
      "auxiliary_loss_mlp": 0.0103318,
      "balance_loss_clip": 1.04011917,
      "balance_loss_mlp": 1.02443004,
      "epoch": 0.795647207358865,
      "flos": 25010741030400.0,
      "grad_norm": 1.7450156998062927,
      "language_loss": 0.76708597,
      "learning_rate": 4.221471677137358e-07,
      "loss": 0.78821301,
      "num_input_tokens_seen": 142547165,
      "step": 6617,
      "time_per_iteration": 3.8003146648406982
    },
    {
      "auxiliary_loss_clip": 0.01131928,
      "auxiliary_loss_mlp": 0.01026644,
      "balance_loss_clip": 1.04188919,
      "balance_loss_mlp": 1.01919365,
      "epoch": 0.795767450249504,
      "flos": 14648358343680.0,
      "grad_norm": 1.5762116160973894,
      "language_loss": 0.70376456,
      "learning_rate": 4.216686189243492e-07,
      "loss": 0.72535032,
      "num_input_tokens_seen": 142565955,
      "step": 6618,
      "time_per_iteration": 2.6910107135772705
    },
    {
      "auxiliary_loss_clip": 0.01144715,
      "auxiliary_loss_mlp": 0.01023381,
      "balance_loss_clip": 1.04481375,
      "balance_loss_mlp": 1.01475,
      "epoch": 0.7958876931401431,
      "flos": 18547897530240.0,
      "grad_norm": 1.6743997222100078,
      "language_loss": 0.73169684,
      "learning_rate": 4.211903095665785e-07,
      "loss": 0.7533778,
      "num_input_tokens_seen": 142585340,
      "step": 6619,
      "time_per_iteration": 2.707733631134033
    },
    {
      "auxiliary_loss_clip": 0.01072626,
      "auxiliary_loss_mlp": 0.0102332,
      "balance_loss_clip": 1.03252304,
      "balance_loss_mlp": 1.01554191,
      "epoch": 0.7960079360307821,
      "flos": 21543960902400.0,
      "grad_norm": 1.7662065565811347,
      "language_loss": 0.75088578,
      "learning_rate": 4.2071223971298277e-07,
      "loss": 0.77184528,
      "num_input_tokens_seen": 142602525,
      "step": 6620,
      "time_per_iteration": 2.8120009899139404
    },
    {
      "auxiliary_loss_clip": 0.01141421,
      "auxiliary_loss_mlp": 0.01026659,
      "balance_loss_clip": 1.04458487,
      "balance_loss_mlp": 1.01843965,
      "epoch": 0.7961281789214213,
      "flos": 25481745095040.0,
      "grad_norm": 2.0811992260600625,
      "language_loss": 0.60964978,
      "learning_rate": 4.2023440943608433e-07,
      "loss": 0.63133061,
      "num_input_tokens_seen": 142622490,
      "step": 6621,
      "time_per_iteration": 2.7589707374572754
    },
    {
      "auxiliary_loss_clip": 0.0111387,
      "auxiliary_loss_mlp": 0.00761755,
      "balance_loss_clip": 1.03874004,
      "balance_loss_mlp": 1.00048161,
      "epoch": 0.7962484218120603,
      "flos": 21944436612480.0,
      "grad_norm": 1.5437135347863407,
      "language_loss": 0.78462315,
      "learning_rate": 4.1975681880837023e-07,
      "loss": 0.80337948,
      "num_input_tokens_seen": 142642495,
      "step": 6622,
      "time_per_iteration": 2.7805228233337402
    },
    {
      "auxiliary_loss_clip": 0.01108126,
      "auxiliary_loss_mlp": 0.01025617,
      "balance_loss_clip": 1.04097319,
      "balance_loss_mlp": 1.01802349,
      "epoch": 0.7963686647026994,
      "flos": 18876264687360.0,
      "grad_norm": 1.6922851151665232,
      "language_loss": 0.82429987,
      "learning_rate": 4.192794679022895e-07,
      "loss": 0.84563732,
      "num_input_tokens_seen": 142660820,
      "step": 6623,
      "time_per_iteration": 2.7513339519500732
    },
    {
      "auxiliary_loss_clip": 0.01123948,
      "auxiliary_loss_mlp": 0.01021718,
      "balance_loss_clip": 1.04009807,
      "balance_loss_mlp": 1.01468766,
      "epoch": 0.7964889075933386,
      "flos": 29716582763520.0,
      "grad_norm": 1.883037924306079,
      "language_loss": 0.72039986,
      "learning_rate": 4.1880235679025743e-07,
      "loss": 0.74185652,
      "num_input_tokens_seen": 142680915,
      "step": 6624,
      "time_per_iteration": 2.7717549800872803
    },
    {
      "auxiliary_loss_clip": 0.01160293,
      "auxiliary_loss_mlp": 0.0102333,
      "balance_loss_clip": 1.04710448,
      "balance_loss_mlp": 1.01481855,
      "epoch": 0.7966091504839776,
      "flos": 29491458272640.0,
      "grad_norm": 2.0850772979407735,
      "language_loss": 0.63903368,
      "learning_rate": 4.1832548554464986e-07,
      "loss": 0.66086996,
      "num_input_tokens_seen": 142699210,
      "step": 6625,
      "time_per_iteration": 2.690922260284424
    },
    {
      "auxiliary_loss_clip": 0.01068449,
      "auxiliary_loss_mlp": 0.01003878,
      "balance_loss_clip": 1.01629162,
      "balance_loss_mlp": 1.00198221,
      "epoch": 0.7967293933746167,
      "flos": 67288697101440.0,
      "grad_norm": 0.7387823656781429,
      "language_loss": 0.58650148,
      "learning_rate": 4.178488542378098e-07,
      "loss": 0.60722476,
      "num_input_tokens_seen": 142756790,
      "step": 6626,
      "time_per_iteration": 3.178666353225708
    },
    {
      "auxiliary_loss_clip": 0.01142601,
      "auxiliary_loss_mlp": 0.0102089,
      "balance_loss_clip": 1.04369915,
      "balance_loss_mlp": 1.01259351,
      "epoch": 0.7968496362652558,
      "flos": 25554679660800.0,
      "grad_norm": 2.7697235949229535,
      "language_loss": 0.8897081,
      "learning_rate": 4.173724629420401e-07,
      "loss": 0.9113431,
      "num_input_tokens_seen": 142778150,
      "step": 6627,
      "time_per_iteration": 2.7168757915496826
    },
    {
      "auxiliary_loss_clip": 0.0108298,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.03809679,
      "balance_loss_mlp": 1.01842701,
      "epoch": 0.7969698791558949,
      "flos": 14501088581760.0,
      "grad_norm": 5.775086855167011,
      "language_loss": 0.68769872,
      "learning_rate": 4.168963117296087e-07,
      "loss": 0.70878941,
      "num_input_tokens_seen": 142795485,
      "step": 6628,
      "time_per_iteration": 4.033246755599976
    },
    {
      "auxiliary_loss_clip": 0.0114102,
      "auxiliary_loss_mlp": 0.01028709,
      "balance_loss_clip": 1.04643095,
      "balance_loss_mlp": 1.02066875,
      "epoch": 0.797090122046534,
      "flos": 22127545169280.0,
      "grad_norm": 2.501196500664708,
      "language_loss": 0.7573024,
      "learning_rate": 4.1642040067274876e-07,
      "loss": 0.77899969,
      "num_input_tokens_seen": 142815155,
      "step": 6629,
      "time_per_iteration": 3.633084774017334
    },
    {
      "auxiliary_loss_clip": 0.01097875,
      "auxiliary_loss_mlp": 0.01019389,
      "balance_loss_clip": 1.03835082,
      "balance_loss_mlp": 1.01175928,
      "epoch": 0.7972103649371731,
      "flos": 19897671830400.0,
      "grad_norm": 2.329217355805171,
      "language_loss": 0.72611111,
      "learning_rate": 4.1594472984365493e-07,
      "loss": 0.7472837,
      "num_input_tokens_seen": 142833840,
      "step": 6630,
      "time_per_iteration": 3.70271897315979
    },
    {
      "auxiliary_loss_clip": 0.01123894,
      "auxiliary_loss_mlp": 0.0102331,
      "balance_loss_clip": 1.03750348,
      "balance_loss_mlp": 1.01550472,
      "epoch": 0.7973306078278122,
      "flos": 36058621847040.0,
      "grad_norm": 1.913456779694075,
      "language_loss": 0.77648759,
      "learning_rate": 4.154692993144862e-07,
      "loss": 0.79795963,
      "num_input_tokens_seen": 142853610,
      "step": 6631,
      "time_per_iteration": 2.890960454940796
    },
    {
      "auxiliary_loss_clip": 0.0110378,
      "auxiliary_loss_mlp": 0.01025777,
      "balance_loss_clip": 1.0426805,
      "balance_loss_mlp": 1.01855922,
      "epoch": 0.7974508507184512,
      "flos": 21360600950400.0,
      "grad_norm": 2.0900394072457495,
      "language_loss": 0.7204119,
      "learning_rate": 4.1499410915736476e-07,
      "loss": 0.74170744,
      "num_input_tokens_seen": 142872540,
      "step": 6632,
      "time_per_iteration": 2.7849693298339844
    },
    {
      "auxiliary_loss_clip": 0.01044968,
      "auxiliary_loss_mlp": 0.01003435,
      "balance_loss_clip": 1.01629925,
      "balance_loss_mlp": 1.00172997,
      "epoch": 0.7975710936090904,
      "flos": 68253115317120.0,
      "grad_norm": 0.768297359854931,
      "language_loss": 0.64173639,
      "learning_rate": 4.145191594443762e-07,
      "loss": 0.66222036,
      "num_input_tokens_seen": 142936895,
      "step": 6633,
      "time_per_iteration": 3.3991200923919678
    },
    {
      "auxiliary_loss_clip": 0.0112921,
      "auxiliary_loss_mlp": 0.01029641,
      "balance_loss_clip": 1.04206741,
      "balance_loss_mlp": 1.02118278,
      "epoch": 0.7976913364997295,
      "flos": 22492433479680.0,
      "grad_norm": 1.8700381638419106,
      "language_loss": 0.7067169,
      "learning_rate": 4.140444502475713e-07,
      "loss": 0.72830546,
      "num_input_tokens_seen": 142956445,
      "step": 6634,
      "time_per_iteration": 2.715564250946045
    },
    {
      "auxiliary_loss_clip": 0.01123811,
      "auxiliary_loss_mlp": 0.01028065,
      "balance_loss_clip": 1.0413115,
      "balance_loss_mlp": 1.02072787,
      "epoch": 0.7978115793903685,
      "flos": 15263220378240.0,
      "grad_norm": 1.8616636324977434,
      "language_loss": 0.70180988,
      "learning_rate": 4.1356998163896216e-07,
      "loss": 0.72332871,
      "num_input_tokens_seen": 142973495,
      "step": 6635,
      "time_per_iteration": 2.76621675491333
    },
    {
      "auxiliary_loss_clip": 0.01123025,
      "auxiliary_loss_mlp": 0.01027811,
      "balance_loss_clip": 1.04173446,
      "balance_loss_mlp": 1.02036679,
      "epoch": 0.7979318222810077,
      "flos": 19719232041600.0,
      "grad_norm": 6.128919169834119,
      "language_loss": 0.74982607,
      "learning_rate": 4.130957536905255e-07,
      "loss": 0.77133441,
      "num_input_tokens_seen": 142991510,
      "step": 6636,
      "time_per_iteration": 2.7429587841033936
    },
    {
      "auxiliary_loss_clip": 0.01136051,
      "auxiliary_loss_mlp": 0.01023295,
      "balance_loss_clip": 1.04672647,
      "balance_loss_mlp": 1.0147661,
      "epoch": 0.7980520651716467,
      "flos": 15560273854080.0,
      "grad_norm": 4.75137905591686,
      "language_loss": 0.71474653,
      "learning_rate": 4.1262176647420134e-07,
      "loss": 0.73633999,
      "num_input_tokens_seen": 143009675,
      "step": 6637,
      "time_per_iteration": 2.725430727005005
    },
    {
      "auxiliary_loss_clip": 0.01115379,
      "auxiliary_loss_mlp": 0.01021816,
      "balance_loss_clip": 1.04142451,
      "balance_loss_mlp": 1.014431,
      "epoch": 0.7981723080622858,
      "flos": 22309432663680.0,
      "grad_norm": 2.026340868232579,
      "language_loss": 0.79677022,
      "learning_rate": 4.121480200618923e-07,
      "loss": 0.81814218,
      "num_input_tokens_seen": 143029330,
      "step": 6638,
      "time_per_iteration": 2.7427923679351807
    },
    {
      "auxiliary_loss_clip": 0.01089745,
      "auxiliary_loss_mlp": 0.01030441,
      "balance_loss_clip": 1.03759754,
      "balance_loss_mlp": 1.0228827,
      "epoch": 0.798292550952925,
      "flos": 22929573997440.0,
      "grad_norm": 1.9516495503226692,
      "language_loss": 0.79890865,
      "learning_rate": 4.116745145254674e-07,
      "loss": 0.82011044,
      "num_input_tokens_seen": 143048865,
      "step": 6639,
      "time_per_iteration": 2.7824103832244873
    },
    {
      "auxiliary_loss_clip": 0.01044848,
      "auxiliary_loss_mlp": 0.01002719,
      "balance_loss_clip": 1.01489687,
      "balance_loss_mlp": 1.00097847,
      "epoch": 0.798412793843564,
      "flos": 64497936890880.0,
      "grad_norm": 0.7665656046324523,
      "language_loss": 0.57988816,
      "learning_rate": 4.1120124993675476e-07,
      "loss": 0.60036379,
      "num_input_tokens_seen": 143113295,
      "step": 6640,
      "time_per_iteration": 3.2815682888031006
    },
    {
      "auxiliary_loss_clip": 0.01053471,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.03668022,
      "balance_loss_mlp": 1.01721156,
      "epoch": 0.7985330367342031,
      "flos": 13586910514560.0,
      "grad_norm": 2.3106576921911204,
      "language_loss": 0.61951494,
      "learning_rate": 4.107282263675498e-07,
      "loss": 0.64030302,
      "num_input_tokens_seen": 143130965,
      "step": 6641,
      "time_per_iteration": 2.9925246238708496
    },
    {
      "auxiliary_loss_clip": 0.01033241,
      "auxiliary_loss_mlp": 0.01003298,
      "balance_loss_clip": 1.0273509,
      "balance_loss_mlp": 1.00113988,
      "epoch": 0.7986532796248422,
      "flos": 67698797656320.0,
      "grad_norm": 0.7797121668588115,
      "language_loss": 0.5247308,
      "learning_rate": 4.1025544388960907e-07,
      "loss": 0.54509622,
      "num_input_tokens_seen": 143192005,
      "step": 6642,
      "time_per_iteration": 3.9547791481018066
    },
    {
      "auxiliary_loss_clip": 0.01141221,
      "auxiliary_loss_mlp": 0.01026497,
      "balance_loss_clip": 1.04497313,
      "balance_loss_mlp": 1.01860833,
      "epoch": 0.7987735225154813,
      "flos": 22455373622400.0,
      "grad_norm": 15.520152328899227,
      "language_loss": 0.71498489,
      "learning_rate": 4.097829025746538e-07,
      "loss": 0.73666203,
      "num_input_tokens_seen": 143213550,
      "step": 6643,
      "time_per_iteration": 3.708036422729492
    },
    {
      "auxiliary_loss_clip": 0.01027453,
      "auxiliary_loss_mlp": 0.01003082,
      "balance_loss_clip": 1.01668048,
      "balance_loss_mlp": 1.0012219,
      "epoch": 0.7988937654061203,
      "flos": 68864098682880.0,
      "grad_norm": 0.6655981250156354,
      "language_loss": 0.60987854,
      "learning_rate": 4.0931060249436757e-07,
      "loss": 0.63018388,
      "num_input_tokens_seen": 143277390,
      "step": 6644,
      "time_per_iteration": 3.3278210163116455
    },
    {
      "auxiliary_loss_clip": 0.01086584,
      "auxiliary_loss_mlp": 0.01025931,
      "balance_loss_clip": 1.04021907,
      "balance_loss_mlp": 1.01837349,
      "epoch": 0.7990140082967595,
      "flos": 20806893820800.0,
      "grad_norm": 2.1277800399554323,
      "language_loss": 0.69987661,
      "learning_rate": 4.088385437203978e-07,
      "loss": 0.72100174,
      "num_input_tokens_seen": 143294400,
      "step": 6645,
      "time_per_iteration": 2.7917416095733643
    },
    {
      "auxiliary_loss_clip": 0.0117001,
      "auxiliary_loss_mlp": 0.01025182,
      "balance_loss_clip": 1.04659486,
      "balance_loss_mlp": 1.01714766,
      "epoch": 0.7991342511873986,
      "flos": 18985289443200.0,
      "grad_norm": 2.7441604544250935,
      "language_loss": 0.77535135,
      "learning_rate": 4.083667263243564e-07,
      "loss": 0.79730326,
      "num_input_tokens_seen": 143312745,
      "step": 6646,
      "time_per_iteration": 2.5779337882995605
    },
    {
      "auxiliary_loss_clip": 0.01159536,
      "auxiliary_loss_mlp": 0.0102234,
      "balance_loss_clip": 1.04749846,
      "balance_loss_mlp": 1.01471031,
      "epoch": 0.7992544940780376,
      "flos": 20816805974400.0,
      "grad_norm": 1.7586971023067375,
      "language_loss": 0.71838737,
      "learning_rate": 4.0789515037781653e-07,
      "loss": 0.74020612,
      "num_input_tokens_seen": 143333470,
      "step": 6647,
      "time_per_iteration": 2.6729061603546143
    },
    {
      "auxiliary_loss_clip": 0.01172498,
      "auxiliary_loss_mlp": 0.0102246,
      "balance_loss_clip": 1.04934919,
      "balance_loss_mlp": 1.01483023,
      "epoch": 0.7993747369686768,
      "flos": 12640772321280.0,
      "grad_norm": 1.9789750598733638,
      "language_loss": 0.82819223,
      "learning_rate": 4.0742381595231755e-07,
      "loss": 0.85014182,
      "num_input_tokens_seen": 143350195,
      "step": 6648,
      "time_per_iteration": 2.576992988586426
    },
    {
      "auxiliary_loss_clip": 0.01133323,
      "auxiliary_loss_mlp": 0.01022733,
      "balance_loss_clip": 1.04110813,
      "balance_loss_mlp": 1.0149188,
      "epoch": 0.7994949798593158,
      "flos": 20078769225600.0,
      "grad_norm": 2.9633262654545662,
      "language_loss": 0.78175807,
      "learning_rate": 4.06952723119359e-07,
      "loss": 0.80331862,
      "num_input_tokens_seen": 143370070,
      "step": 6649,
      "time_per_iteration": 2.692178726196289
    },
    {
      "auxiliary_loss_clip": 0.01110181,
      "auxiliary_loss_mlp": 0.01025491,
      "balance_loss_clip": 1.04442477,
      "balance_loss_mlp": 1.01773095,
      "epoch": 0.7996152227499549,
      "flos": 38654209509120.0,
      "grad_norm": 1.7593739671794728,
      "language_loss": 0.66999722,
      "learning_rate": 4.0648187195040504e-07,
      "loss": 0.69135392,
      "num_input_tokens_seen": 143392275,
      "step": 6650,
      "time_per_iteration": 2.856375217437744
    },
    {
      "auxiliary_loss_clip": 0.01078807,
      "auxiliary_loss_mlp": 0.01002424,
      "balance_loss_clip": 1.01796722,
      "balance_loss_mlp": 1.00054061,
      "epoch": 0.799735465640594,
      "flos": 70243821947520.0,
      "grad_norm": 0.8174354361538628,
      "language_loss": 0.67523921,
      "learning_rate": 4.060112625168848e-07,
      "loss": 0.69605154,
      "num_input_tokens_seen": 143457385,
      "step": 6651,
      "time_per_iteration": 3.2736921310424805
    },
    {
      "auxiliary_loss_clip": 0.01118627,
      "auxiliary_loss_mlp": 0.0076285,
      "balance_loss_clip": 1.0430963,
      "balance_loss_mlp": 1.00055933,
      "epoch": 0.7998557085312331,
      "flos": 24240995550720.0,
      "grad_norm": 1.8369240678427181,
      "language_loss": 0.74121588,
      "learning_rate": 4.055408948901886e-07,
      "loss": 0.76003063,
      "num_input_tokens_seen": 143478785,
      "step": 6652,
      "time_per_iteration": 2.8292088508605957
    },
    {
      "auxiliary_loss_clip": 0.011611,
      "auxiliary_loss_mlp": 0.00762899,
      "balance_loss_clip": 1.04722238,
      "balance_loss_mlp": 1.00058758,
      "epoch": 0.7999759514218722,
      "flos": 27564025449600.0,
      "grad_norm": 2.4784150948643298,
      "language_loss": 0.71626997,
      "learning_rate": 4.050707691416708e-07,
      "loss": 0.73550999,
      "num_input_tokens_seen": 143500095,
      "step": 6653,
      "time_per_iteration": 2.664585590362549
    },
    {
      "auxiliary_loss_clip": 0.01068493,
      "auxiliary_loss_mlp": 0.00753862,
      "balance_loss_clip": 1.01631701,
      "balance_loss_mlp": 1.00105941,
      "epoch": 0.8000961943125112,
      "flos": 67337428878720.0,
      "grad_norm": 0.6771900444748714,
      "language_loss": 0.59746718,
      "learning_rate": 4.046008853426495e-07,
      "loss": 0.61569071,
      "num_input_tokens_seen": 143563410,
      "step": 6654,
      "time_per_iteration": 4.220674991607666
    },
    {
      "auxiliary_loss_clip": 0.0110354,
      "auxiliary_loss_mlp": 0.01019685,
      "balance_loss_clip": 1.04203868,
      "balance_loss_mlp": 1.01197863,
      "epoch": 0.8002164372031504,
      "flos": 28733815676160.0,
      "grad_norm": 3.3989311378328217,
      "language_loss": 0.62683707,
      "learning_rate": 4.0413124356440464e-07,
      "loss": 0.64806932,
      "num_input_tokens_seen": 143587455,
      "step": 6655,
      "time_per_iteration": 3.8677351474761963
    },
    {
      "auxiliary_loss_clip": 0.01156135,
      "auxiliary_loss_mlp": 0.00762387,
      "balance_loss_clip": 1.04586244,
      "balance_loss_mlp": 1.00046611,
      "epoch": 0.8003366800937894,
      "flos": 17639429725440.0,
      "grad_norm": 1.8751348294657317,
      "language_loss": 0.82243681,
      "learning_rate": 4.0366184387818223e-07,
      "loss": 0.84162205,
      "num_input_tokens_seen": 143605915,
      "step": 6656,
      "time_per_iteration": 3.5897953510284424
    },
    {
      "auxiliary_loss_clip": 0.01163533,
      "auxiliary_loss_mlp": 0.01027628,
      "balance_loss_clip": 1.04912031,
      "balance_loss_mlp": 1.01887238,
      "epoch": 0.8004569229844285,
      "flos": 25995303797760.0,
      "grad_norm": 2.4538162976685265,
      "language_loss": 0.85288531,
      "learning_rate": 4.0319268635518797e-07,
      "loss": 0.87479693,
      "num_input_tokens_seen": 143626490,
      "step": 6657,
      "time_per_iteration": 2.764122724533081
    },
    {
      "auxiliary_loss_clip": 0.0107999,
      "auxiliary_loss_mlp": 0.01026696,
      "balance_loss_clip": 1.03844416,
      "balance_loss_mlp": 1.01884651,
      "epoch": 0.8005771658750677,
      "flos": 20812352688000.0,
      "grad_norm": 1.9365999389549269,
      "language_loss": 0.75160122,
      "learning_rate": 4.027237710665943e-07,
      "loss": 0.77266812,
      "num_input_tokens_seen": 143644955,
      "step": 6658,
      "time_per_iteration": 2.950021982192993
    },
    {
      "auxiliary_loss_clip": 0.01116711,
      "auxiliary_loss_mlp": 0.01026999,
      "balance_loss_clip": 1.0391655,
      "balance_loss_mlp": 1.01824892,
      "epoch": 0.8006974087657067,
      "flos": 25812626204160.0,
      "grad_norm": 2.025021941648308,
      "language_loss": 0.6919055,
      "learning_rate": 4.022550980835344e-07,
      "loss": 0.71334261,
      "num_input_tokens_seen": 143667200,
      "step": 6659,
      "time_per_iteration": 2.8535845279693604
    },
    {
      "auxiliary_loss_clip": 0.01139278,
      "auxiliary_loss_mlp": 0.0076237,
      "balance_loss_clip": 1.04015791,
      "balance_loss_mlp": 1.00052416,
      "epoch": 0.8008176516563458,
      "flos": 17164690646400.0,
      "grad_norm": 1.9742662680663114,
      "language_loss": 0.80014169,
      "learning_rate": 4.017866674771051e-07,
      "loss": 0.8191582,
      "num_input_tokens_seen": 143684685,
      "step": 6660,
      "time_per_iteration": 2.694704532623291
    },
    {
      "auxiliary_loss_clip": 0.01171022,
      "auxiliary_loss_mlp": 0.01028871,
      "balance_loss_clip": 1.04792929,
      "balance_loss_mlp": 1.02031732,
      "epoch": 0.8009378945469849,
      "flos": 24207311571840.0,
      "grad_norm": 2.453420033511315,
      "language_loss": 0.74531591,
      "learning_rate": 4.013184793183688e-07,
      "loss": 0.76731485,
      "num_input_tokens_seen": 143706780,
      "step": 6661,
      "time_per_iteration": 2.753182888031006
    },
    {
      "auxiliary_loss_clip": 0.01143691,
      "auxiliary_loss_mlp": 0.01023931,
      "balance_loss_clip": 1.04489565,
      "balance_loss_mlp": 1.01639724,
      "epoch": 0.801058137437624,
      "flos": 19787318271360.0,
      "grad_norm": 1.7013069130964815,
      "language_loss": 0.72705734,
      "learning_rate": 4.008505336783472e-07,
      "loss": 0.74873352,
      "num_input_tokens_seen": 143724505,
      "step": 6662,
      "time_per_iteration": 2.691535711288452
    },
    {
      "auxiliary_loss_clip": 0.01105277,
      "auxiliary_loss_mlp": 0.01025867,
      "balance_loss_clip": 1.03859901,
      "balance_loss_mlp": 1.01899767,
      "epoch": 0.801178380328263,
      "flos": 18659400324480.0,
      "grad_norm": 2.107236540836284,
      "language_loss": 0.80913794,
      "learning_rate": 4.003828306280284e-07,
      "loss": 0.8304494,
      "num_input_tokens_seen": 143742180,
      "step": 6663,
      "time_per_iteration": 2.812931776046753
    },
    {
      "auxiliary_loss_clip": 0.01154874,
      "auxiliary_loss_mlp": 0.01023534,
      "balance_loss_clip": 1.04493999,
      "balance_loss_mlp": 1.01627421,
      "epoch": 0.8012986232189022,
      "flos": 15706573948800.0,
      "grad_norm": 2.412753063476424,
      "language_loss": 0.78207338,
      "learning_rate": 3.999153702383626e-07,
      "loss": 0.80385745,
      "num_input_tokens_seen": 143760070,
      "step": 6664,
      "time_per_iteration": 2.685077428817749
    },
    {
      "auxiliary_loss_clip": 0.01146781,
      "auxiliary_loss_mlp": 0.00762542,
      "balance_loss_clip": 1.04430687,
      "balance_loss_mlp": 1.00058174,
      "epoch": 0.8014188661095413,
      "flos": 28584139703040.0,
      "grad_norm": 2.21199602581072,
      "language_loss": 0.74124932,
      "learning_rate": 3.9944815258026263e-07,
      "loss": 0.76034254,
      "num_input_tokens_seen": 143781890,
      "step": 6665,
      "time_per_iteration": 2.7519850730895996
    },
    {
      "auxiliary_loss_clip": 0.01159202,
      "auxiliary_loss_mlp": 0.00762439,
      "balance_loss_clip": 1.04571068,
      "balance_loss_mlp": 1.00054526,
      "epoch": 0.8015391090001803,
      "flos": 29310360877440.0,
      "grad_norm": 4.571387103735272,
      "language_loss": 0.83478272,
      "learning_rate": 3.989811777246057e-07,
      "loss": 0.85399908,
      "num_input_tokens_seen": 143802060,
      "step": 6666,
      "time_per_iteration": 2.7074570655822754
    },
    {
      "auxiliary_loss_clip": 0.01035182,
      "auxiliary_loss_mlp": 0.01002383,
      "balance_loss_clip": 1.01466393,
      "balance_loss_mlp": 1.00052381,
      "epoch": 0.8016593518908195,
      "flos": 70397340675840.0,
      "grad_norm": 0.8507505191966784,
      "language_loss": 0.66172975,
      "learning_rate": 3.985144457422305e-07,
      "loss": 0.68210542,
      "num_input_tokens_seen": 143856345,
      "step": 6667,
      "time_per_iteration": 3.289255142211914
    },
    {
      "auxiliary_loss_clip": 0.01082721,
      "auxiliary_loss_mlp": 0.01025647,
      "balance_loss_clip": 1.0416832,
      "balance_loss_mlp": 1.01771998,
      "epoch": 0.8017795947814585,
      "flos": 26026114688640.0,
      "grad_norm": 2.4552658823528044,
      "language_loss": 0.76830709,
      "learning_rate": 3.9804795670394096e-07,
      "loss": 0.7893908,
      "num_input_tokens_seen": 143876470,
      "step": 6668,
      "time_per_iteration": 4.093900918960571
    },
    {
      "auxiliary_loss_clip": 0.01154983,
      "auxiliary_loss_mlp": 0.01027592,
      "balance_loss_clip": 1.04993999,
      "balance_loss_mlp": 1.01977229,
      "epoch": 0.8018998376720976,
      "flos": 22087181260800.0,
      "grad_norm": 2.1035886841501195,
      "language_loss": 0.70582032,
      "learning_rate": 3.975817106805022e-07,
      "loss": 0.72764611,
      "num_input_tokens_seen": 143895170,
      "step": 6669,
      "time_per_iteration": 2.8019211292266846
    },
    {
      "auxiliary_loss_clip": 0.01157085,
      "auxiliary_loss_mlp": 0.01023565,
      "balance_loss_clip": 1.04700291,
      "balance_loss_mlp": 1.01510108,
      "epoch": 0.8020200805627368,
      "flos": 34568545023360.0,
      "grad_norm": 4.875413161199642,
      "language_loss": 0.65160096,
      "learning_rate": 3.97115707742645e-07,
      "loss": 0.67340744,
      "num_input_tokens_seen": 143915845,
      "step": 6670,
      "time_per_iteration": 2.7911202907562256
    },
    {
      "auxiliary_loss_clip": 0.01124489,
      "auxiliary_loss_mlp": 0.01023362,
      "balance_loss_clip": 1.04566312,
      "balance_loss_mlp": 1.01532745,
      "epoch": 0.8021403234533758,
      "flos": 20120354196480.0,
      "grad_norm": 1.9017770777448253,
      "language_loss": 0.64979047,
      "learning_rate": 3.966499479610599e-07,
      "loss": 0.671269,
      "num_input_tokens_seen": 143933940,
      "step": 6671,
      "time_per_iteration": 2.7303009033203125
    },
    {
      "auxiliary_loss_clip": 0.01127141,
      "auxiliary_loss_mlp": 0.00761772,
      "balance_loss_clip": 1.04441214,
      "balance_loss_mlp": 1.00056767,
      "epoch": 0.8022605663440149,
      "flos": 27746200252800.0,
      "grad_norm": 1.7014875155262517,
      "language_loss": 0.64849997,
      "learning_rate": 3.9618443140640225e-07,
      "loss": 0.66738909,
      "num_input_tokens_seen": 143952850,
      "step": 6672,
      "time_per_iteration": 2.8022711277008057
    },
    {
      "auxiliary_loss_clip": 0.01030752,
      "auxiliary_loss_mlp": 0.00754096,
      "balance_loss_clip": 1.01015329,
      "balance_loss_mlp": 1.00106144,
      "epoch": 0.802380809234654,
      "flos": 60244998768000.0,
      "grad_norm": 0.7576894413282491,
      "language_loss": 0.51383287,
      "learning_rate": 3.957191581492918e-07,
      "loss": 0.5316813,
      "num_input_tokens_seen": 144013610,
      "step": 6673,
      "time_per_iteration": 3.342388153076172
    },
    {
      "auxiliary_loss_clip": 0.01094673,
      "auxiliary_loss_mlp": 0.01027501,
      "balance_loss_clip": 1.03861117,
      "balance_loss_mlp": 1.01930487,
      "epoch": 0.8025010521252931,
      "flos": 15080722352640.0,
      "grad_norm": 3.0715150228701016,
      "language_loss": 0.71418047,
      "learning_rate": 3.952541282603097e-07,
      "loss": 0.73540223,
      "num_input_tokens_seen": 144028715,
      "step": 6674,
      "time_per_iteration": 2.8334949016571045
    },
    {
      "auxiliary_loss_clip": 0.01086553,
      "auxiliary_loss_mlp": 0.01027734,
      "balance_loss_clip": 1.03663504,
      "balance_loss_mlp": 1.01962209,
      "epoch": 0.8026212950159322,
      "flos": 22163527618560.0,
      "grad_norm": 2.3486265583506967,
      "language_loss": 0.83278179,
      "learning_rate": 3.9478934181000013e-07,
      "loss": 0.85392463,
      "num_input_tokens_seen": 144048740,
      "step": 6675,
      "time_per_iteration": 2.7880001068115234
    },
    {
      "auxiliary_loss_clip": 0.0112612,
      "auxiliary_loss_mlp": 0.0102675,
      "balance_loss_clip": 1.04135275,
      "balance_loss_mlp": 1.01827455,
      "epoch": 0.8027415379065713,
      "flos": 17675986792320.0,
      "grad_norm": 2.226378265363101,
      "language_loss": 0.84055275,
      "learning_rate": 3.943247988688714e-07,
      "loss": 0.86208141,
      "num_input_tokens_seen": 144067435,
      "step": 6676,
      "time_per_iteration": 2.688209056854248
    },
    {
      "auxiliary_loss_clip": 0.01123343,
      "auxiliary_loss_mlp": 0.01025067,
      "balance_loss_clip": 1.04359698,
      "balance_loss_mlp": 1.01828957,
      "epoch": 0.8028617807972104,
      "flos": 21979593048960.0,
      "grad_norm": 1.7724388521135779,
      "language_loss": 0.72121704,
      "learning_rate": 3.938604995073933e-07,
      "loss": 0.74270117,
      "num_input_tokens_seen": 144085905,
      "step": 6677,
      "time_per_iteration": 2.865591049194336
    },
    {
      "auxiliary_loss_clip": 0.01155006,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 1.04669857,
      "balance_loss_mlp": 1.01802182,
      "epoch": 0.8029820236878494,
      "flos": 26428457905920.0,
      "grad_norm": 1.6408918434427497,
      "language_loss": 0.65532929,
      "learning_rate": 3.9339644379600157e-07,
      "loss": 0.67713141,
      "num_input_tokens_seen": 144105735,
      "step": 6678,
      "time_per_iteration": 2.690091609954834
    },
    {
      "auxiliary_loss_clip": 0.01112889,
      "auxiliary_loss_mlp": 0.01025266,
      "balance_loss_clip": 1.04322159,
      "balance_loss_mlp": 1.01771998,
      "epoch": 0.8031022665784886,
      "flos": 17676489582720.0,
      "grad_norm": 2.0250213583121828,
      "language_loss": 0.7129066,
      "learning_rate": 3.929326318050907e-07,
      "loss": 0.73428822,
      "num_input_tokens_seen": 144123405,
      "step": 6679,
      "time_per_iteration": 2.832008123397827
    },
    {
      "auxiliary_loss_clip": 0.01117235,
      "auxiliary_loss_mlp": 0.01024808,
      "balance_loss_clip": 1.04360378,
      "balance_loss_mlp": 1.01714838,
      "epoch": 0.8032225094691277,
      "flos": 15450279431040.0,
      "grad_norm": 1.9881369941659852,
      "language_loss": 0.79088104,
      "learning_rate": 3.924690636050225e-07,
      "loss": 0.81230152,
      "num_input_tokens_seen": 144140815,
      "step": 6680,
      "time_per_iteration": 3.737025022506714
    },
    {
      "auxiliary_loss_clip": 0.01100464,
      "auxiliary_loss_mlp": 0.01024486,
      "balance_loss_clip": 1.04201651,
      "balance_loss_mlp": 1.01688004,
      "epoch": 0.8033427523597667,
      "flos": 26179202453760.0,
      "grad_norm": 1.827510314531956,
      "language_loss": 0.73193097,
      "learning_rate": 3.9200573926611915e-07,
      "loss": 0.75318044,
      "num_input_tokens_seen": 144162230,
      "step": 6681,
      "time_per_iteration": 3.8156580924987793
    },
    {
      "auxiliary_loss_clip": 0.01110165,
      "auxiliary_loss_mlp": 0.00762369,
      "balance_loss_clip": 1.0412209,
      "balance_loss_mlp": 1.00045788,
      "epoch": 0.8034629952504058,
      "flos": 21324905809920.0,
      "grad_norm": 6.843488719574074,
      "language_loss": 0.72923934,
      "learning_rate": 3.9154265885866613e-07,
      "loss": 0.74796468,
      "num_input_tokens_seen": 144181540,
      "step": 6682,
      "time_per_iteration": 2.8140318393707275
    },
    {
      "auxiliary_loss_clip": 0.01135037,
      "auxiliary_loss_mlp": 0.01029595,
      "balance_loss_clip": 1.04422045,
      "balance_loss_mlp": 1.021209,
      "epoch": 0.8035832381410449,
      "flos": 21651585027840.0,
      "grad_norm": 2.7207378265981244,
      "language_loss": 0.7494911,
      "learning_rate": 3.9107982245291394e-07,
      "loss": 0.77113742,
      "num_input_tokens_seen": 144199665,
      "step": 6683,
      "time_per_iteration": 2.763806104660034
    },
    {
      "auxiliary_loss_clip": 0.01130747,
      "auxiliary_loss_mlp": 0.01022391,
      "balance_loss_clip": 1.04549444,
      "balance_loss_mlp": 1.01447511,
      "epoch": 0.803703481031684,
      "flos": 20518818744960.0,
      "grad_norm": 2.1986713498758252,
      "language_loss": 0.77372795,
      "learning_rate": 3.9061723011907245e-07,
      "loss": 0.79525936,
      "num_input_tokens_seen": 144219020,
      "step": 6684,
      "time_per_iteration": 2.705151081085205
    },
    {
      "auxiliary_loss_clip": 0.01162747,
      "auxiliary_loss_mlp": 0.0102783,
      "balance_loss_clip": 1.04940796,
      "balance_loss_mlp": 1.0200727,
      "epoch": 0.803823723922323,
      "flos": 22854807838080.0,
      "grad_norm": 1.692390291297176,
      "language_loss": 0.79611844,
      "learning_rate": 3.901548819273179e-07,
      "loss": 0.81802416,
      "num_input_tokens_seen": 144239035,
      "step": 6685,
      "time_per_iteration": 2.8246726989746094
    },
    {
      "auxiliary_loss_clip": 0.01138035,
      "auxiliary_loss_mlp": 0.01022974,
      "balance_loss_clip": 1.04450321,
      "balance_loss_mlp": 1.01481414,
      "epoch": 0.8039439668129622,
      "flos": 21362145235200.0,
      "grad_norm": 2.40240204394377,
      "language_loss": 0.69454581,
      "learning_rate": 3.896927779477881e-07,
      "loss": 0.71615595,
      "num_input_tokens_seen": 144258295,
      "step": 6686,
      "time_per_iteration": 2.6942038536071777
    },
    {
      "auxiliary_loss_clip": 0.01086106,
      "auxiliary_loss_mlp": 0.01026586,
      "balance_loss_clip": 1.04025066,
      "balance_loss_mlp": 1.01877165,
      "epoch": 0.8040642097036013,
      "flos": 23802382575360.0,
      "grad_norm": 2.0959664097587645,
      "language_loss": 0.67288065,
      "learning_rate": 3.892309182505833e-07,
      "loss": 0.69400758,
      "num_input_tokens_seen": 144276110,
      "step": 6687,
      "time_per_iteration": 2.868769884109497
    },
    {
      "auxiliary_loss_clip": 0.01116498,
      "auxiliary_loss_mlp": 0.01024971,
      "balance_loss_clip": 1.04149246,
      "balance_loss_mlp": 1.01759768,
      "epoch": 0.8041844525942403,
      "flos": 25922046009600.0,
      "grad_norm": 2.2507373792218983,
      "language_loss": 0.86601847,
      "learning_rate": 3.887693029057675e-07,
      "loss": 0.88743317,
      "num_input_tokens_seen": 144295620,
      "step": 6688,
      "time_per_iteration": 2.7966320514678955
    },
    {
      "auxiliary_loss_clip": 0.01094479,
      "auxiliary_loss_mlp": 0.01023623,
      "balance_loss_clip": 1.03792334,
      "balance_loss_mlp": 1.01606548,
      "epoch": 0.8043046954848795,
      "flos": 25191120153600.0,
      "grad_norm": 1.682518877279646,
      "language_loss": 0.81440938,
      "learning_rate": 3.8830793198336684e-07,
      "loss": 0.83559042,
      "num_input_tokens_seen": 144315210,
      "step": 6689,
      "time_per_iteration": 2.7704081535339355
    },
    {
      "auxiliary_loss_clip": 0.01129394,
      "auxiliary_loss_mlp": 0.01028244,
      "balance_loss_clip": 1.03820467,
      "balance_loss_mlp": 1.02039409,
      "epoch": 0.8044249383755185,
      "flos": 41719185123840.0,
      "grad_norm": 1.6687568754525333,
      "language_loss": 0.70844769,
      "learning_rate": 3.878468055533721e-07,
      "loss": 0.7300241,
      "num_input_tokens_seen": 144337750,
      "step": 6690,
      "time_per_iteration": 2.874537467956543
    },
    {
      "auxiliary_loss_clip": 0.01130308,
      "auxiliary_loss_mlp": 0.01029729,
      "balance_loss_clip": 1.04303086,
      "balance_loss_mlp": 1.02207553,
      "epoch": 0.8045451812661576,
      "flos": 20631434860800.0,
      "grad_norm": 2.7096777796516047,
      "language_loss": 0.84686434,
      "learning_rate": 3.8738592368573464e-07,
      "loss": 0.86846471,
      "num_input_tokens_seen": 144355305,
      "step": 6691,
      "time_per_iteration": 2.7569971084594727
    },
    {
      "auxiliary_loss_clip": 0.01134033,
      "auxiliary_loss_mlp": 0.01028905,
      "balance_loss_clip": 1.04564691,
      "balance_loss_mlp": 1.02087307,
      "epoch": 0.8046654241567968,
      "flos": 29711806254720.0,
      "grad_norm": 2.053952261820916,
      "language_loss": 0.88398468,
      "learning_rate": 3.8692528645037137e-07,
      "loss": 0.90561402,
      "num_input_tokens_seen": 144374485,
      "step": 6692,
      "time_per_iteration": 2.760373830795288
    },
    {
      "auxiliary_loss_clip": 0.01113313,
      "auxiliary_loss_mlp": 0.01024106,
      "balance_loss_clip": 1.04031515,
      "balance_loss_mlp": 1.01623511,
      "epoch": 0.8047856670474358,
      "flos": 17671389851520.0,
      "grad_norm": 12.55942078056677,
      "language_loss": 0.77928537,
      "learning_rate": 3.8646489391715907e-07,
      "loss": 0.80065954,
      "num_input_tokens_seen": 144388780,
      "step": 6693,
      "time_per_iteration": 3.6014811992645264
    },
    {
      "auxiliary_loss_clip": 0.01143353,
      "auxiliary_loss_mlp": 0.01028174,
      "balance_loss_clip": 1.04565692,
      "balance_loss_mlp": 1.02051461,
      "epoch": 0.8049059099380749,
      "flos": 17120699464320.0,
      "grad_norm": 2.3692502146855445,
      "language_loss": 0.88385284,
      "learning_rate": 3.8600474615593903e-07,
      "loss": 0.90556812,
      "num_input_tokens_seen": 144403395,
      "step": 6694,
      "time_per_iteration": 2.6369338035583496
    },
    {
      "auxiliary_loss_clip": 0.01041922,
      "auxiliary_loss_mlp": 0.01003429,
      "balance_loss_clip": 1.01682472,
      "balance_loss_mlp": 1.00167692,
      "epoch": 0.805026152828714,
      "flos": 62212903240320.0,
      "grad_norm": 0.7778914609163993,
      "language_loss": 0.59606719,
      "learning_rate": 3.8554484323651605e-07,
      "loss": 0.6165207,
      "num_input_tokens_seen": 144465265,
      "step": 6695,
      "time_per_iteration": 3.293403387069702
    },
    {
      "auxiliary_loss_clip": 0.01088632,
      "auxiliary_loss_mlp": 0.01027064,
      "balance_loss_clip": 1.03668904,
      "balance_loss_mlp": 1.0197866,
      "epoch": 0.8051463957193531,
      "flos": 21688608971520.0,
      "grad_norm": 1.8871430731589258,
      "language_loss": 0.7934863,
      "learning_rate": 3.85085185228657e-07,
      "loss": 0.81464326,
      "num_input_tokens_seen": 144484235,
      "step": 6696,
      "time_per_iteration": 2.859769821166992
    },
    {
      "auxiliary_loss_clip": 0.01121077,
      "auxiliary_loss_mlp": 0.01024066,
      "balance_loss_clip": 1.04440188,
      "balance_loss_mlp": 1.01643038,
      "epoch": 0.8052666386099921,
      "flos": 32051458535040.0,
      "grad_norm": 2.024474992468899,
      "language_loss": 0.73193049,
      "learning_rate": 3.8462577220209114e-07,
      "loss": 0.75338197,
      "num_input_tokens_seen": 144504610,
      "step": 6697,
      "time_per_iteration": 2.9165287017822266
    },
    {
      "auxiliary_loss_clip": 0.01063726,
      "auxiliary_loss_mlp": 0.00753836,
      "balance_loss_clip": 1.01304913,
      "balance_loss_mlp": 1.00102234,
      "epoch": 0.8053868815006313,
      "flos": 67157875768320.0,
      "grad_norm": 0.7159138234980023,
      "language_loss": 0.58978963,
      "learning_rate": 3.8416660422651127e-07,
      "loss": 0.60796529,
      "num_input_tokens_seen": 144574260,
      "step": 6698,
      "time_per_iteration": 3.322113513946533
    },
    {
      "auxiliary_loss_clip": 0.01154737,
      "auxiliary_loss_mlp": 0.00762612,
      "balance_loss_clip": 1.04612863,
      "balance_loss_mlp": 1.00053227,
      "epoch": 0.8055071243912704,
      "flos": 23837000307840.0,
      "grad_norm": 2.6407483452928884,
      "language_loss": 0.68362635,
      "learning_rate": 3.837076813715723e-07,
      "loss": 0.7027998,
      "num_input_tokens_seen": 144594145,
      "step": 6699,
      "time_per_iteration": 2.6944162845611572
    },
    {
      "auxiliary_loss_clip": 0.01145649,
      "auxiliary_loss_mlp": 0.01029688,
      "balance_loss_clip": 1.04706311,
      "balance_loss_mlp": 1.0207895,
      "epoch": 0.8056273672819094,
      "flos": 21324510760320.0,
      "grad_norm": 1.788415594536025,
      "language_loss": 0.75379986,
      "learning_rate": 3.832490037068941e-07,
      "loss": 0.77555323,
      "num_input_tokens_seen": 144612935,
      "step": 6700,
      "time_per_iteration": 2.7781355381011963
    },
    {
      "auxiliary_loss_clip": 0.01136484,
      "auxiliary_loss_mlp": 0.01024182,
      "balance_loss_clip": 1.04153371,
      "balance_loss_mlp": 1.01666248,
      "epoch": 0.8057476101725486,
      "flos": 25768383626880.0,
      "grad_norm": 6.685514966828123,
      "language_loss": 0.76227927,
      "learning_rate": 3.827905713020554e-07,
      "loss": 0.78388596,
      "num_input_tokens_seen": 144630580,
      "step": 6701,
      "time_per_iteration": 2.7247090339660645
    },
    {
      "auxiliary_loss_clip": 0.01143119,
      "auxiliary_loss_mlp": 0.01029954,
      "balance_loss_clip": 1.0501231,
      "balance_loss_mlp": 1.02104282,
      "epoch": 0.8058678530631876,
      "flos": 24535283679360.0,
      "grad_norm": 2.046122862532861,
      "language_loss": 0.69163436,
      "learning_rate": 3.823323842266017e-07,
      "loss": 0.71336508,
      "num_input_tokens_seen": 144649975,
      "step": 6702,
      "time_per_iteration": 2.7431044578552246
    },
    {
      "auxiliary_loss_clip": 0.01105231,
      "auxiliary_loss_mlp": 0.01024155,
      "balance_loss_clip": 1.04058576,
      "balance_loss_mlp": 1.01603079,
      "epoch": 0.8059880959538267,
      "flos": 24753728240640.0,
      "grad_norm": 5.580717576687097,
      "language_loss": 0.73390388,
      "learning_rate": 3.818744425500393e-07,
      "loss": 0.75519776,
      "num_input_tokens_seen": 144667990,
      "step": 6703,
      "time_per_iteration": 2.8319811820983887
    },
    {
      "auxiliary_loss_clip": 0.01135218,
      "auxiliary_loss_mlp": 0.01024826,
      "balance_loss_clip": 1.04582345,
      "balance_loss_mlp": 1.01685715,
      "epoch": 0.8061083388444659,
      "flos": 22196349671040.0,
      "grad_norm": 2.269197526363721,
      "language_loss": 0.80718398,
      "learning_rate": 3.8141674634183675e-07,
      "loss": 0.82878447,
      "num_input_tokens_seen": 144687020,
      "step": 6704,
      "time_per_iteration": 2.7333037853240967
    },
    {
      "auxiliary_loss_clip": 0.01117716,
      "auxiliary_loss_mlp": 0.01028372,
      "balance_loss_clip": 1.04356432,
      "balance_loss_mlp": 1.02083206,
      "epoch": 0.8062285817351049,
      "flos": 30044195735040.0,
      "grad_norm": 1.9304721987985447,
      "language_loss": 0.66082692,
      "learning_rate": 3.809592956714278e-07,
      "loss": 0.68228781,
      "num_input_tokens_seen": 144710255,
      "step": 6705,
      "time_per_iteration": 2.7422966957092285
    },
    {
      "auxiliary_loss_clip": 0.01082575,
      "auxiliary_loss_mlp": 0.01023869,
      "balance_loss_clip": 1.04145455,
      "balance_loss_mlp": 1.01585269,
      "epoch": 0.806348824625744,
      "flos": 22782591544320.0,
      "grad_norm": 2.680016298186056,
      "language_loss": 0.75102496,
      "learning_rate": 3.805020906082057e-07,
      "loss": 0.77208936,
      "num_input_tokens_seen": 144728830,
      "step": 6706,
      "time_per_iteration": 3.9396541118621826
    },
    {
      "auxiliary_loss_clip": 0.01150186,
      "auxiliary_loss_mlp": 0.01028629,
      "balance_loss_clip": 1.04431891,
      "balance_loss_mlp": 1.02065396,
      "epoch": 0.8064690675163831,
      "flos": 23404600385280.0,
      "grad_norm": 5.604946434918686,
      "language_loss": 0.80747676,
      "learning_rate": 3.8004513122152917e-07,
      "loss": 0.82926494,
      "num_input_tokens_seen": 144747140,
      "step": 6707,
      "time_per_iteration": 4.7062273025512695
    },
    {
      "auxiliary_loss_clip": 0.01125502,
      "auxiliary_loss_mlp": 0.010215,
      "balance_loss_clip": 1.04271948,
      "balance_loss_mlp": 1.01416874,
      "epoch": 0.8065893104070222,
      "flos": 24060903736320.0,
      "grad_norm": 2.10418936170909,
      "language_loss": 0.67361403,
      "learning_rate": 3.79588417580718e-07,
      "loss": 0.69508398,
      "num_input_tokens_seen": 144765250,
      "step": 6708,
      "time_per_iteration": 2.715327024459839
    },
    {
      "auxiliary_loss_clip": 0.01109608,
      "auxiliary_loss_mlp": 0.01025921,
      "balance_loss_clip": 1.04101777,
      "balance_loss_mlp": 1.01791632,
      "epoch": 0.8067095532976613,
      "flos": 22305410340480.0,
      "grad_norm": 1.9051208567419649,
      "language_loss": 0.76069206,
      "learning_rate": 3.791319497550558e-07,
      "loss": 0.78204739,
      "num_input_tokens_seen": 144783080,
      "step": 6709,
      "time_per_iteration": 2.7535266876220703
    },
    {
      "auxiliary_loss_clip": 0.01129151,
      "auxiliary_loss_mlp": 0.01021554,
      "balance_loss_clip": 1.04053605,
      "balance_loss_mlp": 1.01362097,
      "epoch": 0.8068297961883004,
      "flos": 17129498296320.0,
      "grad_norm": 2.641826698846458,
      "language_loss": 0.70540881,
      "learning_rate": 3.78675727813788e-07,
      "loss": 0.72691584,
      "num_input_tokens_seen": 144800645,
      "step": 6710,
      "time_per_iteration": 2.7423739433288574
    },
    {
      "auxiliary_loss_clip": 0.01128012,
      "auxiliary_loss_mlp": 0.01024922,
      "balance_loss_clip": 1.04212809,
      "balance_loss_mlp": 1.01735198,
      "epoch": 0.8069500390789395,
      "flos": 22018843635840.0,
      "grad_norm": 2.81435387803254,
      "language_loss": 0.73370624,
      "learning_rate": 3.782197518261225e-07,
      "loss": 0.75523561,
      "num_input_tokens_seen": 144820085,
      "step": 6711,
      "time_per_iteration": 2.744102716445923
    },
    {
      "auxiliary_loss_clip": 0.01130247,
      "auxiliary_loss_mlp": 0.01023021,
      "balance_loss_clip": 1.04257202,
      "balance_loss_mlp": 1.01471829,
      "epoch": 0.8070702819695785,
      "flos": 19244241567360.0,
      "grad_norm": 2.2532034502378733,
      "language_loss": 0.95332956,
      "learning_rate": 3.777640218612319e-07,
      "loss": 0.97486222,
      "num_input_tokens_seen": 144838070,
      "step": 6712,
      "time_per_iteration": 2.6801395416259766
    },
    {
      "auxiliary_loss_clip": 0.01139302,
      "auxiliary_loss_mlp": 0.01023348,
      "balance_loss_clip": 1.04422283,
      "balance_loss_mlp": 1.01583743,
      "epoch": 0.8071905248602176,
      "flos": 21544320038400.0,
      "grad_norm": 2.5249048857570306,
      "language_loss": 0.72410607,
      "learning_rate": 3.773085379882488e-07,
      "loss": 0.74573261,
      "num_input_tokens_seen": 144857125,
      "step": 6713,
      "time_per_iteration": 2.7128610610961914
    },
    {
      "auxiliary_loss_clip": 0.01082259,
      "auxiliary_loss_mlp": 0.01033214,
      "balance_loss_clip": 1.03943825,
      "balance_loss_mlp": 1.02505457,
      "epoch": 0.8073107677508568,
      "flos": 37268309105280.0,
      "grad_norm": 1.8487140930312846,
      "language_loss": 0.75891823,
      "learning_rate": 3.768533002762715e-07,
      "loss": 0.78007293,
      "num_input_tokens_seen": 144880660,
      "step": 6714,
      "time_per_iteration": 2.979749917984009
    },
    {
      "auxiliary_loss_clip": 0.01145788,
      "auxiliary_loss_mlp": 0.0102863,
      "balance_loss_clip": 1.04337192,
      "balance_loss_mlp": 1.02095926,
      "epoch": 0.8074310106414958,
      "flos": 28366269759360.0,
      "grad_norm": 3.9927883914602678,
      "language_loss": 0.76973093,
      "learning_rate": 3.763983087943572e-07,
      "loss": 0.79147506,
      "num_input_tokens_seen": 144900050,
      "step": 6715,
      "time_per_iteration": 2.7649056911468506
    },
    {
      "auxiliary_loss_clip": 0.01136195,
      "auxiliary_loss_mlp": 0.01024376,
      "balance_loss_clip": 1.04190063,
      "balance_loss_mlp": 1.01713967,
      "epoch": 0.8075512535321349,
      "flos": 24281646768000.0,
      "grad_norm": 2.7020214215453784,
      "language_loss": 0.81045055,
      "learning_rate": 3.759435636115282e-07,
      "loss": 0.83205628,
      "num_input_tokens_seen": 144920835,
      "step": 6716,
      "time_per_iteration": 2.736603021621704
    },
    {
      "auxiliary_loss_clip": 0.01122949,
      "auxiliary_loss_mlp": 0.01024125,
      "balance_loss_clip": 1.04362047,
      "balance_loss_mlp": 1.01608992,
      "epoch": 0.807671496422774,
      "flos": 26030855283840.0,
      "grad_norm": 1.9730434866802853,
      "language_loss": 0.73207414,
      "learning_rate": 3.7548906479676967e-07,
      "loss": 0.75354487,
      "num_input_tokens_seen": 144940430,
      "step": 6717,
      "time_per_iteration": 2.809823989868164
    },
    {
      "auxiliary_loss_clip": 0.01156608,
      "auxiliary_loss_mlp": 0.01026444,
      "balance_loss_clip": 1.04393518,
      "balance_loss_mlp": 1.01793206,
      "epoch": 0.8077917393134131,
      "flos": 23730740899200.0,
      "grad_norm": 2.619638223097784,
      "language_loss": 0.71475625,
      "learning_rate": 3.7503481241902855e-07,
      "loss": 0.73658681,
      "num_input_tokens_seen": 144960405,
      "step": 6718,
      "time_per_iteration": 2.6896955966949463
    },
    {
      "auxiliary_loss_clip": 0.01126559,
      "auxiliary_loss_mlp": 0.00761507,
      "balance_loss_clip": 1.04543924,
      "balance_loss_mlp": 1.0004878,
      "epoch": 0.8079119822040521,
      "flos": 18402028398720.0,
      "grad_norm": 1.6946592022907474,
      "language_loss": 0.80421019,
      "learning_rate": 3.745808065472145e-07,
      "loss": 0.82309085,
      "num_input_tokens_seen": 144977700,
      "step": 6719,
      "time_per_iteration": 3.761537790298462
    },
    {
      "auxiliary_loss_clip": 0.01135126,
      "auxiliary_loss_mlp": 0.0076196,
      "balance_loss_clip": 1.04486012,
      "balance_loss_mlp": 1.00052786,
      "epoch": 0.8080322250946913,
      "flos": 23621787970560.0,
      "grad_norm": 1.729762615476165,
      "language_loss": 0.76367748,
      "learning_rate": 3.741270472501994e-07,
      "loss": 0.78264832,
      "num_input_tokens_seen": 144998340,
      "step": 6720,
      "time_per_iteration": 2.7443909645080566
    },
    {
      "auxiliary_loss_clip": 0.01109368,
      "auxiliary_loss_mlp": 0.01025733,
      "balance_loss_clip": 1.03931046,
      "balance_loss_mlp": 1.01865482,
      "epoch": 0.8081524679853304,
      "flos": 22820692896000.0,
      "grad_norm": 6.009639534276223,
      "language_loss": 0.73177409,
      "learning_rate": 3.736735345968183e-07,
      "loss": 0.75312507,
      "num_input_tokens_seen": 145017950,
      "step": 6721,
      "time_per_iteration": 2.798297166824341
    },
    {
      "auxiliary_loss_clip": 0.01135475,
      "auxiliary_loss_mlp": 0.00762535,
      "balance_loss_clip": 1.04378724,
      "balance_loss_mlp": 1.00051999,
      "epoch": 0.8082727108759694,
      "flos": 17640004343040.0,
      "grad_norm": 1.5262445820078523,
      "language_loss": 0.78891391,
      "learning_rate": 3.7322026865586986e-07,
      "loss": 0.80789405,
      "num_input_tokens_seen": 145036985,
      "step": 6722,
      "time_per_iteration": 2.6486318111419678
    },
    {
      "auxiliary_loss_clip": 0.01068846,
      "auxiliary_loss_mlp": 0.00762819,
      "balance_loss_clip": 1.03946304,
      "balance_loss_mlp": 1.00055528,
      "epoch": 0.8083929537666086,
      "flos": 25958172113280.0,
      "grad_norm": 1.989868994348347,
      "language_loss": 0.73832059,
      "learning_rate": 3.7276724949611206e-07,
      "loss": 0.75663722,
      "num_input_tokens_seen": 145057095,
      "step": 6723,
      "time_per_iteration": 2.918166399002075
    },
    {
      "auxiliary_loss_clip": 0.01117844,
      "auxiliary_loss_mlp": 0.0102531,
      "balance_loss_clip": 1.04142189,
      "balance_loss_mlp": 1.01675057,
      "epoch": 0.8085131966572476,
      "flos": 27089178629760.0,
      "grad_norm": 2.0510804659520803,
      "language_loss": 0.75387698,
      "learning_rate": 3.723144771862694e-07,
      "loss": 0.77530861,
      "num_input_tokens_seen": 145077735,
      "step": 6724,
      "time_per_iteration": 2.978336811065674
    },
    {
      "auxiliary_loss_clip": 0.01089094,
      "auxiliary_loss_mlp": 0.01021298,
      "balance_loss_clip": 1.04138267,
      "balance_loss_mlp": 1.01375759,
      "epoch": 0.8086334395478867,
      "flos": 23988543788160.0,
      "grad_norm": 3.3161063012117378,
      "language_loss": 0.76998556,
      "learning_rate": 3.718619517950263e-07,
      "loss": 0.79108942,
      "num_input_tokens_seen": 145098330,
      "step": 6725,
      "time_per_iteration": 2.838764190673828
    },
    {
      "auxiliary_loss_clip": 0.01078495,
      "auxiliary_loss_mlp": 0.01025342,
      "balance_loss_clip": 1.04140186,
      "balance_loss_mlp": 1.01764727,
      "epoch": 0.8087536824385259,
      "flos": 20405879406720.0,
      "grad_norm": 2.008310538328633,
      "language_loss": 0.77417171,
      "learning_rate": 3.714096733910301e-07,
      "loss": 0.79521,
      "num_input_tokens_seen": 145115855,
      "step": 6726,
      "time_per_iteration": 2.785625457763672
    },
    {
      "auxiliary_loss_clip": 0.01085064,
      "auxiliary_loss_mlp": 0.01035055,
      "balance_loss_clip": 1.04096365,
      "balance_loss_mlp": 1.02566695,
      "epoch": 0.8088739253291649,
      "flos": 25919639798400.0,
      "grad_norm": 4.6370866089486045,
      "language_loss": 0.70763016,
      "learning_rate": 3.709576420428926e-07,
      "loss": 0.72883135,
      "num_input_tokens_seen": 145136655,
      "step": 6727,
      "time_per_iteration": 2.8295652866363525
    },
    {
      "auxiliary_loss_clip": 0.01160558,
      "auxiliary_loss_mlp": 0.01021513,
      "balance_loss_clip": 1.04636025,
      "balance_loss_mlp": 1.0141573,
      "epoch": 0.808994168219804,
      "flos": 28402072640640.0,
      "grad_norm": 2.5281129249594265,
      "language_loss": 0.73494679,
      "learning_rate": 3.7050585781918463e-07,
      "loss": 0.75676751,
      "num_input_tokens_seen": 145156955,
      "step": 6728,
      "time_per_iteration": 2.827946662902832
    },
    {
      "auxiliary_loss_clip": 0.01071703,
      "auxiliary_loss_mlp": 0.01029606,
      "balance_loss_clip": 1.03795934,
      "balance_loss_mlp": 1.02098083,
      "epoch": 0.8091144111104431,
      "flos": 17421056991360.0,
      "grad_norm": 2.185081357506909,
      "language_loss": 0.68825614,
      "learning_rate": 3.700543207884428e-07,
      "loss": 0.70926923,
      "num_input_tokens_seen": 145173865,
      "step": 6729,
      "time_per_iteration": 2.9124526977539062
    },
    {
      "auxiliary_loss_clip": 0.01117023,
      "auxiliary_loss_mlp": 0.01023264,
      "balance_loss_clip": 1.04127216,
      "balance_loss_mlp": 1.0156939,
      "epoch": 0.8092346540010822,
      "flos": 32153803361280.0,
      "grad_norm": 1.760432010035572,
      "language_loss": 0.71166873,
      "learning_rate": 3.6960303101916466e-07,
      "loss": 0.73307157,
      "num_input_tokens_seen": 145193780,
      "step": 6730,
      "time_per_iteration": 3.301241397857666
    },
    {
      "auxiliary_loss_clip": 0.01045509,
      "auxiliary_loss_mlp": 0.01000535,
      "balance_loss_clip": 1.00972307,
      "balance_loss_mlp": 0.99932528,
      "epoch": 0.8093548968917212,
      "flos": 58035093390720.0,
      "grad_norm": 0.7469044931154704,
      "language_loss": 0.55505604,
      "learning_rate": 3.6915198857981047e-07,
      "loss": 0.57551646,
      "num_input_tokens_seen": 145258980,
      "step": 6731,
      "time_per_iteration": 3.3002233505249023
    },
    {
      "auxiliary_loss_clip": 0.01158569,
      "auxiliary_loss_mlp": 0.01030315,
      "balance_loss_clip": 1.05015445,
      "balance_loss_mlp": 1.02186859,
      "epoch": 0.8094751397823604,
      "flos": 27381599251200.0,
      "grad_norm": 2.0510588976274686,
      "language_loss": 0.68005097,
      "learning_rate": 3.687011935388027e-07,
      "loss": 0.70193982,
      "num_input_tokens_seen": 145281875,
      "step": 6732,
      "time_per_iteration": 3.60884952545166
    },
    {
      "auxiliary_loss_clip": 0.01066612,
      "auxiliary_loss_mlp": 0.01023278,
      "balance_loss_clip": 1.0344193,
      "balance_loss_mlp": 1.01583886,
      "epoch": 0.8095953826729995,
      "flos": 24061083304320.0,
      "grad_norm": 1.9933782710470715,
      "language_loss": 0.73201001,
      "learning_rate": 3.6825064596452646e-07,
      "loss": 0.75290895,
      "num_input_tokens_seen": 145302220,
      "step": 6733,
      "time_per_iteration": 4.800199270248413
    },
    {
      "auxiliary_loss_clip": 0.0116814,
      "auxiliary_loss_mlp": 0.01020748,
      "balance_loss_clip": 1.04788375,
      "balance_loss_mlp": 1.01358986,
      "epoch": 0.8097156255636385,
      "flos": 23951412103680.0,
      "grad_norm": 1.818393289666404,
      "language_loss": 0.70685482,
      "learning_rate": 3.678003459253305e-07,
      "loss": 0.72874367,
      "num_input_tokens_seen": 145323070,
      "step": 6734,
      "time_per_iteration": 3.229617118835449
    },
    {
      "auxiliary_loss_clip": 0.01085468,
      "auxiliary_loss_mlp": 0.01024165,
      "balance_loss_clip": 1.04054427,
      "balance_loss_mlp": 1.0162493,
      "epoch": 0.8098358684542777,
      "flos": 21799142098560.0,
      "grad_norm": 5.6650518013692075,
      "language_loss": 0.74097216,
      "learning_rate": 3.673502934895236e-07,
      "loss": 0.76206851,
      "num_input_tokens_seen": 145342575,
      "step": 6735,
      "time_per_iteration": 2.9528303146362305
    },
    {
      "auxiliary_loss_clip": 0.01030178,
      "auxiliary_loss_mlp": 0.00754733,
      "balance_loss_clip": 1.02441096,
      "balance_loss_mlp": 1.00123715,
      "epoch": 0.8099561113449167,
      "flos": 68809515966720.0,
      "grad_norm": 0.6906342388119268,
      "language_loss": 0.58035535,
      "learning_rate": 3.669004887253802e-07,
      "loss": 0.59820443,
      "num_input_tokens_seen": 145408865,
      "step": 6736,
      "time_per_iteration": 3.5138967037200928
    },
    {
      "auxiliary_loss_clip": 0.01124236,
      "auxiliary_loss_mlp": 0.00762108,
      "balance_loss_clip": 1.04601276,
      "balance_loss_mlp": 1.00053573,
      "epoch": 0.8100763542355558,
      "flos": 23586056916480.0,
      "grad_norm": 1.7170326701084302,
      "language_loss": 0.79277748,
      "learning_rate": 3.664509317011335e-07,
      "loss": 0.81164092,
      "num_input_tokens_seen": 145429200,
      "step": 6737,
      "time_per_iteration": 2.8065028190612793
    },
    {
      "auxiliary_loss_clip": 0.01137438,
      "auxiliary_loss_mlp": 0.01033857,
      "balance_loss_clip": 1.04498541,
      "balance_loss_mlp": 1.02489865,
      "epoch": 0.810196597126195,
      "flos": 31650408207360.0,
      "grad_norm": 2.3580499748622996,
      "language_loss": 0.74012375,
      "learning_rate": 3.6600162248498134e-07,
      "loss": 0.76183671,
      "num_input_tokens_seen": 145452830,
      "step": 6738,
      "time_per_iteration": 2.8228790760040283
    },
    {
      "auxiliary_loss_clip": 0.0111763,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.03860331,
      "balance_loss_mlp": 1.01973927,
      "epoch": 0.810316840016834,
      "flos": 24900459298560.0,
      "grad_norm": 2.692748658009543,
      "language_loss": 0.76205498,
      "learning_rate": 3.6555256114508426e-07,
      "loss": 0.78349954,
      "num_input_tokens_seen": 145472625,
      "step": 6739,
      "time_per_iteration": 2.7888667583465576
    },
    {
      "auxiliary_loss_clip": 0.01129185,
      "auxiliary_loss_mlp": 0.0102456,
      "balance_loss_clip": 1.04298973,
      "balance_loss_mlp": 1.01623929,
      "epoch": 0.8104370829074731,
      "flos": 27965003950080.0,
      "grad_norm": 2.4574408025368517,
      "language_loss": 0.73171961,
      "learning_rate": 3.651037477495642e-07,
      "loss": 0.7532571,
      "num_input_tokens_seen": 145494075,
      "step": 6740,
      "time_per_iteration": 2.818256378173828
    },
    {
      "auxiliary_loss_clip": 0.0112677,
      "auxiliary_loss_mlp": 0.01023652,
      "balance_loss_clip": 1.0394876,
      "balance_loss_mlp": 1.01570106,
      "epoch": 0.8105573257981122,
      "flos": 24640752988800.0,
      "grad_norm": 2.1164299698394338,
      "language_loss": 0.6838761,
      "learning_rate": 3.6465518236650584e-07,
      "loss": 0.70538032,
      "num_input_tokens_seen": 145514220,
      "step": 6741,
      "time_per_iteration": 2.78474497795105
    },
    {
      "auxiliary_loss_clip": 0.01127906,
      "auxiliary_loss_mlp": 0.00761567,
      "balance_loss_clip": 1.0439626,
      "balance_loss_mlp": 1.00045264,
      "epoch": 0.8106775686887513,
      "flos": 26358935132160.0,
      "grad_norm": 2.1379954370965435,
      "language_loss": 0.7841177,
      "learning_rate": 3.642068650639558e-07,
      "loss": 0.80301249,
      "num_input_tokens_seen": 145533965,
      "step": 6742,
      "time_per_iteration": 2.8119542598724365
    },
    {
      "auxiliary_loss_clip": 0.01095042,
      "auxiliary_loss_mlp": 0.01023597,
      "balance_loss_clip": 1.03682852,
      "balance_loss_mlp": 1.0161171,
      "epoch": 0.8107978115793903,
      "flos": 27271892136960.0,
      "grad_norm": 2.100099854803366,
      "language_loss": 0.64767718,
      "learning_rate": 3.6375879590992334e-07,
      "loss": 0.66886353,
      "num_input_tokens_seen": 145554310,
      "step": 6743,
      "time_per_iteration": 2.8123788833618164
    },
    {
      "auxiliary_loss_clip": 0.01172853,
      "auxiliary_loss_mlp": 0.00762516,
      "balance_loss_clip": 1.04871237,
      "balance_loss_mlp": 1.00049496,
      "epoch": 0.8109180544700295,
      "flos": 24934322845440.0,
      "grad_norm": 2.0217294985412217,
      "language_loss": 0.81261188,
      "learning_rate": 3.6331097497238173e-07,
      "loss": 0.83196557,
      "num_input_tokens_seen": 145573755,
      "step": 6744,
      "time_per_iteration": 3.772015333175659
    },
    {
      "auxiliary_loss_clip": 0.01136012,
      "auxiliary_loss_mlp": 0.01026321,
      "balance_loss_clip": 1.03991079,
      "balance_loss_mlp": 1.01894546,
      "epoch": 0.8110382973606686,
      "flos": 21105383840640.0,
      "grad_norm": 2.137507230086795,
      "language_loss": 0.80366355,
      "learning_rate": 3.628634023192627e-07,
      "loss": 0.82528687,
      "num_input_tokens_seen": 145594000,
      "step": 6745,
      "time_per_iteration": 2.73640775680542
    },
    {
      "auxiliary_loss_clip": 0.01115892,
      "auxiliary_loss_mlp": 0.0102253,
      "balance_loss_clip": 1.03886557,
      "balance_loss_mlp": 1.01451957,
      "epoch": 0.8111585402513076,
      "flos": 15414081500160.0,
      "grad_norm": 2.0259644287671694,
      "language_loss": 0.75224733,
      "learning_rate": 3.624160780184644e-07,
      "loss": 0.77363157,
      "num_input_tokens_seen": 145611215,
      "step": 6746,
      "time_per_iteration": 2.7509312629699707
    },
    {
      "auxiliary_loss_clip": 0.01143919,
      "auxiliary_loss_mlp": 0.01021166,
      "balance_loss_clip": 1.04456973,
      "balance_loss_mlp": 1.01295269,
      "epoch": 0.8112787831419467,
      "flos": 24095736950400.0,
      "grad_norm": 2.2026198044713117,
      "language_loss": 0.74616635,
      "learning_rate": 3.6196900213784496e-07,
      "loss": 0.76781726,
      "num_input_tokens_seen": 145630530,
      "step": 6747,
      "time_per_iteration": 2.733983039855957
    },
    {
      "auxiliary_loss_clip": 0.01121093,
      "auxiliary_loss_mlp": 0.01024747,
      "balance_loss_clip": 1.04185581,
      "balance_loss_mlp": 1.01705194,
      "epoch": 0.8113990260325858,
      "flos": 20483374999680.0,
      "grad_norm": 2.1265478890868676,
      "language_loss": 0.86415815,
      "learning_rate": 3.6152217474522527e-07,
      "loss": 0.88561654,
      "num_input_tokens_seen": 145647345,
      "step": 6748,
      "time_per_iteration": 2.7322657108306885
    },
    {
      "auxiliary_loss_clip": 0.01083733,
      "auxiliary_loss_mlp": 0.01028864,
      "balance_loss_clip": 1.03959346,
      "balance_loss_mlp": 1.02155638,
      "epoch": 0.8115192689232249,
      "flos": 24901141656960.0,
      "grad_norm": 1.6245379772128632,
      "language_loss": 0.72921109,
      "learning_rate": 3.6107559590838975e-07,
      "loss": 0.75033706,
      "num_input_tokens_seen": 145666330,
      "step": 6749,
      "time_per_iteration": 2.9193930625915527
    },
    {
      "auxiliary_loss_clip": 0.0111144,
      "auxiliary_loss_mlp": 0.01028586,
      "balance_loss_clip": 1.0389576,
      "balance_loss_mlp": 1.02058697,
      "epoch": 0.811639511813864,
      "flos": 24057204635520.0,
      "grad_norm": 2.578844047306769,
      "language_loss": 0.668257,
      "learning_rate": 3.606292656950822e-07,
      "loss": 0.68965727,
      "num_input_tokens_seen": 145684740,
      "step": 6750,
      "time_per_iteration": 2.718980312347412
    },
    {
      "auxiliary_loss_clip": 0.01118375,
      "auxiliary_loss_mlp": 0.00762417,
      "balance_loss_clip": 1.0408994,
      "balance_loss_mlp": 1.00054204,
      "epoch": 0.8117597547045031,
      "flos": 23185150243200.0,
      "grad_norm": 2.0577019990685157,
      "language_loss": 0.86669934,
      "learning_rate": 3.601831841730121e-07,
      "loss": 0.88550723,
      "num_input_tokens_seen": 145702660,
      "step": 6751,
      "time_per_iteration": 2.739884614944458
    },
    {
      "auxiliary_loss_clip": 0.01152367,
      "auxiliary_loss_mlp": 0.00762362,
      "balance_loss_clip": 1.04518771,
      "balance_loss_mlp": 1.00049019,
      "epoch": 0.8118799975951422,
      "flos": 23040250778880.0,
      "grad_norm": 1.8630318982093703,
      "language_loss": 0.73017919,
      "learning_rate": 3.5973735140984916e-07,
      "loss": 0.74932647,
      "num_input_tokens_seen": 145722830,
      "step": 6752,
      "time_per_iteration": 2.637025833129883
    },
    {
      "auxiliary_loss_clip": 0.01058311,
      "auxiliary_loss_mlp": 0.01027384,
      "balance_loss_clip": 1.03659225,
      "balance_loss_mlp": 1.02012992,
      "epoch": 0.8120002404857812,
      "flos": 24639962889600.0,
      "grad_norm": 2.103451552665598,
      "language_loss": 0.80028224,
      "learning_rate": 3.5929176747322607e-07,
      "loss": 0.82113922,
      "num_input_tokens_seen": 145741935,
      "step": 6753,
      "time_per_iteration": 2.989349126815796
    },
    {
      "auxiliary_loss_clip": 0.01035264,
      "auxiliary_loss_mlp": 0.01002778,
      "balance_loss_clip": 1.01169753,
      "balance_loss_mlp": 1.00162745,
      "epoch": 0.8121204833764204,
      "flos": 57415742156160.0,
      "grad_norm": 0.8053253524831532,
      "language_loss": 0.56227225,
      "learning_rate": 3.588464324307372e-07,
      "loss": 0.58265263,
      "num_input_tokens_seen": 145805560,
      "step": 6754,
      "time_per_iteration": 4.384123086929321
    },
    {
      "auxiliary_loss_clip": 0.01114786,
      "auxiliary_loss_mlp": 0.01021663,
      "balance_loss_clip": 1.04398191,
      "balance_loss_mlp": 1.01393557,
      "epoch": 0.8122407262670595,
      "flos": 19464589549440.0,
      "grad_norm": 2.417722507785576,
      "language_loss": 0.75588638,
      "learning_rate": 3.584013463499391e-07,
      "loss": 0.77725089,
      "num_input_tokens_seen": 145824180,
      "step": 6755,
      "time_per_iteration": 2.8217129707336426
    },
    {
      "auxiliary_loss_clip": 0.01061637,
      "auxiliary_loss_mlp": 0.0100092,
      "balance_loss_clip": 1.01190352,
      "balance_loss_mlp": 0.99975759,
      "epoch": 0.8123609691576985,
      "flos": 56425325472000.0,
      "grad_norm": 0.7322604069574086,
      "language_loss": 0.64441788,
      "learning_rate": 3.579565092983521e-07,
      "loss": 0.66504347,
      "num_input_tokens_seen": 145885300,
      "step": 6756,
      "time_per_iteration": 3.121438980102539
    },
    {
      "auxiliary_loss_clip": 0.01109206,
      "auxiliary_loss_mlp": 0.01025497,
      "balance_loss_clip": 1.03665042,
      "balance_loss_mlp": 1.01809692,
      "epoch": 0.8124812120483377,
      "flos": 20631973564800.0,
      "grad_norm": 2.1095440797762097,
      "language_loss": 0.83935034,
      "learning_rate": 3.575119213434565e-07,
      "loss": 0.86069733,
      "num_input_tokens_seen": 145903815,
      "step": 6757,
      "time_per_iteration": 2.8194892406463623
    },
    {
      "auxiliary_loss_clip": 0.01102772,
      "auxiliary_loss_mlp": 0.01029117,
      "balance_loss_clip": 1.04018188,
      "balance_loss_mlp": 1.02181232,
      "epoch": 0.8126014549389767,
      "flos": 22492397566080.0,
      "grad_norm": 1.9738886984227755,
      "language_loss": 0.82151842,
      "learning_rate": 3.5706758255269765e-07,
      "loss": 0.84283733,
      "num_input_tokens_seen": 145922270,
      "step": 6758,
      "time_per_iteration": 4.6511430740356445
    },
    {
      "auxiliary_loss_clip": 0.01104415,
      "auxiliary_loss_mlp": 0.01029715,
      "balance_loss_clip": 1.04302359,
      "balance_loss_mlp": 1.02138853,
      "epoch": 0.8127216978296158,
      "flos": 23287961946240.0,
      "grad_norm": 1.6971971234719851,
      "language_loss": 0.69956887,
      "learning_rate": 3.566234929934795e-07,
      "loss": 0.72091019,
      "num_input_tokens_seen": 145941470,
      "step": 6759,
      "time_per_iteration": 3.6868345737457275
    },
    {
      "auxiliary_loss_clip": 0.01108749,
      "auxiliary_loss_mlp": 0.01024834,
      "balance_loss_clip": 1.03776145,
      "balance_loss_mlp": 1.0169667,
      "epoch": 0.812841940720255,
      "flos": 25154994049920.0,
      "grad_norm": 1.4413834005146564,
      "language_loss": 0.71647614,
      "learning_rate": 3.561796527331706e-07,
      "loss": 0.73781198,
      "num_input_tokens_seen": 145963145,
      "step": 6760,
      "time_per_iteration": 2.799614429473877
    },
    {
      "auxiliary_loss_clip": 0.01119796,
      "auxiliary_loss_mlp": 0.01022664,
      "balance_loss_clip": 1.04102707,
      "balance_loss_mlp": 1.01469517,
      "epoch": 0.812962183610894,
      "flos": 26648446752000.0,
      "grad_norm": 2.039129762517899,
      "language_loss": 0.7764557,
      "learning_rate": 3.5573606183910163e-07,
      "loss": 0.79788029,
      "num_input_tokens_seen": 145983150,
      "step": 6761,
      "time_per_iteration": 2.8565986156463623
    },
    {
      "auxiliary_loss_clip": 0.01050459,
      "auxiliary_loss_mlp": 0.01026315,
      "balance_loss_clip": 1.03222466,
      "balance_loss_mlp": 1.01835799,
      "epoch": 0.8130824265015331,
      "flos": 24966965329920.0,
      "grad_norm": 1.8912101881058787,
      "language_loss": 0.78910768,
      "learning_rate": 3.5529272037856493e-07,
      "loss": 0.80987549,
      "num_input_tokens_seen": 146001365,
      "step": 6762,
      "time_per_iteration": 3.2912325859069824
    },
    {
      "auxiliary_loss_clip": 0.01024106,
      "auxiliary_loss_mlp": 0.01001199,
      "balance_loss_clip": 1.01108575,
      "balance_loss_mlp": 0.99989396,
      "epoch": 0.8132026693921722,
      "flos": 67622918175360.0,
      "grad_norm": 0.7201996601044541,
      "language_loss": 0.53761667,
      "learning_rate": 3.548496284188149e-07,
      "loss": 0.55786967,
      "num_input_tokens_seen": 146061570,
      "step": 6763,
      "time_per_iteration": 3.599233865737915
    },
    {
      "auxiliary_loss_clip": 0.01112605,
      "auxiliary_loss_mlp": 0.01023363,
      "balance_loss_clip": 1.03888464,
      "balance_loss_mlp": 1.0156498,
      "epoch": 0.8133229122828113,
      "flos": 19495149045120.0,
      "grad_norm": 1.8187394788969133,
      "language_loss": 0.79559648,
      "learning_rate": 3.544067860270681e-07,
      "loss": 0.81695604,
      "num_input_tokens_seen": 146079145,
      "step": 6764,
      "time_per_iteration": 2.7388453483581543
    },
    {
      "auxiliary_loss_clip": 0.01146805,
      "auxiliary_loss_mlp": 0.01026092,
      "balance_loss_clip": 1.04452586,
      "balance_loss_mlp": 1.01768768,
      "epoch": 0.8134431551734503,
      "flos": 20668135582080.0,
      "grad_norm": 1.9482392892246996,
      "language_loss": 0.7121796,
      "learning_rate": 3.539641932705029e-07,
      "loss": 0.73390859,
      "num_input_tokens_seen": 146097625,
      "step": 6765,
      "time_per_iteration": 2.66463565826416
    },
    {
      "auxiliary_loss_clip": 0.01073886,
      "auxiliary_loss_mlp": 0.01027944,
      "balance_loss_clip": 1.03737855,
      "balance_loss_mlp": 1.01917601,
      "epoch": 0.8135633980640895,
      "flos": 21507332008320.0,
      "grad_norm": 2.180185575016654,
      "language_loss": 0.77103031,
      "learning_rate": 3.53521850216262e-07,
      "loss": 0.79204857,
      "num_input_tokens_seen": 146117195,
      "step": 6766,
      "time_per_iteration": 2.845231294631958
    },
    {
      "auxiliary_loss_clip": 0.01160524,
      "auxiliary_loss_mlp": 0.01024356,
      "balance_loss_clip": 1.04819012,
      "balance_loss_mlp": 1.01633918,
      "epoch": 0.8136836409547286,
      "flos": 20554442058240.0,
      "grad_norm": 3.133210066603323,
      "language_loss": 0.76878548,
      "learning_rate": 3.530797569314461e-07,
      "loss": 0.79063427,
      "num_input_tokens_seen": 146136220,
      "step": 6767,
      "time_per_iteration": 2.6901731491088867
    },
    {
      "auxiliary_loss_clip": 0.01157188,
      "auxiliary_loss_mlp": 0.0102616,
      "balance_loss_clip": 1.04620731,
      "balance_loss_mlp": 1.01877463,
      "epoch": 0.8138038838453676,
      "flos": 20299045380480.0,
      "grad_norm": 2.4404970283213916,
      "language_loss": 0.77662385,
      "learning_rate": 3.5263791348312235e-07,
      "loss": 0.79845726,
      "num_input_tokens_seen": 146155415,
      "step": 6768,
      "time_per_iteration": 2.695261001586914
    },
    {
      "auxiliary_loss_clip": 0.01138159,
      "auxiliary_loss_mlp": 0.00762131,
      "balance_loss_clip": 1.04378963,
      "balance_loss_mlp": 1.0005697,
      "epoch": 0.8139241267360068,
      "flos": 29789840551680.0,
      "grad_norm": 2.2288004149533736,
      "language_loss": 0.7068302,
      "learning_rate": 3.521963199383171e-07,
      "loss": 0.72583306,
      "num_input_tokens_seen": 146178370,
      "step": 6769,
      "time_per_iteration": 2.7737743854522705
    },
    {
      "auxiliary_loss_clip": 0.01173178,
      "auxiliary_loss_mlp": 0.01020546,
      "balance_loss_clip": 1.0486424,
      "balance_loss_mlp": 1.01254702,
      "epoch": 0.8140443696266458,
      "flos": 19713270384000.0,
      "grad_norm": 2.8240704837788795,
      "language_loss": 0.77157092,
      "learning_rate": 3.517549763640197e-07,
      "loss": 0.79350817,
      "num_input_tokens_seen": 146196010,
      "step": 6770,
      "time_per_iteration": 4.86752724647522
    },
    {
      "auxiliary_loss_clip": 0.01132911,
      "auxiliary_loss_mlp": 0.01029786,
      "balance_loss_clip": 1.04491603,
      "balance_loss_mlp": 1.02226436,
      "epoch": 0.8141646125172849,
      "flos": 27160568910720.0,
      "grad_norm": 4.37886691396201,
      "language_loss": 0.70939428,
      "learning_rate": 3.513138828271829e-07,
      "loss": 0.73102123,
      "num_input_tokens_seen": 146215880,
      "step": 6771,
      "time_per_iteration": 2.922849178314209
    },
    {
      "auxiliary_loss_clip": 0.01083067,
      "auxiliary_loss_mlp": 0.01025963,
      "balance_loss_clip": 1.03479791,
      "balance_loss_mlp": 1.01798761,
      "epoch": 0.8142848554079241,
      "flos": 39673102700160.0,
      "grad_norm": 2.669300048707549,
      "language_loss": 0.70283246,
      "learning_rate": 3.508730393947179e-07,
      "loss": 0.72392273,
      "num_input_tokens_seen": 146239135,
      "step": 6772,
      "time_per_iteration": 3.012685537338257
    },
    {
      "auxiliary_loss_clip": 0.01168601,
      "auxiliary_loss_mlp": 0.01021453,
      "balance_loss_clip": 1.04869533,
      "balance_loss_mlp": 1.01370752,
      "epoch": 0.8144050982985631,
      "flos": 22237288197120.0,
      "grad_norm": 1.8301475078795064,
      "language_loss": 0.71965837,
      "learning_rate": 3.504324461335024e-07,
      "loss": 0.74155891,
      "num_input_tokens_seen": 146259245,
      "step": 6773,
      "time_per_iteration": 2.8064773082733154
    },
    {
      "auxiliary_loss_clip": 0.01146273,
      "auxiliary_loss_mlp": 0.01027475,
      "balance_loss_clip": 1.04633677,
      "balance_loss_mlp": 1.01885653,
      "epoch": 0.8145253411892022,
      "flos": 23038239617280.0,
      "grad_norm": 1.7775259829903047,
      "language_loss": 0.88523698,
      "learning_rate": 3.499921031103732e-07,
      "loss": 0.90697443,
      "num_input_tokens_seen": 146280015,
      "step": 6774,
      "time_per_iteration": 2.71319317817688
    },
    {
      "auxiliary_loss_clip": 0.01081664,
      "auxiliary_loss_mlp": 0.01026409,
      "balance_loss_clip": 1.03923202,
      "balance_loss_mlp": 1.01868403,
      "epoch": 0.8146455840798413,
      "flos": 24827668387200.0,
      "grad_norm": 1.9058617615506857,
      "language_loss": 0.78598809,
      "learning_rate": 3.4955201039212987e-07,
      "loss": 0.80706882,
      "num_input_tokens_seen": 146300935,
      "step": 6775,
      "time_per_iteration": 3.027911901473999
    },
    {
      "auxiliary_loss_clip": 0.01116643,
      "auxiliary_loss_mlp": 0.01031872,
      "balance_loss_clip": 1.0476644,
      "balance_loss_mlp": 1.02305686,
      "epoch": 0.8147658269704804,
      "flos": 19974520978560.0,
      "grad_norm": 2.048902876427154,
      "language_loss": 0.6567145,
      "learning_rate": 3.4911216804553465e-07,
      "loss": 0.67819965,
      "num_input_tokens_seen": 146319835,
      "step": 6776,
      "time_per_iteration": 2.8115601539611816
    },
    {
      "auxiliary_loss_clip": 0.01147208,
      "auxiliary_loss_mlp": 0.01026356,
      "balance_loss_clip": 1.04602194,
      "balance_loss_mlp": 1.01768696,
      "epoch": 0.8148860698611194,
      "flos": 21178031097600.0,
      "grad_norm": 2.183481942811383,
      "language_loss": 0.70228577,
      "learning_rate": 3.4867257613731017e-07,
      "loss": 0.72402138,
      "num_input_tokens_seen": 146339030,
      "step": 6777,
      "time_per_iteration": 2.643439769744873
    },
    {
      "auxiliary_loss_clip": 0.01155387,
      "auxiliary_loss_mlp": 0.01020568,
      "balance_loss_clip": 1.04667866,
      "balance_loss_mlp": 1.0131948,
      "epoch": 0.8150063127517585,
      "flos": 19606903234560.0,
      "grad_norm": 1.7823291607266725,
      "language_loss": 0.85646874,
      "learning_rate": 3.4823323473414343e-07,
      "loss": 0.87822831,
      "num_input_tokens_seen": 146358550,
      "step": 6778,
      "time_per_iteration": 2.6706011295318604
    },
    {
      "auxiliary_loss_clip": 0.01099142,
      "auxiliary_loss_mlp": 0.01024763,
      "balance_loss_clip": 1.04048514,
      "balance_loss_mlp": 1.01531553,
      "epoch": 0.8151265556423977,
      "flos": 22638374438400.0,
      "grad_norm": 2.645649315577339,
      "language_loss": 0.75940037,
      "learning_rate": 3.477941439026812e-07,
      "loss": 0.78063953,
      "num_input_tokens_seen": 146376770,
      "step": 6779,
      "time_per_iteration": 2.7761690616607666
    },
    {
      "auxiliary_loss_clip": 0.01121141,
      "auxiliary_loss_mlp": 0.01024921,
      "balance_loss_clip": 1.04774141,
      "balance_loss_mlp": 1.01759005,
      "epoch": 0.8152467985330367,
      "flos": 17968048277760.0,
      "grad_norm": 1.784003233074465,
      "language_loss": 0.73075378,
      "learning_rate": 3.473553037095349e-07,
      "loss": 0.75221443,
      "num_input_tokens_seen": 146395795,
      "step": 6780,
      "time_per_iteration": 2.7118797302246094
    },
    {
      "auxiliary_loss_clip": 0.01122479,
      "auxiliary_loss_mlp": 0.00761376,
      "balance_loss_clip": 1.04409337,
      "balance_loss_mlp": 1.00044608,
      "epoch": 0.8153670414236758,
      "flos": 24969012405120.0,
      "grad_norm": 1.8911213228107149,
      "language_loss": 0.83394027,
      "learning_rate": 3.469167142212743e-07,
      "loss": 0.85277879,
      "num_input_tokens_seen": 146417640,
      "step": 6781,
      "time_per_iteration": 2.8040220737457275
    },
    {
      "auxiliary_loss_clip": 0.0117231,
      "auxiliary_loss_mlp": 0.0102921,
      "balance_loss_clip": 1.04893827,
      "balance_loss_mlp": 1.02112174,
      "epoch": 0.8154872843143149,
      "flos": 31066069754880.0,
      "grad_norm": 2.963359611811099,
      "language_loss": 0.63035697,
      "learning_rate": 3.4647837550443337e-07,
      "loss": 0.65237218,
      "num_input_tokens_seen": 146436205,
      "step": 6782,
      "time_per_iteration": 2.7219173908233643
    },
    {
      "auxiliary_loss_clip": 0.01124291,
      "auxiliary_loss_mlp": 0.01024873,
      "balance_loss_clip": 1.04274189,
      "balance_loss_mlp": 1.01721382,
      "epoch": 0.815607527204954,
      "flos": 19391654983680.0,
      "grad_norm": 3.120396252939884,
      "language_loss": 0.74902785,
      "learning_rate": 3.460402876255086e-07,
      "loss": 0.7705195,
      "num_input_tokens_seen": 146453595,
      "step": 6783,
      "time_per_iteration": 3.7068097591400146
    },
    {
      "auxiliary_loss_clip": 0.01118566,
      "auxiliary_loss_mlp": 0.01021511,
      "balance_loss_clip": 1.042377,
      "balance_loss_mlp": 1.0137862,
      "epoch": 0.815727770095593,
      "flos": 26140418743680.0,
      "grad_norm": 2.1385740860495193,
      "language_loss": 0.72098339,
      "learning_rate": 3.456024506509574e-07,
      "loss": 0.74238414,
      "num_input_tokens_seen": 146474515,
      "step": 6784,
      "time_per_iteration": 4.671556234359741
    },
    {
      "auxiliary_loss_clip": 0.01090962,
      "auxiliary_loss_mlp": 0.01025994,
      "balance_loss_clip": 1.03865361,
      "balance_loss_mlp": 1.01807857,
      "epoch": 0.8158480129862322,
      "flos": 25337527989120.0,
      "grad_norm": 1.7060251578707437,
      "language_loss": 0.73772299,
      "learning_rate": 3.4516486464719873e-07,
      "loss": 0.75889254,
      "num_input_tokens_seen": 146493905,
      "step": 6785,
      "time_per_iteration": 2.8663785457611084
    },
    {
      "auxiliary_loss_clip": 0.01099131,
      "auxiliary_loss_mlp": 0.00762649,
      "balance_loss_clip": 1.04130781,
      "balance_loss_mlp": 1.00050831,
      "epoch": 0.8159682558768713,
      "flos": 34423645559040.0,
      "grad_norm": 1.7525790796122693,
      "language_loss": 0.62041092,
      "learning_rate": 3.4472752968061445e-07,
      "loss": 0.63902879,
      "num_input_tokens_seen": 146518335,
      "step": 6786,
      "time_per_iteration": 2.9194352626800537
    },
    {
      "auxiliary_loss_clip": 0.01138123,
      "auxiliary_loss_mlp": 0.01024603,
      "balance_loss_clip": 1.04680157,
      "balance_loss_mlp": 1.01645494,
      "epoch": 0.8160884987675103,
      "flos": 18653223185280.0,
      "grad_norm": 2.3839808570081598,
      "language_loss": 0.73918623,
      "learning_rate": 3.442904458175475e-07,
      "loss": 0.76081347,
      "num_input_tokens_seen": 146535655,
      "step": 6787,
      "time_per_iteration": 2.6581966876983643
    },
    {
      "auxiliary_loss_clip": 0.01150654,
      "auxiliary_loss_mlp": 0.00762564,
      "balance_loss_clip": 1.0452987,
      "balance_loss_mlp": 1.0005362,
      "epoch": 0.8162087416581495,
      "flos": 31430527102080.0,
      "grad_norm": 1.611958710779965,
      "language_loss": 0.76186484,
      "learning_rate": 3.438536131243044e-07,
      "loss": 0.7809971,
      "num_input_tokens_seen": 146556815,
      "step": 6788,
      "time_per_iteration": 2.7350292205810547
    },
    {
      "auxiliary_loss_clip": 0.0116141,
      "auxiliary_loss_mlp": 0.00763349,
      "balance_loss_clip": 1.04527712,
      "balance_loss_mlp": 1.00045514,
      "epoch": 0.8163289845487885,
      "flos": 37593910915200.0,
      "grad_norm": 2.569408230941796,
      "language_loss": 0.6176213,
      "learning_rate": 3.434170316671503e-07,
      "loss": 0.63686889,
      "num_input_tokens_seen": 146581845,
      "step": 6789,
      "time_per_iteration": 2.806349754333496
    },
    {
      "auxiliary_loss_clip": 0.01110207,
      "auxiliary_loss_mlp": 0.01021621,
      "balance_loss_clip": 1.04153728,
      "balance_loss_mlp": 1.01443565,
      "epoch": 0.8164492274394276,
      "flos": 13953989554560.0,
      "grad_norm": 2.650501285298303,
      "language_loss": 0.89911872,
      "learning_rate": 3.4298070151231583e-07,
      "loss": 0.92043698,
      "num_input_tokens_seen": 146597245,
      "step": 6790,
      "time_per_iteration": 2.8007407188415527
    },
    {
      "auxiliary_loss_clip": 0.01168641,
      "auxiliary_loss_mlp": 0.0102516,
      "balance_loss_clip": 1.04596353,
      "balance_loss_mlp": 1.01718497,
      "epoch": 0.8165694703300668,
      "flos": 28986554747520.0,
      "grad_norm": 1.9626209418956377,
      "language_loss": 0.59870648,
      "learning_rate": 3.425446227259916e-07,
      "loss": 0.62064451,
      "num_input_tokens_seen": 146618210,
      "step": 6791,
      "time_per_iteration": 2.684788703918457
    },
    {
      "auxiliary_loss_clip": 0.01136361,
      "auxiliary_loss_mlp": 0.01023514,
      "balance_loss_clip": 1.04585552,
      "balance_loss_mlp": 1.01575649,
      "epoch": 0.8166897132207058,
      "flos": 25118365155840.0,
      "grad_norm": 1.9438944255504118,
      "language_loss": 0.82473516,
      "learning_rate": 3.421087953743296e-07,
      "loss": 0.84633386,
      "num_input_tokens_seen": 146637975,
      "step": 6792,
      "time_per_iteration": 2.690943717956543
    },
    {
      "auxiliary_loss_clip": 0.01135488,
      "auxiliary_loss_mlp": 0.00763244,
      "balance_loss_clip": 1.04206085,
      "balance_loss_mlp": 1.00050199,
      "epoch": 0.8168099561113449,
      "flos": 23148593176320.0,
      "grad_norm": 3.1757237459897003,
      "language_loss": 0.80194789,
      "learning_rate": 3.416732195234464e-07,
      "loss": 0.82093519,
      "num_input_tokens_seen": 146658030,
      "step": 6793,
      "time_per_iteration": 2.786339044570923
    },
    {
      "auxiliary_loss_clip": 0.01152996,
      "auxiliary_loss_mlp": 0.01027253,
      "balance_loss_clip": 1.04325008,
      "balance_loss_mlp": 1.0196768,
      "epoch": 0.816930199001984,
      "flos": 18407666833920.0,
      "grad_norm": 1.4486776025217025,
      "language_loss": 0.79499859,
      "learning_rate": 3.4123789523941613e-07,
      "loss": 0.81680107,
      "num_input_tokens_seen": 146677855,
      "step": 6794,
      "time_per_iteration": 2.6814522743225098
    },
    {
      "auxiliary_loss_clip": 0.01052739,
      "auxiliary_loss_mlp": 0.00762913,
      "balance_loss_clip": 1.03209567,
      "balance_loss_mlp": 1.00051415,
      "epoch": 0.8170504418926231,
      "flos": 21251324799360.0,
      "grad_norm": 1.844713963366405,
      "language_loss": 0.63480169,
      "learning_rate": 3.4080282258827884e-07,
      "loss": 0.65295827,
      "num_input_tokens_seen": 146696230,
      "step": 6795,
      "time_per_iteration": 3.8657307624816895
    },
    {
      "auxiliary_loss_clip": 0.01109905,
      "auxiliary_loss_mlp": 0.01024127,
      "balance_loss_clip": 1.04063261,
      "balance_loss_mlp": 1.01648605,
      "epoch": 0.8171706847832622,
      "flos": 19099234362240.0,
      "grad_norm": 2.144550603931844,
      "language_loss": 0.7276727,
      "learning_rate": 3.403680016360342e-07,
      "loss": 0.74901295,
      "num_input_tokens_seen": 146714835,
      "step": 6796,
      "time_per_iteration": 2.8815994262695312
    },
    {
      "auxiliary_loss_clip": 0.01094254,
      "auxiliary_loss_mlp": 0.01025395,
      "balance_loss_clip": 1.04293942,
      "balance_loss_mlp": 1.01685941,
      "epoch": 0.8172909276739013,
      "flos": 21470128496640.0,
      "grad_norm": 1.6893037016412318,
      "language_loss": 0.67884088,
      "learning_rate": 3.3993343244864403e-07,
      "loss": 0.70003736,
      "num_input_tokens_seen": 146734425,
      "step": 6797,
      "time_per_iteration": 2.801077127456665
    },
    {
      "auxiliary_loss_clip": 0.01129022,
      "auxiliary_loss_mlp": 0.01025517,
      "balance_loss_clip": 1.04662466,
      "balance_loss_mlp": 1.01837039,
      "epoch": 0.8174111705645404,
      "flos": 27599792417280.0,
      "grad_norm": 1.5336663518950888,
      "language_loss": 0.72875798,
      "learning_rate": 3.394991150920323e-07,
      "loss": 0.75030339,
      "num_input_tokens_seen": 146757545,
      "step": 6798,
      "time_per_iteration": 2.8427610397338867
    },
    {
      "auxiliary_loss_clip": 0.01094844,
      "auxiliary_loss_mlp": 0.01032122,
      "balance_loss_clip": 1.04091048,
      "balance_loss_mlp": 1.02293062,
      "epoch": 0.8175314134551794,
      "flos": 14064594508800.0,
      "grad_norm": 2.3102999896040224,
      "language_loss": 0.74682921,
      "learning_rate": 3.3906504963208396e-07,
      "loss": 0.76809883,
      "num_input_tokens_seen": 146774240,
      "step": 6799,
      "time_per_iteration": 2.7854909896850586
    },
    {
      "auxiliary_loss_clip": 0.01143835,
      "auxiliary_loss_mlp": 0.0102844,
      "balance_loss_clip": 1.04596603,
      "balance_loss_mlp": 1.02027369,
      "epoch": 0.8176516563458186,
      "flos": 22708076780160.0,
      "grad_norm": 2.381336717563685,
      "language_loss": 0.66613907,
      "learning_rate": 3.3863123613464774e-07,
      "loss": 0.68786186,
      "num_input_tokens_seen": 146793140,
      "step": 6800,
      "time_per_iteration": 2.6976287364959717
    },
    {
      "auxiliary_loss_clip": 0.01154416,
      "auxiliary_loss_mlp": 0.01024825,
      "balance_loss_clip": 1.04769874,
      "balance_loss_mlp": 1.01689136,
      "epoch": 0.8177718992364577,
      "flos": 21945406279680.0,
      "grad_norm": 1.9965034770684045,
      "language_loss": 0.75227261,
      "learning_rate": 3.381976746655317e-07,
      "loss": 0.77406502,
      "num_input_tokens_seen": 146812895,
      "step": 6801,
      "time_per_iteration": 2.6828339099884033
    },
    {
      "auxiliary_loss_clip": 0.01076522,
      "auxiliary_loss_mlp": 0.01020315,
      "balance_loss_clip": 1.03828764,
      "balance_loss_mlp": 1.01288795,
      "epoch": 0.8178921421270967,
      "flos": 22017443005440.0,
      "grad_norm": 2.182770141095646,
      "language_loss": 0.67624968,
      "learning_rate": 3.3776436529050756e-07,
      "loss": 0.69721806,
      "num_input_tokens_seen": 146832445,
      "step": 6802,
      "time_per_iteration": 2.8295297622680664
    },
    {
      "auxiliary_loss_clip": 0.01138654,
      "auxiliary_loss_mlp": 0.01028912,
      "balance_loss_clip": 1.04245842,
      "balance_loss_mlp": 1.02101994,
      "epoch": 0.8180123850177359,
      "flos": 33183111496320.0,
      "grad_norm": 1.6770070578611929,
      "language_loss": 0.72501731,
      "learning_rate": 3.373313080753073e-07,
      "loss": 0.7466929,
      "num_input_tokens_seen": 146856505,
      "step": 6803,
      "time_per_iteration": 2.7592849731445312
    },
    {
      "auxiliary_loss_clip": 0.01136468,
      "auxiliary_loss_mlp": 0.01027517,
      "balance_loss_clip": 1.04355884,
      "balance_loss_mlp": 1.01939321,
      "epoch": 0.8181326279083749,
      "flos": 22091167670400.0,
      "grad_norm": 1.5385329887462451,
      "language_loss": 0.77592051,
      "learning_rate": 3.3689850308562527e-07,
      "loss": 0.79756039,
      "num_input_tokens_seen": 146876950,
      "step": 6804,
      "time_per_iteration": 2.6969518661499023
    },
    {
      "auxiliary_loss_clip": 0.01136466,
      "auxiliary_loss_mlp": 0.01023127,
      "balance_loss_clip": 1.04608345,
      "balance_loss_mlp": 1.01581037,
      "epoch": 0.818252870799014,
      "flos": 15705747936000.0,
      "grad_norm": 2.0703978001592738,
      "language_loss": 0.77932179,
      "learning_rate": 3.364659503871183e-07,
      "loss": 0.80091774,
      "num_input_tokens_seen": 146894885,
      "step": 6805,
      "time_per_iteration": 2.6657705307006836
    },
    {
      "auxiliary_loss_clip": 0.01119478,
      "auxiliary_loss_mlp": 0.01021838,
      "balance_loss_clip": 1.04126215,
      "balance_loss_mlp": 1.01495636,
      "epoch": 0.8183731136896532,
      "flos": 18770687637120.0,
      "grad_norm": 2.053675328988986,
      "language_loss": 0.83996546,
      "learning_rate": 3.3603365004540417e-07,
      "loss": 0.86137855,
      "num_input_tokens_seen": 146913180,
      "step": 6806,
      "time_per_iteration": 2.735485076904297
    },
    {
      "auxiliary_loss_clip": 0.01088691,
      "auxiliary_loss_mlp": 0.01022741,
      "balance_loss_clip": 1.04227328,
      "balance_loss_mlp": 1.01461136,
      "epoch": 0.8184933565802922,
      "flos": 26541792293760.0,
      "grad_norm": 2.0969085900743094,
      "language_loss": 0.77114421,
      "learning_rate": 3.356016021260624e-07,
      "loss": 0.7922585,
      "num_input_tokens_seen": 146933510,
      "step": 6807,
      "time_per_iteration": 2.8501813411712646
    },
    {
      "auxiliary_loss_clip": 0.01132911,
      "auxiliary_loss_mlp": 0.01025181,
      "balance_loss_clip": 1.04340506,
      "balance_loss_mlp": 1.01719356,
      "epoch": 0.8186135994709313,
      "flos": 17530117660800.0,
      "grad_norm": 2.7381104200680992,
      "language_loss": 0.66356945,
      "learning_rate": 3.35169806694634e-07,
      "loss": 0.68515027,
      "num_input_tokens_seen": 146951760,
      "step": 6808,
      "time_per_iteration": 2.694434642791748
    },
    {
      "auxiliary_loss_clip": 0.01007505,
      "auxiliary_loss_mlp": 0.00753963,
      "balance_loss_clip": 1.01043522,
      "balance_loss_mlp": 1.00101936,
      "epoch": 0.8187338423615703,
      "flos": 63480300675840.0,
      "grad_norm": 0.7156633992506962,
      "language_loss": 0.60612947,
      "learning_rate": 3.3473826381662186e-07,
      "loss": 0.62374413,
      "num_input_tokens_seen": 147022900,
      "step": 6809,
      "time_per_iteration": 4.726808786392212
    },
    {
      "auxiliary_loss_clip": 0.01137098,
      "auxiliary_loss_mlp": 0.01023663,
      "balance_loss_clip": 1.04620135,
      "balance_loss_mlp": 1.01599216,
      "epoch": 0.8188540852522095,
      "flos": 17529974006400.0,
      "grad_norm": 3.555420666408629,
      "language_loss": 0.81814277,
      "learning_rate": 3.3430697355749216e-07,
      "loss": 0.83975035,
      "num_input_tokens_seen": 147040590,
      "step": 6810,
      "time_per_iteration": 4.598565340042114
    },
    {
      "auxiliary_loss_clip": 0.01113848,
      "auxiliary_loss_mlp": 0.01025363,
      "balance_loss_clip": 1.04260254,
      "balance_loss_mlp": 1.01717925,
      "epoch": 0.8189743281428485,
      "flos": 14392530702720.0,
      "grad_norm": 2.1933554439973872,
      "language_loss": 0.75531608,
      "learning_rate": 3.3387593598266907e-07,
      "loss": 0.77670825,
      "num_input_tokens_seen": 147057200,
      "step": 6811,
      "time_per_iteration": 2.904736042022705
    },
    {
      "auxiliary_loss_clip": 0.01133242,
      "auxiliary_loss_mlp": 0.01025375,
      "balance_loss_clip": 1.04566467,
      "balance_loss_mlp": 1.0175488,
      "epoch": 0.8190945710334876,
      "flos": 25080479285760.0,
      "grad_norm": 1.7861007725505802,
      "language_loss": 0.78372139,
      "learning_rate": 3.3344515115754225e-07,
      "loss": 0.80530751,
      "num_input_tokens_seen": 147076180,
      "step": 6812,
      "time_per_iteration": 2.715740442276001
    },
    {
      "auxiliary_loss_clip": 0.01095335,
      "auxiliary_loss_mlp": 0.01027584,
      "balance_loss_clip": 1.04123163,
      "balance_loss_mlp": 1.01904845,
      "epoch": 0.8192148139241268,
      "flos": 21507152440320.0,
      "grad_norm": 3.1269401964061703,
      "language_loss": 0.80288935,
      "learning_rate": 3.33014619147461e-07,
      "loss": 0.82411849,
      "num_input_tokens_seen": 147094205,
      "step": 6813,
      "time_per_iteration": 2.7866246700286865
    },
    {
      "auxiliary_loss_clip": 0.01087026,
      "auxiliary_loss_mlp": 0.0076282,
      "balance_loss_clip": 1.0417105,
      "balance_loss_mlp": 1.00051999,
      "epoch": 0.8193350568147658,
      "flos": 23952166289280.0,
      "grad_norm": 5.255285970032278,
      "language_loss": 0.71770585,
      "learning_rate": 3.325843400177362e-07,
      "loss": 0.73620427,
      "num_input_tokens_seen": 147115545,
      "step": 6814,
      "time_per_iteration": 2.9356305599212646
    },
    {
      "auxiliary_loss_clip": 0.01114144,
      "auxiliary_loss_mlp": 0.01025935,
      "balance_loss_clip": 1.03946388,
      "balance_loss_mlp": 1.01819229,
      "epoch": 0.8194552997054049,
      "flos": 20559469962240.0,
      "grad_norm": 2.162367058196308,
      "language_loss": 0.73583585,
      "learning_rate": 3.32154313833642e-07,
      "loss": 0.7572366,
      "num_input_tokens_seen": 147135700,
      "step": 6815,
      "time_per_iteration": 3.1124684810638428
    },
    {
      "auxiliary_loss_clip": 0.01136677,
      "auxiliary_loss_mlp": 0.01025235,
      "balance_loss_clip": 1.04157484,
      "balance_loss_mlp": 1.01716399,
      "epoch": 0.819575542596044,
      "flos": 26031753123840.0,
      "grad_norm": 2.0003947346938165,
      "language_loss": 0.59400046,
      "learning_rate": 3.3172454066041164e-07,
      "loss": 0.6156196,
      "num_input_tokens_seen": 147155205,
      "step": 6816,
      "time_per_iteration": 2.784541606903076
    },
    {
      "auxiliary_loss_clip": 0.01140349,
      "auxiliary_loss_mlp": 0.01021125,
      "balance_loss_clip": 1.04297543,
      "balance_loss_mlp": 1.01345372,
      "epoch": 0.8196957854866831,
      "flos": 29096944220160.0,
      "grad_norm": 5.20759989955292,
      "language_loss": 0.76063311,
      "learning_rate": 3.3129502056324234e-07,
      "loss": 0.78224778,
      "num_input_tokens_seen": 147176570,
      "step": 6817,
      "time_per_iteration": 2.757352352142334
    },
    {
      "auxiliary_loss_clip": 0.01068137,
      "auxiliary_loss_mlp": 0.00753975,
      "balance_loss_clip": 1.01025391,
      "balance_loss_mlp": 1.00091743,
      "epoch": 0.8198160283773221,
      "flos": 69033631898880.0,
      "grad_norm": 0.7986187062716016,
      "language_loss": 0.59770191,
      "learning_rate": 3.3086575360729165e-07,
      "loss": 0.61592305,
      "num_input_tokens_seen": 147234105,
      "step": 6818,
      "time_per_iteration": 3.004528045654297
    },
    {
      "auxiliary_loss_clip": 0.01104677,
      "auxiliary_loss_mlp": 0.01025407,
      "balance_loss_clip": 1.03828919,
      "balance_loss_mlp": 1.01796246,
      "epoch": 0.8199362712679613,
      "flos": 16618058496000.0,
      "grad_norm": 1.8771020207057945,
      "language_loss": 0.71796399,
      "learning_rate": 3.3043673985767906e-07,
      "loss": 0.73926485,
      "num_input_tokens_seen": 147253170,
      "step": 6819,
      "time_per_iteration": 2.7805819511413574
    },
    {
      "auxiliary_loss_clip": 0.01108304,
      "auxiliary_loss_mlp": 0.0102359,
      "balance_loss_clip": 1.03853738,
      "balance_loss_mlp": 1.01623166,
      "epoch": 0.8200565141586004,
      "flos": 21757664868480.0,
      "grad_norm": 2.3315429949318967,
      "language_loss": 0.77396357,
      "learning_rate": 3.3000797937948564e-07,
      "loss": 0.79528248,
      "num_input_tokens_seen": 147271465,
      "step": 6820,
      "time_per_iteration": 3.7207467555999756
    },
    {
      "auxiliary_loss_clip": 0.01049143,
      "auxiliary_loss_mlp": 0.01000621,
      "balance_loss_clip": 1.01018524,
      "balance_loss_mlp": 0.99938679,
      "epoch": 0.8201767570492394,
      "flos": 69807112392960.0,
      "grad_norm": 0.9444487484677827,
      "language_loss": 0.65010267,
      "learning_rate": 3.295794722377534e-07,
      "loss": 0.6706003,
      "num_input_tokens_seen": 147335070,
      "step": 6821,
      "time_per_iteration": 3.2826642990112305
    },
    {
      "auxiliary_loss_clip": 0.01142903,
      "auxiliary_loss_mlp": 0.01025564,
      "balance_loss_clip": 1.04342651,
      "balance_loss_mlp": 1.01803625,
      "epoch": 0.8202969999398786,
      "flos": 23111892455040.0,
      "grad_norm": 1.7890153974210823,
      "language_loss": 0.79818666,
      "learning_rate": 3.291512184974876e-07,
      "loss": 0.81987137,
      "num_input_tokens_seen": 147355460,
      "step": 6822,
      "time_per_iteration": 2.716813087463379
    },
    {
      "auxiliary_loss_clip": 0.01155265,
      "auxiliary_loss_mlp": 0.01021208,
      "balance_loss_clip": 1.04597306,
      "balance_loss_mlp": 1.01388276,
      "epoch": 0.8204172428305176,
      "flos": 28220616109440.0,
      "grad_norm": 1.6861530620495062,
      "language_loss": 0.66463518,
      "learning_rate": 3.2872321822365346e-07,
      "loss": 0.68639994,
      "num_input_tokens_seen": 147375675,
      "step": 6823,
      "time_per_iteration": 2.7252612113952637
    },
    {
      "auxiliary_loss_clip": 0.01062505,
      "auxiliary_loss_mlp": 0.01022436,
      "balance_loss_clip": 1.03648543,
      "balance_loss_mlp": 1.0146935,
      "epoch": 0.8205374857211567,
      "flos": 20887011106560.0,
      "grad_norm": 1.8176986956440302,
      "language_loss": 0.73591089,
      "learning_rate": 3.282954714811783e-07,
      "loss": 0.75676024,
      "num_input_tokens_seen": 147394580,
      "step": 6824,
      "time_per_iteration": 3.2931313514709473
    },
    {
      "auxiliary_loss_clip": 0.01123978,
      "auxiliary_loss_mlp": 0.01025261,
      "balance_loss_clip": 1.04059637,
      "balance_loss_mlp": 1.01755404,
      "epoch": 0.8206577286117959,
      "flos": 13152140294400.0,
      "grad_norm": 2.5430800073238857,
      "language_loss": 0.71080804,
      "learning_rate": 3.2786797833495093e-07,
      "loss": 0.73230046,
      "num_input_tokens_seen": 147409935,
      "step": 6825,
      "time_per_iteration": 3.6818151473999023
    },
    {
      "auxiliary_loss_clip": 0.01112296,
      "auxiliary_loss_mlp": 0.01028435,
      "balance_loss_clip": 1.04562378,
      "balance_loss_mlp": 1.02042699,
      "epoch": 0.8207779715024349,
      "flos": 25265634917760.0,
      "grad_norm": 1.846633716981437,
      "language_loss": 0.72651279,
      "learning_rate": 3.274407388498213e-07,
      "loss": 0.7479201,
      "num_input_tokens_seen": 147428065,
      "step": 6826,
      "time_per_iteration": 2.810333251953125
    },
    {
      "auxiliary_loss_clip": 0.01124074,
      "auxiliary_loss_mlp": 0.01024143,
      "balance_loss_clip": 1.03885007,
      "balance_loss_mlp": 1.01674032,
      "epoch": 0.820898214393074,
      "flos": 19610243199360.0,
      "grad_norm": 2.146621153624563,
      "language_loss": 0.74430788,
      "learning_rate": 3.270137530906021e-07,
      "loss": 0.7657901,
      "num_input_tokens_seen": 147447300,
      "step": 6827,
      "time_per_iteration": 2.7656090259552
    },
    {
      "auxiliary_loss_clip": 0.01086062,
      "auxiliary_loss_mlp": 0.01025409,
      "balance_loss_clip": 1.03924203,
      "balance_loss_mlp": 1.01792288,
      "epoch": 0.8210184572837131,
      "flos": 15596615439360.0,
      "grad_norm": 2.799352846768968,
      "language_loss": 0.83049512,
      "learning_rate": 3.265870211220665e-07,
      "loss": 0.85160983,
      "num_input_tokens_seen": 147465135,
      "step": 6828,
      "time_per_iteration": 2.927344799041748
    },
    {
      "auxiliary_loss_clip": 0.01130677,
      "auxiliary_loss_mlp": 0.01030405,
      "balance_loss_clip": 1.04496896,
      "balance_loss_mlp": 1.02232838,
      "epoch": 0.8211387001743522,
      "flos": 20813932886400.0,
      "grad_norm": 5.16847160283191,
      "language_loss": 0.81767762,
      "learning_rate": 3.2616054300894934e-07,
      "loss": 0.83928847,
      "num_input_tokens_seen": 147484585,
      "step": 6829,
      "time_per_iteration": 2.767908811569214
    },
    {
      "auxiliary_loss_clip": 0.01082026,
      "auxiliary_loss_mlp": 0.01023808,
      "balance_loss_clip": 1.03155422,
      "balance_loss_mlp": 1.01592207,
      "epoch": 0.8212589430649913,
      "flos": 27704579368320.0,
      "grad_norm": 1.9958193043120362,
      "language_loss": 0.84675598,
      "learning_rate": 3.2573431881594693e-07,
      "loss": 0.86781442,
      "num_input_tokens_seen": 147504130,
      "step": 6830,
      "time_per_iteration": 2.896458625793457
    },
    {
      "auxiliary_loss_clip": 0.01127466,
      "auxiliary_loss_mlp": 0.01022853,
      "balance_loss_clip": 1.04084468,
      "balance_loss_mlp": 1.01455593,
      "epoch": 0.8213791859556304,
      "flos": 22455625017600.0,
      "grad_norm": 2.448334491883716,
      "language_loss": 0.65770829,
      "learning_rate": 3.2530834860771663e-07,
      "loss": 0.6792115,
      "num_input_tokens_seen": 147523510,
      "step": 6831,
      "time_per_iteration": 2.7349612712860107
    },
    {
      "auxiliary_loss_clip": 0.01110274,
      "auxiliary_loss_mlp": 0.0102867,
      "balance_loss_clip": 1.0405035,
      "balance_loss_mlp": 1.02096343,
      "epoch": 0.8214994288462695,
      "flos": 16654471908480.0,
      "grad_norm": 2.1918280245670365,
      "language_loss": 0.74095738,
      "learning_rate": 3.248826324488794e-07,
      "loss": 0.7623468,
      "num_input_tokens_seen": 147540805,
      "step": 6832,
      "time_per_iteration": 2.8546197414398193
    },
    {
      "auxiliary_loss_clip": 0.01142415,
      "auxiliary_loss_mlp": 0.01029079,
      "balance_loss_clip": 1.04734325,
      "balance_loss_mlp": 1.02159572,
      "epoch": 0.8216196717369085,
      "flos": 25221787390080.0,
      "grad_norm": 3.837987113857424,
      "language_loss": 0.88116217,
      "learning_rate": 3.244571704040138e-07,
      "loss": 0.90287709,
      "num_input_tokens_seen": 147560965,
      "step": 6833,
      "time_per_iteration": 2.7691688537597656
    },
    {
      "auxiliary_loss_clip": 0.01125988,
      "auxiliary_loss_mlp": 0.01028387,
      "balance_loss_clip": 1.0419637,
      "balance_loss_mlp": 1.02053761,
      "epoch": 0.8217399146275477,
      "flos": 25371930240000.0,
      "grad_norm": 1.9667007824524758,
      "language_loss": 0.73899907,
      "learning_rate": 3.2403196253766374e-07,
      "loss": 0.76054281,
      "num_input_tokens_seen": 147580045,
      "step": 6834,
      "time_per_iteration": 3.706721544265747
    },
    {
      "auxiliary_loss_clip": 0.01148512,
      "auxiliary_loss_mlp": 0.01025741,
      "balance_loss_clip": 1.0503912,
      "balance_loss_mlp": 1.01741385,
      "epoch": 0.8218601575181868,
      "flos": 25629625388160.0,
      "grad_norm": 2.758728600092583,
      "language_loss": 0.78913283,
      "learning_rate": 3.2360700891433254e-07,
      "loss": 0.81087542,
      "num_input_tokens_seen": 147599070,
      "step": 6835,
      "time_per_iteration": 3.912916898727417
    },
    {
      "auxiliary_loss_clip": 0.01020343,
      "auxiliary_loss_mlp": 0.01002045,
      "balance_loss_clip": 1.0157603,
      "balance_loss_mlp": 1.00067985,
      "epoch": 0.8219804004088258,
      "flos": 67660229427840.0,
      "grad_norm": 0.7933288612831837,
      "language_loss": 0.57272673,
      "learning_rate": 3.231823095984847e-07,
      "loss": 0.59295058,
      "num_input_tokens_seen": 147653710,
      "step": 6836,
      "time_per_iteration": 4.229254245758057
    },
    {
      "auxiliary_loss_clip": 0.01138651,
      "auxiliary_loss_mlp": 0.01019844,
      "balance_loss_clip": 1.04430556,
      "balance_loss_mlp": 1.01223218,
      "epoch": 0.822100643299465,
      "flos": 19464266327040.0,
      "grad_norm": 2.0148976309501596,
      "language_loss": 0.75949264,
      "learning_rate": 3.2275786465454814e-07,
      "loss": 0.78107762,
      "num_input_tokens_seen": 147670360,
      "step": 6837,
      "time_per_iteration": 2.698622941970825
    },
    {
      "auxiliary_loss_clip": 0.01122593,
      "auxiliary_loss_mlp": 0.01023679,
      "balance_loss_clip": 1.04402316,
      "balance_loss_mlp": 1.01615381,
      "epoch": 0.822220886190104,
      "flos": 24681368292480.0,
      "grad_norm": 1.9714629142475235,
      "language_loss": 0.75981432,
      "learning_rate": 3.2233367414690917e-07,
      "loss": 0.78127706,
      "num_input_tokens_seen": 147692550,
      "step": 6838,
      "time_per_iteration": 2.7815396785736084
    },
    {
      "auxiliary_loss_clip": 0.01142302,
      "auxiliary_loss_mlp": 0.01021061,
      "balance_loss_clip": 1.04244161,
      "balance_loss_mlp": 1.01390219,
      "epoch": 0.8223411290807431,
      "flos": 27819062991360.0,
      "grad_norm": 10.32764948440025,
      "language_loss": 0.84730291,
      "learning_rate": 3.219097381399183e-07,
      "loss": 0.86893654,
      "num_input_tokens_seen": 147709725,
      "step": 6839,
      "time_per_iteration": 2.7581872940063477
    },
    {
      "auxiliary_loss_clip": 0.01170623,
      "auxiliary_loss_mlp": 0.00762372,
      "balance_loss_clip": 1.04793024,
      "balance_loss_mlp": 1.00054896,
      "epoch": 0.8224613719713821,
      "flos": 23218546913280.0,
      "grad_norm": 2.312811917971646,
      "language_loss": 0.81229281,
      "learning_rate": 3.2148605669788584e-07,
      "loss": 0.83162284,
      "num_input_tokens_seen": 147729615,
      "step": 6840,
      "time_per_iteration": 2.665546417236328
    },
    {
      "auxiliary_loss_clip": 0.01145027,
      "auxiliary_loss_mlp": 0.01024087,
      "balance_loss_clip": 1.04393387,
      "balance_loss_mlp": 1.01657128,
      "epoch": 0.8225816148620213,
      "flos": 15706250726400.0,
      "grad_norm": 2.9493648609996863,
      "language_loss": 0.78028166,
      "learning_rate": 3.2106262988508405e-07,
      "loss": 0.80197281,
      "num_input_tokens_seen": 147747665,
      "step": 6841,
      "time_per_iteration": 2.71531343460083
    },
    {
      "auxiliary_loss_clip": 0.01146445,
      "auxiliary_loss_mlp": 0.01025327,
      "balance_loss_clip": 1.04337955,
      "balance_loss_mlp": 1.01781726,
      "epoch": 0.8227018577526604,
      "flos": 18515111391360.0,
      "grad_norm": 2.483749995879231,
      "language_loss": 0.74784738,
      "learning_rate": 3.206394577657465e-07,
      "loss": 0.76956505,
      "num_input_tokens_seen": 147765445,
      "step": 6842,
      "time_per_iteration": 2.6015708446502686
    },
    {
      "auxiliary_loss_clip": 0.01146606,
      "auxiliary_loss_mlp": 0.01021824,
      "balance_loss_clip": 1.04486775,
      "balance_loss_mlp": 1.01365852,
      "epoch": 0.8228221006432994,
      "flos": 22236785406720.0,
      "grad_norm": 3.364492527461418,
      "language_loss": 0.72761357,
      "learning_rate": 3.202165404040675e-07,
      "loss": 0.74929786,
      "num_input_tokens_seen": 147783365,
      "step": 6843,
      "time_per_iteration": 2.756870746612549
    },
    {
      "auxiliary_loss_clip": 0.01118336,
      "auxiliary_loss_mlp": 0.00762862,
      "balance_loss_clip": 1.04211426,
      "balance_loss_mlp": 1.00049615,
      "epoch": 0.8229423435339386,
      "flos": 24097532630400.0,
      "grad_norm": 2.306569017714687,
      "language_loss": 0.74631166,
      "learning_rate": 3.1979387786420396e-07,
      "loss": 0.76512372,
      "num_input_tokens_seen": 147803605,
      "step": 6844,
      "time_per_iteration": 2.7641940116882324
    },
    {
      "auxiliary_loss_clip": 0.01137529,
      "auxiliary_loss_mlp": 0.01021509,
      "balance_loss_clip": 1.04303598,
      "balance_loss_mlp": 1.01428509,
      "epoch": 0.8230625864245776,
      "flos": 23878549365120.0,
      "grad_norm": 3.6809254247563494,
      "language_loss": 0.82320726,
      "learning_rate": 3.1937147021027346e-07,
      "loss": 0.84479767,
      "num_input_tokens_seen": 147822060,
      "step": 6845,
      "time_per_iteration": 3.713677167892456
    },
    {
      "auxiliary_loss_clip": 0.0111247,
      "auxiliary_loss_mlp": 0.0102381,
      "balance_loss_clip": 1.04390597,
      "balance_loss_mlp": 1.01669288,
      "epoch": 0.8231828293152167,
      "flos": 16581106379520.0,
      "grad_norm": 2.50105160629128,
      "language_loss": 0.76867008,
      "learning_rate": 3.189493175063547e-07,
      "loss": 0.79003286,
      "num_input_tokens_seen": 147839295,
      "step": 6846,
      "time_per_iteration": 2.8330912590026855
    },
    {
      "auxiliary_loss_clip": 0.01119353,
      "auxiliary_loss_mlp": 0.00763143,
      "balance_loss_clip": 1.04334569,
      "balance_loss_mlp": 1.00063372,
      "epoch": 0.8233030722058559,
      "flos": 18880071528960.0,
      "grad_norm": 1.866295226029775,
      "language_loss": 0.67141891,
      "learning_rate": 3.1852741981648776e-07,
      "loss": 0.69024384,
      "num_input_tokens_seen": 147857945,
      "step": 6847,
      "time_per_iteration": 2.837103843688965
    },
    {
      "auxiliary_loss_clip": 0.01143746,
      "auxiliary_loss_mlp": 0.01023642,
      "balance_loss_clip": 1.04490852,
      "balance_loss_mlp": 1.01621222,
      "epoch": 0.8234233150964949,
      "flos": 28439024757120.0,
      "grad_norm": 2.2516391512181477,
      "language_loss": 0.6944629,
      "learning_rate": 3.1810577720467404e-07,
      "loss": 0.71613681,
      "num_input_tokens_seen": 147879675,
      "step": 6848,
      "time_per_iteration": 2.8945086002349854
    },
    {
      "auxiliary_loss_clip": 0.01137705,
      "auxiliary_loss_mlp": 0.01022299,
      "balance_loss_clip": 1.04399943,
      "balance_loss_mlp": 1.0139842,
      "epoch": 0.823543557987134,
      "flos": 33765941577600.0,
      "grad_norm": 1.7863777646735812,
      "language_loss": 0.56685084,
      "learning_rate": 3.176843897348769e-07,
      "loss": 0.58845091,
      "num_input_tokens_seen": 147902870,
      "step": 6849,
      "time_per_iteration": 2.845952033996582
    },
    {
      "auxiliary_loss_clip": 0.01157095,
      "auxiliary_loss_mlp": 0.00762597,
      "balance_loss_clip": 1.04826868,
      "balance_loss_mlp": 1.00055599,
      "epoch": 0.8236638008777731,
      "flos": 17092366611840.0,
      "grad_norm": 2.7117774998455064,
      "language_loss": 0.75982076,
      "learning_rate": 3.1726325747102034e-07,
      "loss": 0.77901769,
      "num_input_tokens_seen": 147921245,
      "step": 6850,
      "time_per_iteration": 2.7373228073120117
    },
    {
      "auxiliary_loss_clip": 0.01147737,
      "auxiliary_loss_mlp": 0.01021242,
      "balance_loss_clip": 1.04526722,
      "balance_loss_mlp": 1.01372874,
      "epoch": 0.8237840437684122,
      "flos": 61639982334720.0,
      "grad_norm": 1.9791722172273842,
      "language_loss": 0.64164543,
      "learning_rate": 3.1684238047698974e-07,
      "loss": 0.6633352,
      "num_input_tokens_seen": 147949515,
      "step": 6851,
      "time_per_iteration": 3.08003568649292
    },
    {
      "auxiliary_loss_clip": 0.01120909,
      "auxiliary_loss_mlp": 0.01025207,
      "balance_loss_clip": 1.03959787,
      "balance_loss_mlp": 1.01730943,
      "epoch": 0.8239042866590512,
      "flos": 27309023821440.0,
      "grad_norm": 2.4030180913920955,
      "language_loss": 0.52681834,
      "learning_rate": 3.1642175881663155e-07,
      "loss": 0.54827952,
      "num_input_tokens_seen": 147969245,
      "step": 6852,
      "time_per_iteration": 2.7603068351745605
    },
    {
      "auxiliary_loss_clip": 0.01135176,
      "auxiliary_loss_mlp": 0.01021943,
      "balance_loss_clip": 1.04237354,
      "balance_loss_mlp": 1.01436162,
      "epoch": 0.8240245295496904,
      "flos": 21726351187200.0,
      "grad_norm": 13.23524093396863,
      "language_loss": 0.83635312,
      "learning_rate": 3.160013925537537e-07,
      "loss": 0.85792422,
      "num_input_tokens_seen": 147990080,
      "step": 6853,
      "time_per_iteration": 2.7122371196746826
    },
    {
      "auxiliary_loss_clip": 0.01154101,
      "auxiliary_loss_mlp": 0.01029503,
      "balance_loss_clip": 1.04489589,
      "balance_loss_mlp": 1.02172446,
      "epoch": 0.8241447724403295,
      "flos": 20009318279040.0,
      "grad_norm": 2.246900870639972,
      "language_loss": 0.75935054,
      "learning_rate": 3.155812817521266e-07,
      "loss": 0.78118658,
      "num_input_tokens_seen": 148010455,
      "step": 6854,
      "time_per_iteration": 2.67087984085083
    },
    {
      "auxiliary_loss_clip": 0.01110693,
      "auxiliary_loss_mlp": 0.01027507,
      "balance_loss_clip": 1.04287577,
      "balance_loss_mlp": 1.01922846,
      "epoch": 0.8242650153309685,
      "flos": 22272983337600.0,
      "grad_norm": 2.4746852084920636,
      "language_loss": 0.78172421,
      "learning_rate": 3.151614264754787e-07,
      "loss": 0.80310619,
      "num_input_tokens_seen": 148028400,
      "step": 6855,
      "time_per_iteration": 2.7931084632873535
    },
    {
      "auxiliary_loss_clip": 0.01148132,
      "auxiliary_loss_mlp": 0.01020367,
      "balance_loss_clip": 1.04420471,
      "balance_loss_mlp": 1.01280928,
      "epoch": 0.8243852582216077,
      "flos": 22309971367680.0,
      "grad_norm": 2.051202846044359,
      "language_loss": 0.79466677,
      "learning_rate": 3.147418267875035e-07,
      "loss": 0.81635177,
      "num_input_tokens_seen": 148046530,
      "step": 6856,
      "time_per_iteration": 2.6884942054748535
    },
    {
      "auxiliary_loss_clip": 0.01114811,
      "auxiliary_loss_mlp": 0.01026981,
      "balance_loss_clip": 1.03863764,
      "balance_loss_mlp": 1.01901793,
      "epoch": 0.8245055011122467,
      "flos": 24645421756800.0,
      "grad_norm": 2.1935224035247822,
      "language_loss": 0.6526109,
      "learning_rate": 3.1432248275185315e-07,
      "loss": 0.67402887,
      "num_input_tokens_seen": 148067040,
      "step": 6857,
      "time_per_iteration": 2.8409066200256348
    },
    {
      "auxiliary_loss_clip": 0.01151308,
      "auxiliary_loss_mlp": 0.01027014,
      "balance_loss_clip": 1.04342389,
      "balance_loss_mlp": 1.0195992,
      "epoch": 0.8246257440028858,
      "flos": 17487275713920.0,
      "grad_norm": 6.338325225980607,
      "language_loss": 0.77438939,
      "learning_rate": 3.139033944321412e-07,
      "loss": 0.79617256,
      "num_input_tokens_seen": 148084400,
      "step": 6858,
      "time_per_iteration": 2.590322732925415
    },
    {
      "auxiliary_loss_clip": 0.01114996,
      "auxiliary_loss_mlp": 0.01026692,
      "balance_loss_clip": 1.04305363,
      "balance_loss_mlp": 1.01841331,
      "epoch": 0.824745986893525,
      "flos": 25010130499200.0,
      "grad_norm": 1.7362810654269045,
      "language_loss": 0.79167867,
      "learning_rate": 3.1348456189194507e-07,
      "loss": 0.81309551,
      "num_input_tokens_seen": 148104860,
      "step": 6859,
      "time_per_iteration": 2.8114867210388184
    },
    {
      "auxiliary_loss_clip": 0.01115228,
      "auxiliary_loss_mlp": 0.01028249,
      "balance_loss_clip": 1.04565573,
      "balance_loss_mlp": 1.02030981,
      "epoch": 0.824866229784164,
      "flos": 18772698798720.0,
      "grad_norm": 1.949065042970968,
      "language_loss": 0.83006734,
      "learning_rate": 3.1306598519479876e-07,
      "loss": 0.85150212,
      "num_input_tokens_seen": 148124680,
      "step": 6860,
      "time_per_iteration": 3.618765354156494
    },
    {
      "auxiliary_loss_clip": 0.01143564,
      "auxiliary_loss_mlp": 0.00761818,
      "balance_loss_clip": 1.04586172,
      "balance_loss_mlp": 1.0005486,
      "epoch": 0.8249864726748031,
      "flos": 23842171866240.0,
      "grad_norm": 2.3557304091977285,
      "language_loss": 0.78299421,
      "learning_rate": 3.1264766440420177e-07,
      "loss": 0.80204809,
      "num_input_tokens_seen": 148147150,
      "step": 6861,
      "time_per_iteration": 3.6293888092041016
    },
    {
      "auxiliary_loss_clip": 0.01095254,
      "auxiliary_loss_mlp": 0.01024695,
      "balance_loss_clip": 1.042413,
      "balance_loss_mlp": 1.01722896,
      "epoch": 0.8251067155654422,
      "flos": 20303103617280.0,
      "grad_norm": 2.353841088523314,
      "language_loss": 0.69388938,
      "learning_rate": 3.122295995836124e-07,
      "loss": 0.71508884,
      "num_input_tokens_seen": 148167020,
      "step": 6862,
      "time_per_iteration": 3.66129207611084
    },
    {
      "auxiliary_loss_clip": 0.01139537,
      "auxiliary_loss_mlp": 0.01023196,
      "balance_loss_clip": 1.04476726,
      "balance_loss_mlp": 1.01520896,
      "epoch": 0.8252269584560813,
      "flos": 25009699536000.0,
      "grad_norm": 2.007656438066275,
      "language_loss": 0.77349752,
      "learning_rate": 3.118117907964508e-07,
      "loss": 0.79512477,
      "num_input_tokens_seen": 148188965,
      "step": 6863,
      "time_per_iteration": 2.7437336444854736
    },
    {
      "auxiliary_loss_clip": 0.01144635,
      "auxiliary_loss_mlp": 0.0102448,
      "balance_loss_clip": 1.04667997,
      "balance_loss_mlp": 1.01690102,
      "epoch": 0.8253472013467203,
      "flos": 17128564542720.0,
      "grad_norm": 1.8921720553095045,
      "language_loss": 0.80573833,
      "learning_rate": 3.1139423810609856e-07,
      "loss": 0.82742947,
      "num_input_tokens_seen": 148205660,
      "step": 6864,
      "time_per_iteration": 2.9023947715759277
    },
    {
      "auxiliary_loss_clip": 0.01120719,
      "auxiliary_loss_mlp": 0.01019175,
      "balance_loss_clip": 1.04146671,
      "balance_loss_mlp": 1.01109815,
      "epoch": 0.8254674442373595,
      "flos": 22414794232320.0,
      "grad_norm": 1.7527828440165223,
      "language_loss": 0.7555539,
      "learning_rate": 3.1097694157589714e-07,
      "loss": 0.77695286,
      "num_input_tokens_seen": 148225545,
      "step": 6865,
      "time_per_iteration": 2.785414934158325
    },
    {
      "auxiliary_loss_clip": 0.01093563,
      "auxiliary_loss_mlp": 0.01025101,
      "balance_loss_clip": 1.04072642,
      "balance_loss_mlp": 1.01737607,
      "epoch": 0.8255876871279986,
      "flos": 24786765774720.0,
      "grad_norm": 3.4021169382288203,
      "language_loss": 0.76061487,
      "learning_rate": 3.105599012691511e-07,
      "loss": 0.78180152,
      "num_input_tokens_seen": 148243975,
      "step": 6866,
      "time_per_iteration": 2.7983345985412598
    },
    {
      "auxiliary_loss_clip": 0.01093381,
      "auxiliary_loss_mlp": 0.01022429,
      "balance_loss_clip": 1.03658664,
      "balance_loss_mlp": 1.01514482,
      "epoch": 0.8257079300186376,
      "flos": 27455431656960.0,
      "grad_norm": 1.8856751622221362,
      "language_loss": 0.82572496,
      "learning_rate": 3.101431172491249e-07,
      "loss": 0.84688306,
      "num_input_tokens_seen": 148265520,
      "step": 6867,
      "time_per_iteration": 2.8553268909454346
    },
    {
      "auxiliary_loss_clip": 0.01113594,
      "auxiliary_loss_mlp": 0.01032064,
      "balance_loss_clip": 1.04155684,
      "balance_loss_mlp": 1.02374959,
      "epoch": 0.8258281729092768,
      "flos": 16471866142080.0,
      "grad_norm": 2.206692738016707,
      "language_loss": 0.72208941,
      "learning_rate": 3.097265895790444e-07,
      "loss": 0.74354601,
      "num_input_tokens_seen": 148283730,
      "step": 6868,
      "time_per_iteration": 2.7907702922821045
    },
    {
      "auxiliary_loss_clip": 0.011359,
      "auxiliary_loss_mlp": 0.0102359,
      "balance_loss_clip": 1.04509497,
      "balance_loss_mlp": 1.01551378,
      "epoch": 0.8259484157999158,
      "flos": 21433822824960.0,
      "grad_norm": 2.0632252313574537,
      "language_loss": 0.83648038,
      "learning_rate": 3.093103183220962e-07,
      "loss": 0.85807526,
      "num_input_tokens_seen": 148303775,
      "step": 6869,
      "time_per_iteration": 2.7550623416900635
    },
    {
      "auxiliary_loss_clip": 0.01042741,
      "auxiliary_loss_mlp": 0.01000494,
      "balance_loss_clip": 1.00905871,
      "balance_loss_mlp": 0.99934405,
      "epoch": 0.8260686586905549,
      "flos": 58322342453760.0,
      "grad_norm": 0.8174495028913511,
      "language_loss": 0.59274387,
      "learning_rate": 3.0889430354142796e-07,
      "loss": 0.61317623,
      "num_input_tokens_seen": 148365285,
      "step": 6870,
      "time_per_iteration": 3.199528455734253
    },
    {
      "auxiliary_loss_clip": 0.01166352,
      "auxiliary_loss_mlp": 0.01025271,
      "balance_loss_clip": 1.04581881,
      "balance_loss_mlp": 1.01745129,
      "epoch": 0.826188901581194,
      "flos": 27527288814720.0,
      "grad_norm": 2.035252184578412,
      "language_loss": 0.69996691,
      "learning_rate": 3.084785453001497e-07,
      "loss": 0.72188312,
      "num_input_tokens_seen": 148386200,
      "step": 6871,
      "time_per_iteration": 3.6430447101593018
    },
    {
      "auxiliary_loss_clip": 0.01120854,
      "auxiliary_loss_mlp": 0.01032854,
      "balance_loss_clip": 1.04097581,
      "balance_loss_mlp": 1.02458096,
      "epoch": 0.8263091444718331,
      "flos": 23696051339520.0,
      "grad_norm": 2.8702459566412486,
      "language_loss": 0.82302284,
      "learning_rate": 3.080630436613314e-07,
      "loss": 0.84455997,
      "num_input_tokens_seen": 148403970,
      "step": 6872,
      "time_per_iteration": 2.877598762512207
    },
    {
      "auxiliary_loss_clip": 0.01086478,
      "auxiliary_loss_mlp": 0.01028107,
      "balance_loss_clip": 1.03944051,
      "balance_loss_mlp": 1.02036166,
      "epoch": 0.8264293873624722,
      "flos": 17165157523200.0,
      "grad_norm": 2.2101157487900647,
      "language_loss": 0.8614676,
      "learning_rate": 3.076477986880039e-07,
      "loss": 0.88261342,
      "num_input_tokens_seen": 148421765,
      "step": 6873,
      "time_per_iteration": 2.980290651321411
    },
    {
      "auxiliary_loss_clip": 0.01135425,
      "auxiliary_loss_mlp": 0.01019134,
      "balance_loss_clip": 1.04352105,
      "balance_loss_mlp": 1.01120615,
      "epoch": 0.8265496302531112,
      "flos": 24098645952000.0,
      "grad_norm": 2.8940770236965023,
      "language_loss": 0.6913358,
      "learning_rate": 3.0723281044315986e-07,
      "loss": 0.71288145,
      "num_input_tokens_seen": 148443720,
      "step": 6874,
      "time_per_iteration": 3.265644073486328
    },
    {
      "auxiliary_loss_clip": 0.01112766,
      "auxiliary_loss_mlp": 0.00762115,
      "balance_loss_clip": 1.03984761,
      "balance_loss_mlp": 1.00047052,
      "epoch": 0.8266698731437504,
      "flos": 14099894599680.0,
      "grad_norm": 2.5533803753077833,
      "language_loss": 0.76836735,
      "learning_rate": 3.068180789897521e-07,
      "loss": 0.78711617,
      "num_input_tokens_seen": 148462130,
      "step": 6875,
      "time_per_iteration": 2.7109858989715576
    },
    {
      "auxiliary_loss_clip": 0.01151153,
      "auxiliary_loss_mlp": 0.00762513,
      "balance_loss_clip": 1.04365218,
      "balance_loss_mlp": 1.00048053,
      "epoch": 0.8267901160343895,
      "flos": 30777563715840.0,
      "grad_norm": 2.0518721163665687,
      "language_loss": 0.81891048,
      "learning_rate": 3.064036043906966e-07,
      "loss": 0.83804715,
      "num_input_tokens_seen": 148485570,
      "step": 6876,
      "time_per_iteration": 2.7229716777801514
    },
    {
      "auxiliary_loss_clip": 0.01111745,
      "auxiliary_loss_mlp": 0.01031023,
      "balance_loss_clip": 1.04164457,
      "balance_loss_mlp": 1.02245235,
      "epoch": 0.8269103589250285,
      "flos": 40624915242240.0,
      "grad_norm": 2.599686178203987,
      "language_loss": 0.68011463,
      "learning_rate": 3.059893867088668e-07,
      "loss": 0.70154226,
      "num_input_tokens_seen": 148509715,
      "step": 6877,
      "time_per_iteration": 3.0049326419830322
    },
    {
      "auxiliary_loss_clip": 0.01130825,
      "auxiliary_loss_mlp": 0.01024582,
      "balance_loss_clip": 1.04203022,
      "balance_loss_mlp": 1.01734924,
      "epoch": 0.8270306018156677,
      "flos": 30263645877120.0,
      "grad_norm": 2.166121526329117,
      "language_loss": 0.66896141,
      "learning_rate": 3.055754260071004e-07,
      "loss": 0.69051552,
      "num_input_tokens_seen": 148532010,
      "step": 6878,
      "time_per_iteration": 2.8137238025665283
    },
    {
      "auxiliary_loss_clip": 0.01129034,
      "auxiliary_loss_mlp": 0.01024689,
      "balance_loss_clip": 1.04257226,
      "balance_loss_mlp": 1.01742566,
      "epoch": 0.8271508447063067,
      "flos": 25226599812480.0,
      "grad_norm": 2.8020067354972813,
      "language_loss": 0.73976624,
      "learning_rate": 3.051617223481948e-07,
      "loss": 0.76130348,
      "num_input_tokens_seen": 148553330,
      "step": 6879,
      "time_per_iteration": 2.7411375045776367
    },
    {
      "auxiliary_loss_clip": 0.01144759,
      "auxiliary_loss_mlp": 0.0102954,
      "balance_loss_clip": 1.04723907,
      "balance_loss_mlp": 1.02146995,
      "epoch": 0.8272710875969458,
      "flos": 17566602900480.0,
      "grad_norm": 3.531834064955368,
      "language_loss": 0.75255787,
      "learning_rate": 3.047482757949078e-07,
      "loss": 0.77430081,
      "num_input_tokens_seen": 148570960,
      "step": 6880,
      "time_per_iteration": 2.912012815475464
    },
    {
      "auxiliary_loss_clip": 0.01112575,
      "auxiliary_loss_mlp": 0.0102434,
      "balance_loss_clip": 1.03791201,
      "balance_loss_mlp": 1.01750636,
      "epoch": 0.827391330487585,
      "flos": 19755465886080.0,
      "grad_norm": 2.1184374337822205,
      "language_loss": 0.85765213,
      "learning_rate": 3.043350864099605e-07,
      "loss": 0.87902129,
      "num_input_tokens_seen": 148589520,
      "step": 6881,
      "time_per_iteration": 2.755288600921631
    },
    {
      "auxiliary_loss_clip": 0.01115296,
      "auxiliary_loss_mlp": 0.0076242,
      "balance_loss_clip": 1.03838301,
      "balance_loss_mlp": 1.00043178,
      "epoch": 0.827511573378224,
      "flos": 16835174254080.0,
      "grad_norm": 2.5120246761512206,
      "language_loss": 0.80861092,
      "learning_rate": 3.039221542560315e-07,
      "loss": 0.82738811,
      "num_input_tokens_seen": 148606085,
      "step": 6882,
      "time_per_iteration": 2.640958547592163
    },
    {
      "auxiliary_loss_clip": 0.01168698,
      "auxiliary_loss_mlp": 0.01024255,
      "balance_loss_clip": 1.04684579,
      "balance_loss_mlp": 1.01642907,
      "epoch": 0.8276318162688631,
      "flos": 18369242259840.0,
      "grad_norm": 2.029075386029689,
      "language_loss": 0.73680389,
      "learning_rate": 3.0350947939576356e-07,
      "loss": 0.75873345,
      "num_input_tokens_seen": 148625240,
      "step": 6883,
      "time_per_iteration": 2.564863681793213
    },
    {
      "auxiliary_loss_clip": 0.01119274,
      "auxiliary_loss_mlp": 0.01027551,
      "balance_loss_clip": 1.04503226,
      "balance_loss_mlp": 1.01945126,
      "epoch": 0.8277520591595022,
      "flos": 19352691705600.0,
      "grad_norm": 1.894466644956274,
      "language_loss": 0.724347,
      "learning_rate": 3.0309706189175876e-07,
      "loss": 0.74581522,
      "num_input_tokens_seen": 148645075,
      "step": 6884,
      "time_per_iteration": 2.7623085975646973
    },
    {
      "auxiliary_loss_clip": 0.01036009,
      "auxiliary_loss_mlp": 0.009999,
      "balance_loss_clip": 1.01127434,
      "balance_loss_mlp": 0.99888688,
      "epoch": 0.8278723020501413,
      "flos": 67918858329600.0,
      "grad_norm": 0.7577743455144089,
      "language_loss": 0.57301438,
      "learning_rate": 3.0268490180658045e-07,
      "loss": 0.59337342,
      "num_input_tokens_seen": 148707855,
      "step": 6885,
      "time_per_iteration": 4.1539306640625
    },
    {
      "auxiliary_loss_clip": 0.01130802,
      "auxiliary_loss_mlp": 0.01026792,
      "balance_loss_clip": 1.04516363,
      "balance_loss_mlp": 1.01916873,
      "epoch": 0.8279925449407803,
      "flos": 18185738653440.0,
      "grad_norm": 2.128819523331006,
      "language_loss": 0.79545468,
      "learning_rate": 3.0227299920275305e-07,
      "loss": 0.81703061,
      "num_input_tokens_seen": 148724170,
      "step": 6886,
      "time_per_iteration": 2.8651978969573975
    },
    {
      "auxiliary_loss_clip": 0.01143594,
      "auxiliary_loss_mlp": 0.00762675,
      "balance_loss_clip": 1.04661357,
      "balance_loss_mlp": 1.00047886,
      "epoch": 0.8281127878314195,
      "flos": 20631434860800.0,
      "grad_norm": 3.315170657761401,
      "language_loss": 0.86192322,
      "learning_rate": 3.018613541427613e-07,
      "loss": 0.88098586,
      "num_input_tokens_seen": 148743690,
      "step": 6887,
      "time_per_iteration": 4.523649215698242
    },
    {
      "auxiliary_loss_clip": 0.0114016,
      "auxiliary_loss_mlp": 0.01022763,
      "balance_loss_clip": 1.04320788,
      "balance_loss_mlp": 1.01514256,
      "epoch": 0.8282330307220586,
      "flos": 18004282122240.0,
      "grad_norm": 1.8945078976372525,
      "language_loss": 0.73738158,
      "learning_rate": 3.0144996668905243e-07,
      "loss": 0.75901079,
      "num_input_tokens_seen": 148761070,
      "step": 6888,
      "time_per_iteration": 2.6920464038848877
    },
    {
      "auxiliary_loss_clip": 0.01088787,
      "auxiliary_loss_mlp": 0.0102592,
      "balance_loss_clip": 1.04107666,
      "balance_loss_mlp": 1.01867843,
      "epoch": 0.8283532736126976,
      "flos": 20084120352000.0,
      "grad_norm": 3.1899732186593526,
      "language_loss": 0.82260406,
      "learning_rate": 3.010388369040331e-07,
      "loss": 0.84375107,
      "num_input_tokens_seen": 148779730,
      "step": 6889,
      "time_per_iteration": 2.971219778060913
    },
    {
      "auxiliary_loss_clip": 0.01140987,
      "auxiliary_loss_mlp": 0.01024261,
      "balance_loss_clip": 1.04340339,
      "balance_loss_mlp": 1.01697171,
      "epoch": 0.8284735165033368,
      "flos": 31868421805440.0,
      "grad_norm": 4.018305768989082,
      "language_loss": 0.82939672,
      "learning_rate": 3.0062796485007156e-07,
      "loss": 0.85104918,
      "num_input_tokens_seen": 148800670,
      "step": 6890,
      "time_per_iteration": 3.109342575073242
    },
    {
      "auxiliary_loss_clip": 0.01139085,
      "auxiliary_loss_mlp": 0.01025679,
      "balance_loss_clip": 1.0444262,
      "balance_loss_mlp": 1.01759684,
      "epoch": 0.8285937593939758,
      "flos": 26651319840000.0,
      "grad_norm": 2.7493591869373253,
      "language_loss": 0.65957749,
      "learning_rate": 3.002173505894965e-07,
      "loss": 0.68122512,
      "num_input_tokens_seen": 148819820,
      "step": 6891,
      "time_per_iteration": 2.714585781097412
    },
    {
      "auxiliary_loss_clip": 0.01130868,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.04798782,
      "balance_loss_mlp": 1.0166769,
      "epoch": 0.8287140022846149,
      "flos": 20193683811840.0,
      "grad_norm": 3.6980466418878306,
      "language_loss": 0.63119745,
      "learning_rate": 2.998069941845973e-07,
      "loss": 0.65276086,
      "num_input_tokens_seen": 148838890,
      "step": 6892,
      "time_per_iteration": 2.7516610622406006
    },
    {
      "auxiliary_loss_clip": 0.01057887,
      "auxiliary_loss_mlp": 0.01002182,
      "balance_loss_clip": 1.00939703,
      "balance_loss_mlp": 1.00118661,
      "epoch": 0.8288342451752541,
      "flos": 70755980019840.0,
      "grad_norm": 0.7084594398314865,
      "language_loss": 0.57430589,
      "learning_rate": 2.993968956976258e-07,
      "loss": 0.59490657,
      "num_input_tokens_seen": 148906635,
      "step": 6893,
      "time_per_iteration": 3.4014689922332764
    },
    {
      "auxiliary_loss_clip": 0.01176222,
      "auxiliary_loss_mlp": 0.0102984,
      "balance_loss_clip": 1.04966235,
      "balance_loss_mlp": 1.02116776,
      "epoch": 0.8289544880658931,
      "flos": 24572235795840.0,
      "grad_norm": 1.9931988797407494,
      "language_loss": 0.70390528,
      "learning_rate": 2.9898705519079313e-07,
      "loss": 0.72596592,
      "num_input_tokens_seen": 148925740,
      "step": 6894,
      "time_per_iteration": 2.6559550762176514
    },
    {
      "auxiliary_loss_clip": 0.01138682,
      "auxiliary_loss_mlp": 0.0102242,
      "balance_loss_clip": 1.04223049,
      "balance_loss_mlp": 1.01483774,
      "epoch": 0.8290747309565322,
      "flos": 22273378387200.0,
      "grad_norm": 1.9493722006820777,
      "language_loss": 0.75019741,
      "learning_rate": 2.985774727262715e-07,
      "loss": 0.77180839,
      "num_input_tokens_seen": 148944585,
      "step": 6895,
      "time_per_iteration": 2.6889472007751465
    },
    {
      "auxiliary_loss_clip": 0.01141431,
      "auxiliary_loss_mlp": 0.01024022,
      "balance_loss_clip": 1.04294825,
      "balance_loss_mlp": 1.01661587,
      "epoch": 0.8291949738471713,
      "flos": 23255570856960.0,
      "grad_norm": 2.9380246291293286,
      "language_loss": 0.81528193,
      "learning_rate": 2.981681483661949e-07,
      "loss": 0.83693647,
      "num_input_tokens_seen": 148964170,
      "step": 6896,
      "time_per_iteration": 2.7088119983673096
    },
    {
      "auxiliary_loss_clip": 0.01119827,
      "auxiliary_loss_mlp": 0.01031004,
      "balance_loss_clip": 1.04463661,
      "balance_loss_mlp": 1.02311814,
      "epoch": 0.8293152167378104,
      "flos": 52555768185600.0,
      "grad_norm": 1.9239904665204013,
      "language_loss": 0.711797,
      "learning_rate": 2.9775908217265633e-07,
      "loss": 0.73330528,
      "num_input_tokens_seen": 148989405,
      "step": 6897,
      "time_per_iteration": 4.015460014343262
    },
    {
      "auxiliary_loss_clip": 0.01024379,
      "auxiliary_loss_mlp": 0.0100205,
      "balance_loss_clip": 1.01119602,
      "balance_loss_mlp": 1.00112045,
      "epoch": 0.8294354596284494,
      "flos": 63356156294400.0,
      "grad_norm": 0.8323382531032344,
      "language_loss": 0.50335014,
      "learning_rate": 2.9735027420771253e-07,
      "loss": 0.52361441,
      "num_input_tokens_seen": 149049740,
      "step": 6898,
      "time_per_iteration": 3.2423760890960693
    },
    {
      "auxiliary_loss_clip": 0.01131082,
      "auxiliary_loss_mlp": 0.00760926,
      "balance_loss_clip": 1.04288292,
      "balance_loss_mlp": 1.00044501,
      "epoch": 0.8295557025190886,
      "flos": 24827021942400.0,
      "grad_norm": 1.98122735257087,
      "language_loss": 0.7156316,
      "learning_rate": 2.969417245333774e-07,
      "loss": 0.73455167,
      "num_input_tokens_seen": 149069120,
      "step": 6899,
      "time_per_iteration": 2.7065563201904297
    },
    {
      "auxiliary_loss_clip": 0.01138499,
      "auxiliary_loss_mlp": 0.0076246,
      "balance_loss_clip": 1.04545152,
      "balance_loss_mlp": 1.00051486,
      "epoch": 0.8296759454097277,
      "flos": 25118580637440.0,
      "grad_norm": 2.4419428818345303,
      "language_loss": 0.78337371,
      "learning_rate": 2.9653343321162915e-07,
      "loss": 0.80238324,
      "num_input_tokens_seen": 149088630,
      "step": 6900,
      "time_per_iteration": 2.7311339378356934
    },
    {
      "auxiliary_loss_clip": 0.0110927,
      "auxiliary_loss_mlp": 0.00762418,
      "balance_loss_clip": 1.04343057,
      "balance_loss_mlp": 1.00043726,
      "epoch": 0.8297961883003667,
      "flos": 24132581326080.0,
      "grad_norm": 2.3566173368043484,
      "language_loss": 0.64888752,
      "learning_rate": 2.9612540030440446e-07,
      "loss": 0.66760433,
      "num_input_tokens_seen": 149109175,
      "step": 6901,
      "time_per_iteration": 2.7611825466156006
    },
    {
      "auxiliary_loss_clip": 0.01027507,
      "auxiliary_loss_mlp": 0.01001835,
      "balance_loss_clip": 1.01203489,
      "balance_loss_mlp": 1.0007447,
      "epoch": 0.8299164311910058,
      "flos": 67446561375360.0,
      "grad_norm": 0.846146365436416,
      "language_loss": 0.64060783,
      "learning_rate": 2.9571762587360206e-07,
      "loss": 0.66090131,
      "num_input_tokens_seen": 149165560,
      "step": 6902,
      "time_per_iteration": 3.2390122413635254
    },
    {
      "auxiliary_loss_clip": 0.01113099,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.03796613,
      "balance_loss_mlp": 1.0169282,
      "epoch": 0.8300366740816449,
      "flos": 25228682801280.0,
      "grad_norm": 1.8009364333523905,
      "language_loss": 0.74042368,
      "learning_rate": 2.953101099810806e-07,
      "loss": 0.76180029,
      "num_input_tokens_seen": 149185165,
      "step": 6903,
      "time_per_iteration": 2.8337278366088867
    },
    {
      "auxiliary_loss_clip": 0.01122046,
      "auxiliary_loss_mlp": 0.01026822,
      "balance_loss_clip": 1.04447663,
      "balance_loss_mlp": 1.01901412,
      "epoch": 0.830156916972284,
      "flos": 18041018757120.0,
      "grad_norm": 2.3982395920219317,
      "language_loss": 0.82522702,
      "learning_rate": 2.9490285268865965e-07,
      "loss": 0.84671569,
      "num_input_tokens_seen": 149202655,
      "step": 6904,
      "time_per_iteration": 2.744112730026245
    },
    {
      "auxiliary_loss_clip": 0.01125264,
      "auxiliary_loss_mlp": 0.01026918,
      "balance_loss_clip": 1.04215515,
      "balance_loss_mlp": 1.01853168,
      "epoch": 0.830277159862923,
      "flos": 26322485806080.0,
      "grad_norm": 2.6147152774934197,
      "language_loss": 0.79805756,
      "learning_rate": 2.9449585405812085e-07,
      "loss": 0.81957936,
      "num_input_tokens_seen": 149220035,
      "step": 6905,
      "time_per_iteration": 2.784485340118408
    },
    {
      "auxiliary_loss_clip": 0.01157403,
      "auxiliary_loss_mlp": 0.01024441,
      "balance_loss_clip": 1.04940367,
      "balance_loss_mlp": 1.01657867,
      "epoch": 0.8303974027535622,
      "flos": 19938861751680.0,
      "grad_norm": 3.6765103314716012,
      "language_loss": 0.73753911,
      "learning_rate": 2.940891141512043e-07,
      "loss": 0.75935757,
      "num_input_tokens_seen": 149238055,
      "step": 6906,
      "time_per_iteration": 2.7835967540740967
    },
    {
      "auxiliary_loss_clip": 0.01139728,
      "auxiliary_loss_mlp": 0.00762194,
      "balance_loss_clip": 1.0467639,
      "balance_loss_mlp": 1.00053537,
      "epoch": 0.8305176456442013,
      "flos": 17165552572800.0,
      "grad_norm": 2.2150741323089327,
      "language_loss": 0.72421992,
      "learning_rate": 2.9368263302961385e-07,
      "loss": 0.74323916,
      "num_input_tokens_seen": 149256755,
      "step": 6907,
      "time_per_iteration": 2.704219341278076
    },
    {
      "auxiliary_loss_clip": 0.01160336,
      "auxiliary_loss_mlp": 0.00762464,
      "balance_loss_clip": 1.04552996,
      "balance_loss_mlp": 1.00048852,
      "epoch": 0.8306378885348403,
      "flos": 25627614226560.0,
      "grad_norm": 1.7984667224257118,
      "language_loss": 0.79929149,
      "learning_rate": 2.9327641075501075e-07,
      "loss": 0.81851947,
      "num_input_tokens_seen": 149275745,
      "step": 6908,
      "time_per_iteration": 3.067758083343506
    },
    {
      "auxiliary_loss_clip": 0.01145487,
      "auxiliary_loss_mlp": 0.01026157,
      "balance_loss_clip": 1.0453496,
      "balance_loss_mlp": 1.01843774,
      "epoch": 0.8307581314254795,
      "flos": 33947864985600.0,
      "grad_norm": 2.352770288645338,
      "language_loss": 0.66744971,
      "learning_rate": 2.9287044738901866e-07,
      "loss": 0.68916613,
      "num_input_tokens_seen": 149293730,
      "step": 6909,
      "time_per_iteration": 2.8081674575805664
    },
    {
      "auxiliary_loss_clip": 0.01107269,
      "auxiliary_loss_mlp": 0.01027898,
      "balance_loss_clip": 1.04137444,
      "balance_loss_mlp": 1.02004766,
      "epoch": 0.8308783743161186,
      "flos": 17562724231680.0,
      "grad_norm": 2.020912680910372,
      "language_loss": 0.90879667,
      "learning_rate": 2.9246474299322274e-07,
      "loss": 0.93014836,
      "num_input_tokens_seen": 149309290,
      "step": 6910,
      "time_per_iteration": 2.6500184535980225
    },
    {
      "auxiliary_loss_clip": 0.01043093,
      "auxiliary_loss_mlp": 0.00999954,
      "balance_loss_clip": 1.00815868,
      "balance_loss_mlp": 0.99888659,
      "epoch": 0.8309986172067576,
      "flos": 69412885649280.0,
      "grad_norm": 0.8895380286315039,
      "language_loss": 0.63094103,
      "learning_rate": 2.920592976291678e-07,
      "loss": 0.65137148,
      "num_input_tokens_seen": 149366620,
      "step": 6911,
      "time_per_iteration": 4.271988153457642
    },
    {
      "auxiliary_loss_clip": 0.01142011,
      "auxiliary_loss_mlp": 0.01022219,
      "balance_loss_clip": 1.04583871,
      "balance_loss_mlp": 1.01453638,
      "epoch": 0.8311188600973968,
      "flos": 22309755886080.0,
      "grad_norm": 2.3615466491510784,
      "language_loss": 0.80709636,
      "learning_rate": 2.916541113583595e-07,
      "loss": 0.82873869,
      "num_input_tokens_seen": 149385120,
      "step": 6912,
      "time_per_iteration": 3.7280008792877197
    },
    {
      "auxiliary_loss_clip": 0.01139472,
      "auxiliary_loss_mlp": 0.00762434,
      "balance_loss_clip": 1.049389,
      "balance_loss_mlp": 1.00046659,
      "epoch": 0.8312391029880358,
      "flos": 18770077105920.0,
      "grad_norm": 2.6779904468835967,
      "language_loss": 0.66846669,
      "learning_rate": 2.912491842422642e-07,
      "loss": 0.68748581,
      "num_input_tokens_seen": 149402825,
      "step": 6913,
      "time_per_iteration": 3.6317286491394043
    },
    {
      "auxiliary_loss_clip": 0.01100649,
      "auxiliary_loss_mlp": 0.01029184,
      "balance_loss_clip": 1.04509187,
      "balance_loss_mlp": 1.0214715,
      "epoch": 0.8313593458786749,
      "flos": 20376648714240.0,
      "grad_norm": 1.6399385989443875,
      "language_loss": 0.71025908,
      "learning_rate": 2.9084451634230857e-07,
      "loss": 0.73155737,
      "num_input_tokens_seen": 149422125,
      "step": 6914,
      "time_per_iteration": 3.028140068054199
    },
    {
      "auxiliary_loss_clip": 0.01129638,
      "auxiliary_loss_mlp": 0.01024176,
      "balance_loss_clip": 1.04271984,
      "balance_loss_mlp": 1.01612401,
      "epoch": 0.831479588769314,
      "flos": 32124069878400.0,
      "grad_norm": 3.1212352422864,
      "language_loss": 0.7148186,
      "learning_rate": 2.9044010771988125e-07,
      "loss": 0.73635674,
      "num_input_tokens_seen": 149441940,
      "step": 6915,
      "time_per_iteration": 3.0328307151794434
    },
    {
      "auxiliary_loss_clip": 0.01144713,
      "auxiliary_loss_mlp": 0.01024,
      "balance_loss_clip": 1.04611635,
      "balance_loss_mlp": 1.01617944,
      "epoch": 0.8315998316599531,
      "flos": 45185929338240.0,
      "grad_norm": 1.7086049262845278,
      "language_loss": 0.72062373,
      "learning_rate": 2.900359584363303e-07,
      "loss": 0.74231088,
      "num_input_tokens_seen": 149465045,
      "step": 6916,
      "time_per_iteration": 2.9222052097320557
    },
    {
      "auxiliary_loss_clip": 0.01104374,
      "auxiliary_loss_mlp": 0.01029059,
      "balance_loss_clip": 1.04558492,
      "balance_loss_mlp": 1.02093506,
      "epoch": 0.8317200745505922,
      "flos": 18363747479040.0,
      "grad_norm": 2.6523299296990337,
      "language_loss": 0.84721881,
      "learning_rate": 2.8963206855296494e-07,
      "loss": 0.86855316,
      "num_input_tokens_seen": 149481285,
      "step": 6917,
      "time_per_iteration": 2.7499372959136963
    },
    {
      "auxiliary_loss_clip": 0.01141008,
      "auxiliary_loss_mlp": 0.00762587,
      "balance_loss_clip": 1.04596257,
      "balance_loss_mlp": 1.00044882,
      "epoch": 0.8318403174412313,
      "flos": 24206557386240.0,
      "grad_norm": 2.095848260035102,
      "language_loss": 0.77342093,
      "learning_rate": 2.892284381310548e-07,
      "loss": 0.79245687,
      "num_input_tokens_seen": 149502700,
      "step": 6918,
      "time_per_iteration": 2.7228009700775146
    },
    {
      "auxiliary_loss_clip": 0.01153756,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.04572356,
      "balance_loss_mlp": 1.01779461,
      "epoch": 0.8319605603318704,
      "flos": 22418780641920.0,
      "grad_norm": 2.5273369709437343,
      "language_loss": 0.72169363,
      "learning_rate": 2.888250672318302e-07,
      "loss": 0.74349326,
      "num_input_tokens_seen": 149520100,
      "step": 6919,
      "time_per_iteration": 2.593596935272217
    },
    {
      "auxiliary_loss_clip": 0.01146299,
      "auxiliary_loss_mlp": 0.01024479,
      "balance_loss_clip": 1.04697967,
      "balance_loss_mlp": 1.01670921,
      "epoch": 0.8320808032225094,
      "flos": 37414501459200.0,
      "grad_norm": 2.4244255033171145,
      "language_loss": 0.68418968,
      "learning_rate": 2.884219559164831e-07,
      "loss": 0.70589745,
      "num_input_tokens_seen": 149543245,
      "step": 6920,
      "time_per_iteration": 2.8407537937164307
    },
    {
      "auxiliary_loss_clip": 0.01122661,
      "auxiliary_loss_mlp": 0.01028327,
      "balance_loss_clip": 1.0462606,
      "balance_loss_mlp": 1.01969028,
      "epoch": 0.8322010461131486,
      "flos": 12787395638400.0,
      "grad_norm": 2.1151574874186347,
      "language_loss": 0.81612206,
      "learning_rate": 2.880191042461635e-07,
      "loss": 0.83763194,
      "num_input_tokens_seen": 149559185,
      "step": 6921,
      "time_per_iteration": 2.6627724170684814
    },
    {
      "auxiliary_loss_clip": 0.01140872,
      "auxiliary_loss_mlp": 0.01025009,
      "balance_loss_clip": 1.0458622,
      "balance_loss_mlp": 1.01760328,
      "epoch": 0.8323212890037877,
      "flos": 15815455050240.0,
      "grad_norm": 2.1581683119862105,
      "language_loss": 0.80704451,
      "learning_rate": 2.876165122819849e-07,
      "loss": 0.82870322,
      "num_input_tokens_seen": 149577165,
      "step": 6922,
      "time_per_iteration": 2.722297191619873
    },
    {
      "auxiliary_loss_clip": 0.01124227,
      "auxiliary_loss_mlp": 0.010271,
      "balance_loss_clip": 1.04068494,
      "balance_loss_mlp": 1.01947045,
      "epoch": 0.8324415318944267,
      "flos": 21719276208000.0,
      "grad_norm": 1.9349822173112368,
      "language_loss": 0.7952044,
      "learning_rate": 2.872141800850201e-07,
      "loss": 0.81671762,
      "num_input_tokens_seen": 149594340,
      "step": 6923,
      "time_per_iteration": 2.6957035064697266
    },
    {
      "auxiliary_loss_clip": 0.01169706,
      "auxiliary_loss_mlp": 0.01021012,
      "balance_loss_clip": 1.04824936,
      "balance_loss_mlp": 1.01388931,
      "epoch": 0.8325617747850659,
      "flos": 34198700636160.0,
      "grad_norm": 2.0689065124583506,
      "language_loss": 0.73350179,
      "learning_rate": 2.868121077163024e-07,
      "loss": 0.75540894,
      "num_input_tokens_seen": 149613895,
      "step": 6924,
      "time_per_iteration": 3.6691527366638184
    },
    {
      "auxiliary_loss_clip": 0.01123706,
      "auxiliary_loss_mlp": 0.01023796,
      "balance_loss_clip": 1.04211211,
      "balance_loss_mlp": 1.01570725,
      "epoch": 0.8326820176757049,
      "flos": 18369457741440.0,
      "grad_norm": 1.8754429844244125,
      "language_loss": 0.72098023,
      "learning_rate": 2.864102952368257e-07,
      "loss": 0.74245524,
      "num_input_tokens_seen": 149631820,
      "step": 6925,
      "time_per_iteration": 2.6798057556152344
    },
    {
      "auxiliary_loss_clip": 0.01152448,
      "auxiliary_loss_mlp": 0.01020361,
      "balance_loss_clip": 1.04748297,
      "balance_loss_mlp": 1.01298189,
      "epoch": 0.832802260566344,
      "flos": 35991325716480.0,
      "grad_norm": 1.4353275343948229,
      "language_loss": 0.59426808,
      "learning_rate": 2.860087427075444e-07,
      "loss": 0.61599624,
      "num_input_tokens_seen": 149656070,
      "step": 6926,
      "time_per_iteration": 2.743070125579834
    },
    {
      "auxiliary_loss_clip": 0.01111232,
      "auxiliary_loss_mlp": 0.010258,
      "balance_loss_clip": 1.04060531,
      "balance_loss_mlp": 1.01784897,
      "epoch": 0.8329225034569832,
      "flos": 14244434928000.0,
      "grad_norm": 2.5031256780864575,
      "language_loss": 0.86237741,
      "learning_rate": 2.856074501893744e-07,
      "loss": 0.8837477,
      "num_input_tokens_seen": 149671270,
      "step": 6927,
      "time_per_iteration": 2.758601665496826
    },
    {
      "auxiliary_loss_clip": 0.0113708,
      "auxiliary_loss_mlp": 0.01024801,
      "balance_loss_clip": 1.04386258,
      "balance_loss_mlp": 1.01671231,
      "epoch": 0.8330427463476222,
      "flos": 18077468083200.0,
      "grad_norm": 1.8477790999596313,
      "language_loss": 0.82073104,
      "learning_rate": 2.8520641774319054e-07,
      "loss": 0.84234983,
      "num_input_tokens_seen": 149689360,
      "step": 6928,
      "time_per_iteration": 2.761632204055786
    },
    {
      "auxiliary_loss_clip": 0.01093614,
      "auxiliary_loss_mlp": 0.01030459,
      "balance_loss_clip": 1.03796434,
      "balance_loss_mlp": 1.02257323,
      "epoch": 0.8331629892382613,
      "flos": 18040839189120.0,
      "grad_norm": 2.4912155747495253,
      "language_loss": 0.76033103,
      "learning_rate": 2.848056454298309e-07,
      "loss": 0.78157175,
      "num_input_tokens_seen": 149706685,
      "step": 6929,
      "time_per_iteration": 2.7595362663269043
    },
    {
      "auxiliary_loss_clip": 0.01124433,
      "auxiliary_loss_mlp": 0.00762639,
      "balance_loss_clip": 1.04096222,
      "balance_loss_mlp": 1.00054646,
      "epoch": 0.8332832321289004,
      "flos": 17457398576640.0,
      "grad_norm": 2.9754972683529948,
      "language_loss": 0.65161562,
      "learning_rate": 2.844051333100905e-07,
      "loss": 0.67048633,
      "num_input_tokens_seen": 149724230,
      "step": 6930,
      "time_per_iteration": 2.7447433471679688
    },
    {
      "auxiliary_loss_clip": 0.01129095,
      "auxiliary_loss_mlp": 0.01022601,
      "balance_loss_clip": 1.0410428,
      "balance_loss_mlp": 1.01572871,
      "epoch": 0.8334034750195395,
      "flos": 15084852416640.0,
      "grad_norm": 2.2784724180833695,
      "language_loss": 0.83684921,
      "learning_rate": 2.840048814447269e-07,
      "loss": 0.85836613,
      "num_input_tokens_seen": 149742395,
      "step": 6931,
      "time_per_iteration": 2.725353479385376
    },
    {
      "auxiliary_loss_clip": 0.01146415,
      "auxiliary_loss_mlp": 0.01023602,
      "balance_loss_clip": 1.04601693,
      "balance_loss_mlp": 1.01618743,
      "epoch": 0.8335237179101785,
      "flos": 19427170556160.0,
      "grad_norm": 2.690471579276379,
      "language_loss": 0.73741668,
      "learning_rate": 2.836048898944587e-07,
      "loss": 0.75911689,
      "num_input_tokens_seen": 149760820,
      "step": 6932,
      "time_per_iteration": 2.661550283432007
    },
    {
      "auxiliary_loss_clip": 0.01124929,
      "auxiliary_loss_mlp": 0.01020277,
      "balance_loss_clip": 1.04161537,
      "balance_loss_mlp": 1.01275516,
      "epoch": 0.8336439608008177,
      "flos": 21762046327680.0,
      "grad_norm": 2.891412067309968,
      "language_loss": 0.72966337,
      "learning_rate": 2.832051587199642e-07,
      "loss": 0.7511155,
      "num_input_tokens_seen": 149778075,
      "step": 6933,
      "time_per_iteration": 2.730618476867676
    },
    {
      "auxiliary_loss_clip": 0.0105498,
      "auxiliary_loss_mlp": 0.01015033,
      "balance_loss_clip": 1.02040517,
      "balance_loss_mlp": 1.01386476,
      "epoch": 0.8337642036914568,
      "flos": 59702783990400.0,
      "grad_norm": 0.8234075281606131,
      "language_loss": 0.57731229,
      "learning_rate": 2.828056879818821e-07,
      "loss": 0.59801239,
      "num_input_tokens_seen": 149837150,
      "step": 6934,
      "time_per_iteration": 3.18058705329895
    },
    {
      "auxiliary_loss_clip": 0.01119032,
      "auxiliary_loss_mlp": 0.01022629,
      "balance_loss_clip": 1.03969884,
      "balance_loss_mlp": 1.01539278,
      "epoch": 0.8338844465820958,
      "flos": 27162185022720.0,
      "grad_norm": 2.5686213179774047,
      "language_loss": 0.83307469,
      "learning_rate": 2.824064777408117e-07,
      "loss": 0.85449135,
      "num_input_tokens_seen": 149856940,
      "step": 6935,
      "time_per_iteration": 2.780998468399048
    },
    {
      "auxiliary_loss_clip": 0.01135173,
      "auxiliary_loss_mlp": 0.01025399,
      "balance_loss_clip": 1.04127145,
      "balance_loss_mlp": 1.01741147,
      "epoch": 0.8340046894727349,
      "flos": 30481264425600.0,
      "grad_norm": 1.7596338999614354,
      "language_loss": 0.75693619,
      "learning_rate": 2.8200752805731263e-07,
      "loss": 0.77854192,
      "num_input_tokens_seen": 149879930,
      "step": 6936,
      "time_per_iteration": 2.8024442195892334
    },
    {
      "auxiliary_loss_clip": 0.0113338,
      "auxiliary_loss_mlp": 0.01026084,
      "balance_loss_clip": 1.04263949,
      "balance_loss_mlp": 1.01841855,
      "epoch": 0.834124932363374,
      "flos": 27126166659840.0,
      "grad_norm": 1.5691903039859603,
      "language_loss": 0.81031376,
      "learning_rate": 2.8160883899190625e-07,
      "loss": 0.8319084,
      "num_input_tokens_seen": 149903200,
      "step": 6937,
      "time_per_iteration": 3.74481201171875
    },
    {
      "auxiliary_loss_clip": 0.0117284,
      "auxiliary_loss_mlp": 0.01023181,
      "balance_loss_clip": 1.05040371,
      "balance_loss_mlp": 1.01581347,
      "epoch": 0.8342451752540131,
      "flos": 24569865498240.0,
      "grad_norm": 2.746338619152569,
      "language_loss": 0.73116642,
      "learning_rate": 2.8121041060507234e-07,
      "loss": 0.75312662,
      "num_input_tokens_seen": 149922230,
      "step": 6938,
      "time_per_iteration": 2.630927085876465
    },
    {
      "auxiliary_loss_clip": 0.01063056,
      "auxiliary_loss_mlp": 0.0102576,
      "balance_loss_clip": 1.03440571,
      "balance_loss_mlp": 1.01782036,
      "epoch": 0.8343654181446521,
      "flos": 26615085995520.0,
      "grad_norm": 2.1871979562555315,
      "language_loss": 0.71491516,
      "learning_rate": 2.808122429572528e-07,
      "loss": 0.73580337,
      "num_input_tokens_seen": 149942435,
      "step": 6939,
      "time_per_iteration": 5.00614595413208
    },
    {
      "auxiliary_loss_clip": 0.01075034,
      "auxiliary_loss_mlp": 0.0102235,
      "balance_loss_clip": 1.03541565,
      "balance_loss_mlp": 1.01492906,
      "epoch": 0.8344856610352913,
      "flos": 20777268078720.0,
      "grad_norm": 2.643223170854409,
      "language_loss": 0.76110953,
      "learning_rate": 2.804143361088489e-07,
      "loss": 0.78208339,
      "num_input_tokens_seen": 149961615,
      "step": 6940,
      "time_per_iteration": 3.104823350906372
    },
    {
      "auxiliary_loss_clip": 0.01073315,
      "auxiliary_loss_mlp": 0.00762232,
      "balance_loss_clip": 1.03893733,
      "balance_loss_mlp": 1.00045991,
      "epoch": 0.8346059039259304,
      "flos": 26095960684800.0,
      "grad_norm": 7.523486794021746,
      "language_loss": 0.77927142,
      "learning_rate": 2.8001669012022277e-07,
      "loss": 0.79762685,
      "num_input_tokens_seen": 149979585,
      "step": 6941,
      "time_per_iteration": 2.9543912410736084
    },
    {
      "auxiliary_loss_clip": 0.01132303,
      "auxiliary_loss_mlp": 0.01020936,
      "balance_loss_clip": 1.04509866,
      "balance_loss_mlp": 1.01330638,
      "epoch": 0.8347261468165694,
      "flos": 29027708755200.0,
      "grad_norm": 1.6049508782178017,
      "language_loss": 0.69262058,
      "learning_rate": 2.7961930505169795e-07,
      "loss": 0.71415299,
      "num_input_tokens_seen": 150003830,
      "step": 6942,
      "time_per_iteration": 2.8684279918670654
    },
    {
      "auxiliary_loss_clip": 0.01091043,
      "auxiliary_loss_mlp": 0.01021661,
      "balance_loss_clip": 1.03806114,
      "balance_loss_mlp": 1.01419842,
      "epoch": 0.8348463897072086,
      "flos": 26396461866240.0,
      "grad_norm": 4.448445791652661,
      "language_loss": 0.7684232,
      "learning_rate": 2.792221809635558e-07,
      "loss": 0.78955024,
      "num_input_tokens_seen": 150024460,
      "step": 6943,
      "time_per_iteration": 2.8200392723083496
    },
    {
      "auxiliary_loss_clip": 0.01108981,
      "auxiliary_loss_mlp": 0.01030136,
      "balance_loss_clip": 1.03911471,
      "balance_loss_mlp": 1.02198255,
      "epoch": 0.8349666325978476,
      "flos": 23367720096000.0,
      "grad_norm": 1.9928068200779676,
      "language_loss": 0.74798918,
      "learning_rate": 2.788253179160411e-07,
      "loss": 0.76938033,
      "num_input_tokens_seen": 150045620,
      "step": 6944,
      "time_per_iteration": 2.7887139320373535
    },
    {
      "auxiliary_loss_clip": 0.01125856,
      "auxiliary_loss_mlp": 0.00761943,
      "balance_loss_clip": 1.04813087,
      "balance_loss_mlp": 1.00038135,
      "epoch": 0.8350868754884867,
      "flos": 12896528135040.0,
      "grad_norm": 2.0252053298581223,
      "language_loss": 0.64732337,
      "learning_rate": 2.7842871596935725e-07,
      "loss": 0.66620135,
      "num_input_tokens_seen": 150064135,
      "step": 6945,
      "time_per_iteration": 2.718386173248291
    },
    {
      "auxiliary_loss_clip": 0.0113712,
      "auxiliary_loss_mlp": 0.01026363,
      "balance_loss_clip": 1.04328847,
      "balance_loss_mlp": 1.01875734,
      "epoch": 0.8352071183791259,
      "flos": 26505522535680.0,
      "grad_norm": 1.7116393252439175,
      "language_loss": 0.68859333,
      "learning_rate": 2.780323751836682e-07,
      "loss": 0.71022815,
      "num_input_tokens_seen": 150085350,
      "step": 6946,
      "time_per_iteration": 2.8638391494750977
    },
    {
      "auxiliary_loss_clip": 0.01137815,
      "auxiliary_loss_mlp": 0.01026599,
      "balance_loss_clip": 1.04546142,
      "balance_loss_mlp": 1.0195899,
      "epoch": 0.8353273612697649,
      "flos": 20668063754880.0,
      "grad_norm": 10.047266749589905,
      "language_loss": 0.78685188,
      "learning_rate": 2.7763629561909876e-07,
      "loss": 0.80849612,
      "num_input_tokens_seen": 150106180,
      "step": 6947,
      "time_per_iteration": 2.703496217727661
    },
    {
      "auxiliary_loss_clip": 0.01109806,
      "auxiliary_loss_mlp": 0.0076247,
      "balance_loss_clip": 1.04060435,
      "balance_loss_mlp": 1.00050557,
      "epoch": 0.835447604160404,
      "flos": 19754137082880.0,
      "grad_norm": 2.1640068569955533,
      "language_loss": 0.76812685,
      "learning_rate": 2.772404773357335e-07,
      "loss": 0.78684962,
      "num_input_tokens_seen": 150125585,
      "step": 6948,
      "time_per_iteration": 2.7285256385803223
    },
    {
      "auxiliary_loss_clip": 0.01132651,
      "auxiliary_loss_mlp": 0.00762376,
      "balance_loss_clip": 1.04418671,
      "balance_loss_mlp": 1.00050092,
      "epoch": 0.8355678470510431,
      "flos": 23435842239360.0,
      "grad_norm": 2.2175782150182703,
      "language_loss": 0.78658032,
      "learning_rate": 2.7684492039361853e-07,
      "loss": 0.80553061,
      "num_input_tokens_seen": 150144810,
      "step": 6949,
      "time_per_iteration": 3.7853243350982666
    },
    {
      "auxiliary_loss_clip": 0.01124488,
      "auxiliary_loss_mlp": 0.01027009,
      "balance_loss_clip": 1.04308128,
      "balance_loss_mlp": 1.01950216,
      "epoch": 0.8356880899416822,
      "flos": 21214588164480.0,
      "grad_norm": 1.6994130988658693,
      "language_loss": 0.8384766,
      "learning_rate": 2.764496248527586e-07,
      "loss": 0.85999155,
      "num_input_tokens_seen": 150163785,
      "step": 6950,
      "time_per_iteration": 2.715268611907959
    },
    {
      "auxiliary_loss_clip": 0.01067458,
      "auxiliary_loss_mlp": 0.01025728,
      "balance_loss_clip": 1.03964412,
      "balance_loss_mlp": 1.01758039,
      "epoch": 0.8358083328323213,
      "flos": 28037543466240.0,
      "grad_norm": 2.317971114352294,
      "language_loss": 0.78677166,
      "learning_rate": 2.760545907731211e-07,
      "loss": 0.8077035,
      "num_input_tokens_seen": 150184360,
      "step": 6951,
      "time_per_iteration": 2.975393533706665
    },
    {
      "auxiliary_loss_clip": 0.01120781,
      "auxiliary_loss_mlp": 0.01028175,
      "balance_loss_clip": 1.04401445,
      "balance_loss_mlp": 1.02014685,
      "epoch": 0.8359285757229604,
      "flos": 27783655159680.0,
      "grad_norm": 2.192675755677699,
      "language_loss": 0.68048382,
      "learning_rate": 2.75659818214631e-07,
      "loss": 0.70197338,
      "num_input_tokens_seen": 150205465,
      "step": 6952,
      "time_per_iteration": 3.1757888793945312
    },
    {
      "auxiliary_loss_clip": 0.01115816,
      "auxiliary_loss_mlp": 0.0102398,
      "balance_loss_clip": 1.04042315,
      "balance_loss_mlp": 1.01653504,
      "epoch": 0.8360488186135995,
      "flos": 21435115714560.0,
      "grad_norm": 1.8006649571501636,
      "language_loss": 0.78359288,
      "learning_rate": 2.752653072371749e-07,
      "loss": 0.80499077,
      "num_input_tokens_seen": 150224900,
      "step": 6953,
      "time_per_iteration": 2.802274227142334
    },
    {
      "auxiliary_loss_clip": 0.01095802,
      "auxiliary_loss_mlp": 0.01022978,
      "balance_loss_clip": 1.03677559,
      "balance_loss_mlp": 1.01546216,
      "epoch": 0.8361690615042385,
      "flos": 27632327160960.0,
      "grad_norm": 1.9583827482076817,
      "language_loss": 0.74842942,
      "learning_rate": 2.7487105790060105e-07,
      "loss": 0.7696172,
      "num_input_tokens_seen": 150244310,
      "step": 6954,
      "time_per_iteration": 2.8829219341278076
    },
    {
      "auxiliary_loss_clip": 0.01104816,
      "auxiliary_loss_mlp": 0.01021929,
      "balance_loss_clip": 1.03755641,
      "balance_loss_mlp": 1.01485419,
      "epoch": 0.8362893043948777,
      "flos": 39202529598720.0,
      "grad_norm": 2.6730512760796348,
      "language_loss": 0.69672883,
      "learning_rate": 2.7447707026471587e-07,
      "loss": 0.71799636,
      "num_input_tokens_seen": 150267285,
      "step": 6955,
      "time_per_iteration": 2.86820125579834
    },
    {
      "auxiliary_loss_clip": 0.01121419,
      "auxiliary_loss_mlp": 0.01023803,
      "balance_loss_clip": 1.04251695,
      "balance_loss_mlp": 1.01632917,
      "epoch": 0.8364095472855168,
      "flos": 24785329230720.0,
      "grad_norm": 3.1423587061226965,
      "language_loss": 0.79983044,
      "learning_rate": 2.740833443892874e-07,
      "loss": 0.82128263,
      "num_input_tokens_seen": 150285455,
      "step": 6956,
      "time_per_iteration": 2.7719593048095703
    },
    {
      "auxiliary_loss_clip": 0.01115159,
      "auxiliary_loss_mlp": 0.01025014,
      "balance_loss_clip": 1.03825045,
      "balance_loss_mlp": 1.01748586,
      "epoch": 0.8365297901761558,
      "flos": 22743412784640.0,
      "grad_norm": 2.0177486980413115,
      "language_loss": 0.79938579,
      "learning_rate": 2.7368988033404327e-07,
      "loss": 0.82078755,
      "num_input_tokens_seen": 150302970,
      "step": 6957,
      "time_per_iteration": 2.7412705421447754
    },
    {
      "auxiliary_loss_clip": 0.01139816,
      "auxiliary_loss_mlp": 0.01021914,
      "balance_loss_clip": 1.04394007,
      "balance_loss_mlp": 1.01397431,
      "epoch": 0.836650033066795,
      "flos": 28396003242240.0,
      "grad_norm": 1.7173331311728708,
      "language_loss": 0.84776175,
      "learning_rate": 2.732966781586712e-07,
      "loss": 0.86937904,
      "num_input_tokens_seen": 150322715,
      "step": 6958,
      "time_per_iteration": 2.781796932220459
    },
    {
      "auxiliary_loss_clip": 0.01151761,
      "auxiliary_loss_mlp": 0.01023641,
      "balance_loss_clip": 1.044245,
      "balance_loss_mlp": 1.01605618,
      "epoch": 0.836770275957434,
      "flos": 22236857233920.0,
      "grad_norm": 1.6616020818643038,
      "language_loss": 0.665932,
      "learning_rate": 2.729037379228205e-07,
      "loss": 0.68768597,
      "num_input_tokens_seen": 150342900,
      "step": 6959,
      "time_per_iteration": 2.7109711170196533
    },
    {
      "auxiliary_loss_clip": 0.01115666,
      "auxiliary_loss_mlp": 0.00762803,
      "balance_loss_clip": 1.04515636,
      "balance_loss_mlp": 1.00053334,
      "epoch": 0.8368905188480731,
      "flos": 22491930689280.0,
      "grad_norm": 1.5997328908781647,
      "language_loss": 0.80485153,
      "learning_rate": 2.725110596860998e-07,
      "loss": 0.82363617,
      "num_input_tokens_seen": 150363580,
      "step": 6960,
      "time_per_iteration": 2.872666597366333
    },
    {
      "auxiliary_loss_clip": 0.01138897,
      "auxiliary_loss_mlp": 0.01022874,
      "balance_loss_clip": 1.04756355,
      "balance_loss_mlp": 1.01584387,
      "epoch": 0.8370107617387123,
      "flos": 13370405287680.0,
      "grad_norm": 43.721113561882035,
      "language_loss": 0.70388305,
      "learning_rate": 2.7211864350807776e-07,
      "loss": 0.72550082,
      "num_input_tokens_seen": 150381780,
      "step": 6961,
      "time_per_iteration": 2.7697691917419434
    },
    {
      "auxiliary_loss_clip": 0.01125996,
      "auxiliary_loss_mlp": 0.01022493,
      "balance_loss_clip": 1.04058945,
      "balance_loss_mlp": 1.01409495,
      "epoch": 0.8371310046293513,
      "flos": 25261289372160.0,
      "grad_norm": 1.7055690143938809,
      "language_loss": 0.73711979,
      "learning_rate": 2.717264894482836e-07,
      "loss": 0.75860465,
      "num_input_tokens_seen": 150402120,
      "step": 6962,
      "time_per_iteration": 2.7451062202453613
    },
    {
      "auxiliary_loss_clip": 0.0114922,
      "auxiliary_loss_mlp": 0.01025457,
      "balance_loss_clip": 1.04553342,
      "balance_loss_mlp": 1.01738703,
      "epoch": 0.8372512475199904,
      "flos": 19792705311360.0,
      "grad_norm": 2.345711085595281,
      "language_loss": 0.80993986,
      "learning_rate": 2.7133459756620646e-07,
      "loss": 0.83168662,
      "num_input_tokens_seen": 150419315,
      "step": 6963,
      "time_per_iteration": 3.6555566787719727
    },
    {
      "auxiliary_loss_clip": 0.01160316,
      "auxiliary_loss_mlp": 0.01028209,
      "balance_loss_clip": 1.04830718,
      "balance_loss_mlp": 1.02090442,
      "epoch": 0.8373714904106295,
      "flos": 19391224020480.0,
      "grad_norm": 11.436423425878967,
      "language_loss": 0.73591781,
      "learning_rate": 2.7094296792129733e-07,
      "loss": 0.75780308,
      "num_input_tokens_seen": 150438915,
      "step": 6964,
      "time_per_iteration": 4.718454360961914
    },
    {
      "auxiliary_loss_clip": 0.0114245,
      "auxiliary_loss_mlp": 0.01025751,
      "balance_loss_clip": 1.04592144,
      "balance_loss_mlp": 1.01847029,
      "epoch": 0.8374917333012686,
      "flos": 14975935401600.0,
      "grad_norm": 1.9807805996504588,
      "language_loss": 0.75206745,
      "learning_rate": 2.7055160057296424e-07,
      "loss": 0.77374941,
      "num_input_tokens_seen": 150456155,
      "step": 6965,
      "time_per_iteration": 2.708911895751953
    },
    {
      "auxiliary_loss_clip": 0.01136106,
      "auxiliary_loss_mlp": 0.0102136,
      "balance_loss_clip": 1.04457366,
      "balance_loss_mlp": 1.0139544,
      "epoch": 0.8376119761919076,
      "flos": 30331839847680.0,
      "grad_norm": 2.0567990593579113,
      "language_loss": 0.72364837,
      "learning_rate": 2.7016049558057896e-07,
      "loss": 0.74522305,
      "num_input_tokens_seen": 150478115,
      "step": 6966,
      "time_per_iteration": 2.762096643447876
    },
    {
      "auxiliary_loss_clip": 0.01141603,
      "auxiliary_loss_mlp": 0.01022824,
      "balance_loss_clip": 1.04354227,
      "balance_loss_mlp": 1.01521516,
      "epoch": 0.8377322190825467,
      "flos": 29423336129280.0,
      "grad_norm": 2.691678254260203,
      "language_loss": 0.71018177,
      "learning_rate": 2.6976965300347074e-07,
      "loss": 0.73182607,
      "num_input_tokens_seen": 150500725,
      "step": 6967,
      "time_per_iteration": 2.754753589630127
    },
    {
      "auxiliary_loss_clip": 0.01118639,
      "auxiliary_loss_mlp": 0.01026251,
      "balance_loss_clip": 1.03838706,
      "balance_loss_mlp": 1.01864517,
      "epoch": 0.8378524619731859,
      "flos": 26687086807680.0,
      "grad_norm": 2.9939426044117807,
      "language_loss": 0.69532335,
      "learning_rate": 2.693790729009309e-07,
      "loss": 0.71677226,
      "num_input_tokens_seen": 150522335,
      "step": 6968,
      "time_per_iteration": 2.72833251953125
    },
    {
      "auxiliary_loss_clip": 0.01133974,
      "auxiliary_loss_mlp": 0.01023152,
      "balance_loss_clip": 1.04170108,
      "balance_loss_mlp": 1.0152365,
      "epoch": 0.8379727048638249,
      "flos": 20703866636160.0,
      "grad_norm": 1.903871399623538,
      "language_loss": 0.88699937,
      "learning_rate": 2.6898875533220946e-07,
      "loss": 0.90857065,
      "num_input_tokens_seen": 150541640,
      "step": 6969,
      "time_per_iteration": 2.711031436920166
    },
    {
      "auxiliary_loss_clip": 0.01108877,
      "auxiliary_loss_mlp": 0.00760752,
      "balance_loss_clip": 1.03954768,
      "balance_loss_mlp": 1.00048351,
      "epoch": 0.838092947754464,
      "flos": 20084084438400.0,
      "grad_norm": 2.072216651444191,
      "language_loss": 0.82089883,
      "learning_rate": 2.685987003565171e-07,
      "loss": 0.83959514,
      "num_input_tokens_seen": 150559680,
      "step": 6970,
      "time_per_iteration": 2.7941477298736572
    },
    {
      "auxiliary_loss_clip": 0.01132443,
      "auxiliary_loss_mlp": 0.01024102,
      "balance_loss_clip": 1.04171181,
      "balance_loss_mlp": 1.01639557,
      "epoch": 0.8382131906451031,
      "flos": 18113270964480.0,
      "grad_norm": 3.052545517623763,
      "language_loss": 0.75533915,
      "learning_rate": 2.6820890803302566e-07,
      "loss": 0.77690458,
      "num_input_tokens_seen": 150575205,
      "step": 6971,
      "time_per_iteration": 2.814697742462158
    },
    {
      "auxiliary_loss_clip": 0.01138989,
      "auxiliary_loss_mlp": 0.01026128,
      "balance_loss_clip": 1.04351568,
      "balance_loss_mlp": 1.01893353,
      "epoch": 0.8383334335357422,
      "flos": 17092653920640.0,
      "grad_norm": 3.6676761080156055,
      "language_loss": 0.81582022,
      "learning_rate": 2.6781937842086557e-07,
      "loss": 0.83747137,
      "num_input_tokens_seen": 150593995,
      "step": 6972,
      "time_per_iteration": 2.6406376361846924
    },
    {
      "auxiliary_loss_clip": 0.01143271,
      "auxiliary_loss_mlp": 0.01022091,
      "balance_loss_clip": 1.04380023,
      "balance_loss_mlp": 1.01465487,
      "epoch": 0.8384536764263812,
      "flos": 20704728562560.0,
      "grad_norm": 2.090818956768083,
      "language_loss": 0.67391646,
      "learning_rate": 2.6743011157912933e-07,
      "loss": 0.69557011,
      "num_input_tokens_seen": 150613715,
      "step": 6973,
      "time_per_iteration": 2.7251908779144287
    },
    {
      "auxiliary_loss_clip": 0.01155397,
      "auxiliary_loss_mlp": 0.01028728,
      "balance_loss_clip": 1.04812562,
      "balance_loss_mlp": 1.02094674,
      "epoch": 0.8385739193170204,
      "flos": 28986842056320.0,
      "grad_norm": 2.6258587254471584,
      "language_loss": 0.65606606,
      "learning_rate": 2.6704110756686725e-07,
      "loss": 0.67790729,
      "num_input_tokens_seen": 150634540,
      "step": 6974,
      "time_per_iteration": 2.705806255340576
    },
    {
      "auxiliary_loss_clip": 0.01143122,
      "auxiliary_loss_mlp": 0.01024602,
      "balance_loss_clip": 1.04474354,
      "balance_loss_mlp": 1.01648402,
      "epoch": 0.8386941622076595,
      "flos": 23438068882560.0,
      "grad_norm": 1.8817495752720128,
      "language_loss": 0.83549547,
      "learning_rate": 2.6665236644309085e-07,
      "loss": 0.85717273,
      "num_input_tokens_seen": 150654850,
      "step": 6975,
      "time_per_iteration": 3.6551764011383057
    },
    {
      "auxiliary_loss_clip": 0.01119275,
      "auxiliary_loss_mlp": 0.01024716,
      "balance_loss_clip": 1.04107022,
      "balance_loss_mlp": 1.01704514,
      "epoch": 0.8388144050982985,
      "flos": 23002724044800.0,
      "grad_norm": 1.9089797006020195,
      "language_loss": 0.79574704,
      "learning_rate": 2.662638882667727e-07,
      "loss": 0.81718695,
      "num_input_tokens_seen": 150673790,
      "step": 6976,
      "time_per_iteration": 2.811980962753296
    },
    {
      "auxiliary_loss_clip": 0.01112397,
      "auxiliary_loss_mlp": 0.01030759,
      "balance_loss_clip": 1.03902459,
      "balance_loss_mlp": 1.0223788,
      "epoch": 0.8389346479889377,
      "flos": 24280353878400.0,
      "grad_norm": 1.8086835862530497,
      "language_loss": 0.73142594,
      "learning_rate": 2.658756730968443e-07,
      "loss": 0.75285745,
      "num_input_tokens_seen": 150692255,
      "step": 6977,
      "time_per_iteration": 2.7979748249053955
    },
    {
      "auxiliary_loss_clip": 0.01117892,
      "auxiliary_loss_mlp": 0.01029207,
      "balance_loss_clip": 1.04600322,
      "balance_loss_mlp": 1.02129149,
      "epoch": 0.8390548908795767,
      "flos": 21215019127680.0,
      "grad_norm": 2.11080213130592,
      "language_loss": 0.88217664,
      "learning_rate": 2.654877209921975e-07,
      "loss": 0.90364766,
      "num_input_tokens_seen": 150709790,
      "step": 6978,
      "time_per_iteration": 2.7550272941589355
    },
    {
      "auxiliary_loss_clip": 0.01111733,
      "auxiliary_loss_mlp": 0.01032205,
      "balance_loss_clip": 1.04340005,
      "balance_loss_mlp": 1.02342582,
      "epoch": 0.8391751337702158,
      "flos": 35627299332480.0,
      "grad_norm": 3.918686789198966,
      "language_loss": 0.62845147,
      "learning_rate": 2.651000320116843e-07,
      "loss": 0.64989084,
      "num_input_tokens_seen": 150730675,
      "step": 6979,
      "time_per_iteration": 2.868948221206665
    },
    {
      "auxiliary_loss_clip": 0.01138955,
      "auxiliary_loss_mlp": 0.01027361,
      "balance_loss_clip": 1.04551113,
      "balance_loss_mlp": 1.01930213,
      "epoch": 0.839295376660855,
      "flos": 21325229032320.0,
      "grad_norm": 1.8087549101279956,
      "language_loss": 0.76363349,
      "learning_rate": 2.647126062141163e-07,
      "loss": 0.78529668,
      "num_input_tokens_seen": 150749750,
      "step": 6980,
      "time_per_iteration": 2.704684019088745
    },
    {
      "auxiliary_loss_clip": 0.01085465,
      "auxiliary_loss_mlp": 0.01026184,
      "balance_loss_clip": 1.03640687,
      "balance_loss_mlp": 1.0185343,
      "epoch": 0.839415619551494,
      "flos": 18442535961600.0,
      "grad_norm": 1.7260393306703377,
      "language_loss": 0.83959967,
      "learning_rate": 2.643254436582669e-07,
      "loss": 0.86071616,
      "num_input_tokens_seen": 150769240,
      "step": 6981,
      "time_per_iteration": 2.7555599212646484
    },
    {
      "auxiliary_loss_clip": 0.01118552,
      "auxiliary_loss_mlp": 0.00762174,
      "balance_loss_clip": 1.04251552,
      "balance_loss_mlp": 1.00050056,
      "epoch": 0.8395358624421331,
      "flos": 23221958705280.0,
      "grad_norm": 1.824291995551202,
      "language_loss": 0.82403821,
      "learning_rate": 2.6393854440286743e-07,
      "loss": 0.84284544,
      "num_input_tokens_seen": 150788410,
      "step": 6982,
      "time_per_iteration": 2.8417327404022217
    },
    {
      "auxiliary_loss_clip": 0.01094538,
      "auxiliary_loss_mlp": 0.01026589,
      "balance_loss_clip": 1.04188967,
      "balance_loss_mlp": 1.01919222,
      "epoch": 0.8396561053327722,
      "flos": 24381657210240.0,
      "grad_norm": 1.8830868748973497,
      "language_loss": 0.70945537,
      "learning_rate": 2.6355190850661045e-07,
      "loss": 0.73066664,
      "num_input_tokens_seen": 150805245,
      "step": 6983,
      "time_per_iteration": 2.795053482055664
    },
    {
      "auxiliary_loss_clip": 0.01110032,
      "auxiliary_loss_mlp": 0.01026648,
      "balance_loss_clip": 1.03983688,
      "balance_loss_mlp": 1.0186727,
      "epoch": 0.8397763482234113,
      "flos": 22237755073920.0,
      "grad_norm": 1.7277331357785919,
      "language_loss": 0.86412144,
      "learning_rate": 2.631655360281486e-07,
      "loss": 0.88548821,
      "num_input_tokens_seen": 150824920,
      "step": 6984,
      "time_per_iteration": 2.669363021850586
    },
    {
      "auxiliary_loss_clip": 0.01129728,
      "auxiliary_loss_mlp": 0.01022939,
      "balance_loss_clip": 1.04266143,
      "balance_loss_mlp": 1.01501179,
      "epoch": 0.8398965911140504,
      "flos": 22163743100160.0,
      "grad_norm": 2.0919913025151566,
      "language_loss": 0.65641761,
      "learning_rate": 2.6277942702609323e-07,
      "loss": 0.6779443,
      "num_input_tokens_seen": 150844400,
      "step": 6985,
      "time_per_iteration": 2.7946553230285645
    },
    {
      "auxiliary_loss_clip": 0.01146291,
      "auxiliary_loss_mlp": 0.01029077,
      "balance_loss_clip": 1.04627419,
      "balance_loss_mlp": 1.02145982,
      "epoch": 0.8400168340046895,
      "flos": 21542775753600.0,
      "grad_norm": 2.1030894743029873,
      "language_loss": 0.87451208,
      "learning_rate": 2.623935815590186e-07,
      "loss": 0.8962658,
      "num_input_tokens_seen": 150862780,
      "step": 6986,
      "time_per_iteration": 2.6824092864990234
    },
    {
      "auxiliary_loss_clip": 0.01169783,
      "auxiliary_loss_mlp": 0.01028629,
      "balance_loss_clip": 1.04834938,
      "balance_loss_mlp": 1.02162528,
      "epoch": 0.8401370768953286,
      "flos": 22491966602880.0,
      "grad_norm": 1.8847887849582523,
      "language_loss": 0.8099165,
      "learning_rate": 2.6200799968545516e-07,
      "loss": 0.83190066,
      "num_input_tokens_seen": 150883075,
      "step": 6987,
      "time_per_iteration": 2.669919490814209
    },
    {
      "auxiliary_loss_clip": 0.01032967,
      "auxiliary_loss_mlp": 0.00998988,
      "balance_loss_clip": 1.010993,
      "balance_loss_mlp": 0.99794531,
      "epoch": 0.8402573197859676,
      "flos": 59238890818560.0,
      "grad_norm": 0.8041388388861916,
      "language_loss": 0.56387335,
      "learning_rate": 2.616226814638969e-07,
      "loss": 0.58419287,
      "num_input_tokens_seen": 150948180,
      "step": 6988,
      "time_per_iteration": 3.3506486415863037
    },
    {
      "auxiliary_loss_clip": 0.01126253,
      "auxiliary_loss_mlp": 0.0102281,
      "balance_loss_clip": 1.04393244,
      "balance_loss_mlp": 1.01544905,
      "epoch": 0.8403775626766068,
      "flos": 22674608282880.0,
      "grad_norm": 1.9399860779093039,
      "language_loss": 0.77429485,
      "learning_rate": 2.612376269527954e-07,
      "loss": 0.79578555,
      "num_input_tokens_seen": 150967885,
      "step": 6989,
      "time_per_iteration": 3.6769068241119385
    },
    {
      "auxiliary_loss_clip": 0.01127542,
      "auxiliary_loss_mlp": 0.00762612,
      "balance_loss_clip": 1.04633975,
      "balance_loss_mlp": 1.00048935,
      "epoch": 0.8404978055672458,
      "flos": 19609704495360.0,
      "grad_norm": 4.542170457541228,
      "language_loss": 0.67485195,
      "learning_rate": 2.608528362105635e-07,
      "loss": 0.69375348,
      "num_input_tokens_seen": 150987255,
      "step": 6990,
      "time_per_iteration": 2.7649126052856445
    },
    {
      "auxiliary_loss_clip": 0.01111411,
      "auxiliary_loss_mlp": 0.01020403,
      "balance_loss_clip": 1.04173493,
      "balance_loss_mlp": 1.01283336,
      "epoch": 0.8406180484578849,
      "flos": 27526929678720.0,
      "grad_norm": 1.8629044787945104,
      "language_loss": 0.73408198,
      "learning_rate": 2.6046830929557374e-07,
      "loss": 0.75540012,
      "num_input_tokens_seen": 151006905,
      "step": 6991,
      "time_per_iteration": 3.6501760482788086
    },
    {
      "auxiliary_loss_clip": 0.01122607,
      "auxiliary_loss_mlp": 0.01029842,
      "balance_loss_clip": 1.04351354,
      "balance_loss_mlp": 1.02180743,
      "epoch": 0.8407382913485241,
      "flos": 22127473342080.0,
      "grad_norm": 4.890625622729905,
      "language_loss": 0.84730196,
      "learning_rate": 2.6008404626615776e-07,
      "loss": 0.86882651,
      "num_input_tokens_seen": 151025405,
      "step": 6992,
      "time_per_iteration": 2.653337240219116
    },
    {
      "auxiliary_loss_clip": 0.01108053,
      "auxiliary_loss_mlp": 0.01025407,
      "balance_loss_clip": 1.0402559,
      "balance_loss_mlp": 1.01766765,
      "epoch": 0.8408585342391631,
      "flos": 13918473982080.0,
      "grad_norm": 2.6064338536229203,
      "language_loss": 0.7371875,
      "learning_rate": 2.597000471806092e-07,
      "loss": 0.75852215,
      "num_input_tokens_seen": 151041970,
      "step": 6993,
      "time_per_iteration": 2.622467517852783
    },
    {
      "auxiliary_loss_clip": 0.0114212,
      "auxiliary_loss_mlp": 0.01023011,
      "balance_loss_clip": 1.04732716,
      "balance_loss_mlp": 1.01436889,
      "epoch": 0.8409787771298022,
      "flos": 20187865808640.0,
      "grad_norm": 1.9497072586428148,
      "language_loss": 0.73120201,
      "learning_rate": 2.593163120971793e-07,
      "loss": 0.75285327,
      "num_input_tokens_seen": 151060835,
      "step": 6994,
      "time_per_iteration": 2.64571475982666
    },
    {
      "auxiliary_loss_clip": 0.01128409,
      "auxiliary_loss_mlp": 0.01022819,
      "balance_loss_clip": 1.04311633,
      "balance_loss_mlp": 1.01546717,
      "epoch": 0.8410990200204413,
      "flos": 23142523777920.0,
      "grad_norm": 1.9524919073359592,
      "language_loss": 0.69130957,
      "learning_rate": 2.5893284107408165e-07,
      "loss": 0.71282184,
      "num_input_tokens_seen": 151078205,
      "step": 6995,
      "time_per_iteration": 2.672945976257324
    },
    {
      "auxiliary_loss_clip": 0.01160931,
      "auxiliary_loss_mlp": 0.00762238,
      "balance_loss_clip": 1.04782534,
      "balance_loss_mlp": 1.00053775,
      "epoch": 0.8412192629110804,
      "flos": 24027219757440.0,
      "grad_norm": 3.0178534987075074,
      "language_loss": 0.77942479,
      "learning_rate": 2.5854963416948726e-07,
      "loss": 0.79865646,
      "num_input_tokens_seen": 151100470,
      "step": 6996,
      "time_per_iteration": 2.6763205528259277
    },
    {
      "auxiliary_loss_clip": 0.01126196,
      "auxiliary_loss_mlp": 0.01027585,
      "balance_loss_clip": 1.04374838,
      "balance_loss_mlp": 1.01942515,
      "epoch": 0.8413395058017195,
      "flos": 25591703604480.0,
      "grad_norm": 2.4233053720272113,
      "language_loss": 0.69652462,
      "learning_rate": 2.5816669144152816e-07,
      "loss": 0.7180624,
      "num_input_tokens_seen": 151121650,
      "step": 6997,
      "time_per_iteration": 2.7799606323242188
    },
    {
      "auxiliary_loss_clip": 0.01043371,
      "auxiliary_loss_mlp": 0.01001332,
      "balance_loss_clip": 1.01344848,
      "balance_loss_mlp": 1.00041986,
      "epoch": 0.8414597486923585,
      "flos": 63635396624640.0,
      "grad_norm": 0.8526813707774851,
      "language_loss": 0.66293263,
      "learning_rate": 2.5778401294829777e-07,
      "loss": 0.68337965,
      "num_input_tokens_seen": 151180390,
      "step": 6998,
      "time_per_iteration": 3.316193103790283
    },
    {
      "auxiliary_loss_clip": 0.01134527,
      "auxiliary_loss_mlp": 0.01024049,
      "balance_loss_clip": 1.04344988,
      "balance_loss_mlp": 1.01658106,
      "epoch": 0.8415799915829977,
      "flos": 19098731571840.0,
      "grad_norm": 8.580901712584945,
      "language_loss": 0.65324688,
      "learning_rate": 2.574015987478473e-07,
      "loss": 0.6748327,
      "num_input_tokens_seen": 151198520,
      "step": 6999,
      "time_per_iteration": 2.6417856216430664
    },
    {
      "auxiliary_loss_clip": 0.0115227,
      "auxiliary_loss_mlp": 0.01022737,
      "balance_loss_clip": 1.04651892,
      "balance_loss_mlp": 1.01475859,
      "epoch": 0.8417002344736367,
      "flos": 19821612781440.0,
      "grad_norm": 2.6950837039840905,
      "language_loss": 0.87313414,
      "learning_rate": 2.570194488981887e-07,
      "loss": 0.89488423,
      "num_input_tokens_seen": 151215065,
      "step": 7000,
      "time_per_iteration": 3.4568512439727783
    },
    {
      "auxiliary_loss_clip": 0.01040897,
      "auxiliary_loss_mlp": 0.01000512,
      "balance_loss_clip": 1.01097643,
      "balance_loss_mlp": 0.99948734,
      "epoch": 0.8418204773642758,
      "flos": 62161516834560.0,
      "grad_norm": 0.8981467654358487,
      "language_loss": 0.60334778,
      "learning_rate": 2.566375634572939e-07,
      "loss": 0.62376189,
      "num_input_tokens_seen": 151275705,
      "step": 7001,
      "time_per_iteration": 3.183722496032715
    },
    {
      "auxiliary_loss_clip": 0.01111826,
      "auxiliary_loss_mlp": 0.01028347,
      "balance_loss_clip": 1.04065418,
      "balance_loss_mlp": 1.02012134,
      "epoch": 0.841940720254915,
      "flos": 17092905315840.0,
      "grad_norm": 1.8879154169276253,
      "language_loss": 0.76372278,
      "learning_rate": 2.562559424830943e-07,
      "loss": 0.78512448,
      "num_input_tokens_seen": 151293665,
      "step": 7002,
      "time_per_iteration": 2.917278289794922
    },
    {
      "auxiliary_loss_clip": 0.011455,
      "auxiliary_loss_mlp": 0.01022397,
      "balance_loss_clip": 1.04760385,
      "balance_loss_mlp": 1.01424885,
      "epoch": 0.842060963145554,
      "flos": 16283586026880.0,
      "grad_norm": 2.3779721923362027,
      "language_loss": 0.70256001,
      "learning_rate": 2.5587458603348256e-07,
      "loss": 0.72423893,
      "num_input_tokens_seen": 151310955,
      "step": 7003,
      "time_per_iteration": 2.589146375656128
    },
    {
      "auxiliary_loss_clip": 0.01127284,
      "auxiliary_loss_mlp": 0.01026302,
      "balance_loss_clip": 1.04155374,
      "balance_loss_mlp": 1.01867318,
      "epoch": 0.8421812060361931,
      "flos": 21908238681600.0,
      "grad_norm": 2.900124168337498,
      "language_loss": 0.83931071,
      "learning_rate": 2.554934941663085e-07,
      "loss": 0.86084652,
      "num_input_tokens_seen": 151328490,
      "step": 7004,
      "time_per_iteration": 2.716183662414551
    },
    {
      "auxiliary_loss_clip": 0.01111421,
      "auxiliary_loss_mlp": 0.00762293,
      "balance_loss_clip": 1.04218459,
      "balance_loss_mlp": 1.00041819,
      "epoch": 0.8423014489268322,
      "flos": 27777693502080.0,
      "grad_norm": 2.043608142283966,
      "language_loss": 0.73340929,
      "learning_rate": 2.5511266693938484e-07,
      "loss": 0.75214642,
      "num_input_tokens_seen": 151346950,
      "step": 7005,
      "time_per_iteration": 2.7961010932922363
    },
    {
      "auxiliary_loss_clip": 0.01125651,
      "auxiliary_loss_mlp": 0.01026965,
      "balance_loss_clip": 1.04235089,
      "balance_loss_mlp": 1.01841211,
      "epoch": 0.8424216918174713,
      "flos": 25117610970240.0,
      "grad_norm": 1.8927738781070302,
      "language_loss": 0.77761924,
      "learning_rate": 2.547321044104822e-07,
      "loss": 0.79914534,
      "num_input_tokens_seen": 151368445,
      "step": 7006,
      "time_per_iteration": 2.727713108062744
    },
    {
      "auxiliary_loss_clip": 0.01134318,
      "auxiliary_loss_mlp": 0.0102859,
      "balance_loss_clip": 1.04590034,
      "balance_loss_mlp": 1.02063894,
      "epoch": 0.8425419347081103,
      "flos": 24748448941440.0,
      "grad_norm": 1.867802628266386,
      "language_loss": 0.76828325,
      "learning_rate": 2.5435180663733113e-07,
      "loss": 0.7899124,
      "num_input_tokens_seen": 151388745,
      "step": 7007,
      "time_per_iteration": 2.755150318145752
    },
    {
      "auxiliary_loss_clip": 0.01111828,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.04189992,
      "balance_loss_mlp": 1.01713562,
      "epoch": 0.8426621775987495,
      "flos": 24820916630400.0,
      "grad_norm": 3.218079732028488,
      "language_loss": 0.71666384,
      "learning_rate": 2.539717736776241e-07,
      "loss": 0.73802966,
      "num_input_tokens_seen": 151404970,
      "step": 7008,
      "time_per_iteration": 2.754270553588867
    },
    {
      "auxiliary_loss_clip": 0.01128685,
      "auxiliary_loss_mlp": 0.01025536,
      "balance_loss_clip": 1.04554594,
      "balance_loss_mlp": 1.01755536,
      "epoch": 0.8427824204893886,
      "flos": 23550074467200.0,
      "grad_norm": 2.852284290265349,
      "language_loss": 0.76444203,
      "learning_rate": 2.535920055890097e-07,
      "loss": 0.78598422,
      "num_input_tokens_seen": 151426265,
      "step": 7009,
      "time_per_iteration": 2.7465429306030273
    },
    {
      "auxiliary_loss_clip": 0.01161051,
      "auxiliary_loss_mlp": 0.01027431,
      "balance_loss_clip": 1.04877734,
      "balance_loss_mlp": 1.01937866,
      "epoch": 0.8429026633800276,
      "flos": 16143858120960.0,
      "grad_norm": 8.993535007287678,
      "language_loss": 0.64563411,
      "learning_rate": 2.5321250242910006e-07,
      "loss": 0.66751897,
      "num_input_tokens_seen": 151444180,
      "step": 7010,
      "time_per_iteration": 2.642763614654541
    },
    {
      "auxiliary_loss_clip": 0.01112641,
      "auxiliary_loss_mlp": 0.01025603,
      "balance_loss_clip": 1.04596138,
      "balance_loss_mlp": 1.01797974,
      "epoch": 0.8430229062706668,
      "flos": 22198540400640.0,
      "grad_norm": 1.7862369194612608,
      "language_loss": 0.86377668,
      "learning_rate": 2.5283326425546493e-07,
      "loss": 0.88515913,
      "num_input_tokens_seen": 151463290,
      "step": 7011,
      "time_per_iteration": 2.7556653022766113
    },
    {
      "auxiliary_loss_clip": 0.01104333,
      "auxiliary_loss_mlp": 0.01028972,
      "balance_loss_clip": 1.04109204,
      "balance_loss_mlp": 1.02127433,
      "epoch": 0.8431431491613058,
      "flos": 35330317683840.0,
      "grad_norm": 3.4749686268074216,
      "language_loss": 0.69403929,
      "learning_rate": 2.5245429112563443e-07,
      "loss": 0.71537232,
      "num_input_tokens_seen": 151483965,
      "step": 7012,
      "time_per_iteration": 2.8503501415252686
    },
    {
      "auxiliary_loss_clip": 0.01156298,
      "auxiliary_loss_mlp": 0.00762228,
      "balance_loss_clip": 1.04826665,
      "balance_loss_mlp": 1.00054336,
      "epoch": 0.8432633920519449,
      "flos": 25812374808960.0,
      "grad_norm": 2.343755623945235,
      "language_loss": 0.82123411,
      "learning_rate": 2.5207558309709865e-07,
      "loss": 0.84041941,
      "num_input_tokens_seen": 151503700,
      "step": 7013,
      "time_per_iteration": 2.696959972381592
    },
    {
      "auxiliary_loss_clip": 0.01034672,
      "auxiliary_loss_mlp": 0.01001564,
      "balance_loss_clip": 1.00977421,
      "balance_loss_mlp": 1.00049675,
      "epoch": 0.8433836349425841,
      "flos": 64959531592320.0,
      "grad_norm": 0.6602378651806871,
      "language_loss": 0.56289887,
      "learning_rate": 2.516971402273065e-07,
      "loss": 0.58326125,
      "num_input_tokens_seen": 151569765,
      "step": 7014,
      "time_per_iteration": 3.2409770488739014
    },
    {
      "auxiliary_loss_clip": 0.01134345,
      "auxiliary_loss_mlp": 0.01023466,
      "balance_loss_clip": 1.04370284,
      "balance_loss_mlp": 1.01593781,
      "epoch": 0.8435038778332231,
      "flos": 20229989483520.0,
      "grad_norm": 2.187355874614902,
      "language_loss": 0.6782499,
      "learning_rate": 2.513189625736687e-07,
      "loss": 0.69982803,
      "num_input_tokens_seen": 151586660,
      "step": 7015,
      "time_per_iteration": 3.7075695991516113
    },
    {
      "auxiliary_loss_clip": 0.01117427,
      "auxiliary_loss_mlp": 0.01025843,
      "balance_loss_clip": 1.04474211,
      "balance_loss_mlp": 1.01792133,
      "epoch": 0.8436241207238622,
      "flos": 20992229020800.0,
      "grad_norm": 2.590441609671895,
      "language_loss": 0.71466613,
      "learning_rate": 2.509410501935534e-07,
      "loss": 0.73609889,
      "num_input_tokens_seen": 151602295,
      "step": 7016,
      "time_per_iteration": 3.6497926712036133
    },
    {
      "auxiliary_loss_clip": 0.01135097,
      "auxiliary_loss_mlp": 0.01032365,
      "balance_loss_clip": 1.04877567,
      "balance_loss_mlp": 1.02403259,
      "epoch": 0.8437443636145013,
      "flos": 14682257804160.0,
      "grad_norm": 2.5960548763365607,
      "language_loss": 0.75469196,
      "learning_rate": 2.5056340314429116e-07,
      "loss": 0.77636659,
      "num_input_tokens_seen": 151619760,
      "step": 7017,
      "time_per_iteration": 2.7218258380889893
    },
    {
      "auxiliary_loss_clip": 0.01122889,
      "auxiliary_loss_mlp": 0.00762943,
      "balance_loss_clip": 1.04007304,
      "balance_loss_mlp": 1.00041246,
      "epoch": 0.8438646065051404,
      "flos": 21608814908160.0,
      "grad_norm": 17.372785203469782,
      "language_loss": 0.80792582,
      "learning_rate": 2.5018602148316904e-07,
      "loss": 0.82678413,
      "num_input_tokens_seen": 151635795,
      "step": 7018,
      "time_per_iteration": 2.728853940963745
    },
    {
      "auxiliary_loss_clip": 0.01127278,
      "auxiliary_loss_mlp": 0.01022663,
      "balance_loss_clip": 1.04391038,
      "balance_loss_mlp": 1.01498604,
      "epoch": 0.8439848493957794,
      "flos": 23289937194240.0,
      "grad_norm": 2.1305010686948753,
      "language_loss": 0.80564493,
      "learning_rate": 2.498089052674359e-07,
      "loss": 0.82714438,
      "num_input_tokens_seen": 151653770,
      "step": 7019,
      "time_per_iteration": 2.671649932861328
    },
    {
      "auxiliary_loss_clip": 0.01173712,
      "auxiliary_loss_mlp": 0.00762499,
      "balance_loss_clip": 1.0503732,
      "balance_loss_mlp": 1.00050771,
      "epoch": 0.8441050922864186,
      "flos": 19719339782400.0,
      "grad_norm": 2.035248304785061,
      "language_loss": 0.75348938,
      "learning_rate": 2.494320545543007e-07,
      "loss": 0.77285147,
      "num_input_tokens_seen": 151673340,
      "step": 7020,
      "time_per_iteration": 2.6615421772003174
    },
    {
      "auxiliary_loss_clip": 0.0113541,
      "auxiliary_loss_mlp": 0.01026906,
      "balance_loss_clip": 1.04671121,
      "balance_loss_mlp": 1.01910949,
      "epoch": 0.8442253351770577,
      "flos": 21835268202240.0,
      "grad_norm": 1.7801764216432934,
      "language_loss": 0.6645546,
      "learning_rate": 2.490554694009308e-07,
      "loss": 0.68617773,
      "num_input_tokens_seen": 151694205,
      "step": 7021,
      "time_per_iteration": 2.747905731201172
    },
    {
      "auxiliary_loss_clip": 0.01056765,
      "auxiliary_loss_mlp": 0.01023454,
      "balance_loss_clip": 1.03175294,
      "balance_loss_mlp": 1.01549697,
      "epoch": 0.8443455780676967,
      "flos": 34346365447680.0,
      "grad_norm": 1.9473394366380712,
      "language_loss": 0.78436708,
      "learning_rate": 2.4867914986445426e-07,
      "loss": 0.80516922,
      "num_input_tokens_seen": 151716595,
      "step": 7022,
      "time_per_iteration": 3.2179579734802246
    },
    {
      "auxiliary_loss_clip": 0.01056378,
      "auxiliary_loss_mlp": 0.01026787,
      "balance_loss_clip": 1.03467906,
      "balance_loss_mlp": 1.01980424,
      "epoch": 0.8444658209583359,
      "flos": 48214599281280.0,
      "grad_norm": 2.07400742430747,
      "language_loss": 0.71024519,
      "learning_rate": 2.483030960019581e-07,
      "loss": 0.73107684,
      "num_input_tokens_seen": 151740525,
      "step": 7023,
      "time_per_iteration": 3.569892168045044
    },
    {
      "auxiliary_loss_clip": 0.01054833,
      "auxiliary_loss_mlp": 0.01001069,
      "balance_loss_clip": 1.01047671,
      "balance_loss_mlp": 1.00001359,
      "epoch": 0.8445860638489749,
      "flos": 68484773105280.0,
      "grad_norm": 0.7415295948014673,
      "language_loss": 0.55431831,
      "learning_rate": 2.479273078704891e-07,
      "loss": 0.57487732,
      "num_input_tokens_seen": 151793890,
      "step": 7024,
      "time_per_iteration": 4.153167486190796
    },
    {
      "auxiliary_loss_clip": 0.0104404,
      "auxiliary_loss_mlp": 0.01000872,
      "balance_loss_clip": 1.0093627,
      "balance_loss_mlp": 0.99982893,
      "epoch": 0.844706306739614,
      "flos": 62833331882880.0,
      "grad_norm": 0.7833108412334198,
      "language_loss": 0.64670825,
      "learning_rate": 2.475517855270552e-07,
      "loss": 0.66715729,
      "num_input_tokens_seen": 151853970,
      "step": 7025,
      "time_per_iteration": 3.2593564987182617
    },
    {
      "auxiliary_loss_clip": 0.010788,
      "auxiliary_loss_mlp": 0.01023182,
      "balance_loss_clip": 1.03812695,
      "balance_loss_mlp": 1.01591301,
      "epoch": 0.8448265496302532,
      "flos": 14976114969600.0,
      "grad_norm": 1.9765416201797832,
      "language_loss": 0.7256223,
      "learning_rate": 2.4717652902862143e-07,
      "loss": 0.74664205,
      "num_input_tokens_seen": 151872945,
      "step": 7026,
      "time_per_iteration": 3.812650203704834
    },
    {
      "auxiliary_loss_clip": 0.01126438,
      "auxiliary_loss_mlp": 0.01024872,
      "balance_loss_clip": 1.03985739,
      "balance_loss_mlp": 1.01716566,
      "epoch": 0.8449467925208922,
      "flos": 23441265192960.0,
      "grad_norm": 1.9806373146152976,
      "language_loss": 0.81270063,
      "learning_rate": 2.4680153843211495e-07,
      "loss": 0.83421373,
      "num_input_tokens_seen": 151892875,
      "step": 7027,
      "time_per_iteration": 2.8935821056365967
    },
    {
      "auxiliary_loss_clip": 0.01109378,
      "auxiliary_loss_mlp": 0.01023167,
      "balance_loss_clip": 1.04283834,
      "balance_loss_mlp": 1.0150727,
      "epoch": 0.8450670354115313,
      "flos": 22748045639040.0,
      "grad_norm": 1.9461898931856023,
      "language_loss": 0.72719127,
      "learning_rate": 2.464268137944212e-07,
      "loss": 0.74851674,
      "num_input_tokens_seen": 151914170,
      "step": 7028,
      "time_per_iteration": 2.7998273372650146
    },
    {
      "auxiliary_loss_clip": 0.01120797,
      "auxiliary_loss_mlp": 0.01024858,
      "balance_loss_clip": 1.04298425,
      "balance_loss_mlp": 1.01662922,
      "epoch": 0.8451872783021703,
      "flos": 29825571605760.0,
      "grad_norm": 1.940014766194017,
      "language_loss": 0.78133249,
      "learning_rate": 2.46052355172385e-07,
      "loss": 0.80278909,
      "num_input_tokens_seen": 151932210,
      "step": 7029,
      "time_per_iteration": 2.8070755004882812
    },
    {
      "auxiliary_loss_clip": 0.01094999,
      "auxiliary_loss_mlp": 0.01023631,
      "balance_loss_clip": 1.04066205,
      "balance_loss_mlp": 1.01607275,
      "epoch": 0.8453075211928095,
      "flos": 21870029589120.0,
      "grad_norm": 1.8502033849543955,
      "language_loss": 0.74430799,
      "learning_rate": 2.456781626228128e-07,
      "loss": 0.76549435,
      "num_input_tokens_seen": 151951715,
      "step": 7030,
      "time_per_iteration": 2.7586758136749268
    },
    {
      "auxiliary_loss_clip": 0.01051352,
      "auxiliary_loss_mlp": 0.01002424,
      "balance_loss_clip": 1.01182723,
      "balance_loss_mlp": 1.00143456,
      "epoch": 0.8454277640834486,
      "flos": 58751869288320.0,
      "grad_norm": 0.9219862605651147,
      "language_loss": 0.6627233,
      "learning_rate": 2.453042362024675e-07,
      "loss": 0.68326116,
      "num_input_tokens_seen": 152004960,
      "step": 7031,
      "time_per_iteration": 3.28208589553833
    },
    {
      "auxiliary_loss_clip": 0.0107494,
      "auxiliary_loss_mlp": 0.01025072,
      "balance_loss_clip": 1.03660798,
      "balance_loss_mlp": 1.01770544,
      "epoch": 0.8455480069740876,
      "flos": 27090076469760.0,
      "grad_norm": 1.6660448528919363,
      "language_loss": 0.73462605,
      "learning_rate": 2.449305759680751e-07,
      "loss": 0.7556262,
      "num_input_tokens_seen": 152026285,
      "step": 7032,
      "time_per_iteration": 3.105490207672119
    },
    {
      "auxiliary_loss_clip": 0.01121134,
      "auxiliary_loss_mlp": 0.00762237,
      "balance_loss_clip": 1.04486024,
      "balance_loss_mlp": 1.0004859,
      "epoch": 0.8456682498647268,
      "flos": 27198670262400.0,
      "grad_norm": 1.719984416117503,
      "language_loss": 0.75494325,
      "learning_rate": 2.445571819763188e-07,
      "loss": 0.77377695,
      "num_input_tokens_seen": 152048585,
      "step": 7033,
      "time_per_iteration": 3.095456600189209
    },
    {
      "auxiliary_loss_clip": 0.01120192,
      "auxiliary_loss_mlp": 0.00762106,
      "balance_loss_clip": 1.04385483,
      "balance_loss_mlp": 1.00047946,
      "epoch": 0.8457884927553658,
      "flos": 20631901737600.0,
      "grad_norm": 2.0949804742648523,
      "language_loss": 0.5845589,
      "learning_rate": 2.4418405428384227e-07,
      "loss": 0.60338187,
      "num_input_tokens_seen": 152068795,
      "step": 7034,
      "time_per_iteration": 2.8066041469573975
    },
    {
      "auxiliary_loss_clip": 0.01110077,
      "auxiliary_loss_mlp": 0.00762609,
      "balance_loss_clip": 1.0372771,
      "balance_loss_mlp": 1.0003978,
      "epoch": 0.8459087356460049,
      "flos": 15299023259520.0,
      "grad_norm": 1.7807453346250042,
      "language_loss": 0.71735764,
      "learning_rate": 2.4381119294724864e-07,
      "loss": 0.73608446,
      "num_input_tokens_seen": 152086240,
      "step": 7035,
      "time_per_iteration": 2.750966787338257
    },
    {
      "auxiliary_loss_clip": 0.01125848,
      "auxiliary_loss_mlp": 0.01027955,
      "balance_loss_clip": 1.04219663,
      "balance_loss_mlp": 1.0201143,
      "epoch": 0.846028978536644,
      "flos": 18843155326080.0,
      "grad_norm": 3.4516324361524564,
      "language_loss": 0.53627563,
      "learning_rate": 2.434385980231004e-07,
      "loss": 0.55781364,
      "num_input_tokens_seen": 152105080,
      "step": 7036,
      "time_per_iteration": 2.781902313232422
    },
    {
      "auxiliary_loss_clip": 0.01137719,
      "auxiliary_loss_mlp": 0.01027305,
      "balance_loss_clip": 1.04199517,
      "balance_loss_mlp": 1.01982808,
      "epoch": 0.8461492214272831,
      "flos": 52661740285440.0,
      "grad_norm": 1.7416165711777198,
      "language_loss": 0.65773779,
      "learning_rate": 2.4306626956792043e-07,
      "loss": 0.67938805,
      "num_input_tokens_seen": 152130025,
      "step": 7037,
      "time_per_iteration": 2.9651827812194824
    },
    {
      "auxiliary_loss_clip": 0.01075659,
      "auxiliary_loss_mlp": 0.01029076,
      "balance_loss_clip": 1.03264117,
      "balance_loss_mlp": 1.02175677,
      "epoch": 0.8462694643179222,
      "flos": 18588405093120.0,
      "grad_norm": 1.7299936420512039,
      "language_loss": 0.75545448,
      "learning_rate": 2.4269420763819017e-07,
      "loss": 0.77650177,
      "num_input_tokens_seen": 152148070,
      "step": 7038,
      "time_per_iteration": 2.9293124675750732
    },
    {
      "auxiliary_loss_clip": 0.0115294,
      "auxiliary_loss_mlp": 0.01023853,
      "balance_loss_clip": 1.04285049,
      "balance_loss_mlp": 1.01693583,
      "epoch": 0.8463897072085613,
      "flos": 24387080163840.0,
      "grad_norm": 3.404986648271063,
      "language_loss": 0.83379966,
      "learning_rate": 2.4232241229035223e-07,
      "loss": 0.85556757,
      "num_input_tokens_seen": 152165825,
      "step": 7039,
      "time_per_iteration": 3.412548542022705
    },
    {
      "auxiliary_loss_clip": 0.01042792,
      "auxiliary_loss_mlp": 0.01001539,
      "balance_loss_clip": 1.01073778,
      "balance_loss_mlp": 1.00052547,
      "epoch": 0.8465099500992004,
      "flos": 68702140258560.0,
      "grad_norm": 0.7529789348621967,
      "language_loss": 0.56718946,
      "learning_rate": 2.419508835808064e-07,
      "loss": 0.58763278,
      "num_input_tokens_seen": 152222380,
      "step": 7040,
      "time_per_iteration": 3.1991019248962402
    },
    {
      "auxiliary_loss_clip": 0.01156566,
      "auxiliary_loss_mlp": 0.01023857,
      "balance_loss_clip": 1.04580832,
      "balance_loss_mlp": 1.01629031,
      "epoch": 0.8466301929898394,
      "flos": 13735724561280.0,
      "grad_norm": 2.5614416386721715,
      "language_loss": 0.63132811,
      "learning_rate": 2.415796215659134e-07,
      "loss": 0.65313232,
      "num_input_tokens_seen": 152239085,
      "step": 7041,
      "time_per_iteration": 4.028004884719849
    },
    {
      "auxiliary_loss_clip": 0.01112004,
      "auxiliary_loss_mlp": 0.01029034,
      "balance_loss_clip": 1.04285598,
      "balance_loss_mlp": 1.02195573,
      "epoch": 0.8467504358804786,
      "flos": 19241260738560.0,
      "grad_norm": 2.102970676108836,
      "language_loss": 0.77036232,
      "learning_rate": 2.412086263019939e-07,
      "loss": 0.79177266,
      "num_input_tokens_seen": 152257110,
      "step": 7042,
      "time_per_iteration": 4.533121824264526
    },
    {
      "auxiliary_loss_clip": 0.01148549,
      "auxiliary_loss_mlp": 0.01026688,
      "balance_loss_clip": 1.04679,
      "balance_loss_mlp": 1.01958859,
      "epoch": 0.8468706787711177,
      "flos": 21324115710720.0,
      "grad_norm": 1.5848400067874553,
      "language_loss": 0.80073696,
      "learning_rate": 2.408378978453276e-07,
      "loss": 0.82248932,
      "num_input_tokens_seen": 152277230,
      "step": 7043,
      "time_per_iteration": 2.676647424697876
    },
    {
      "auxiliary_loss_clip": 0.01023838,
      "auxiliary_loss_mlp": 0.01002898,
      "balance_loss_clip": 1.01024842,
      "balance_loss_mlp": 1.00176513,
      "epoch": 0.8469909216617567,
      "flos": 64877439058560.0,
      "grad_norm": 0.9214418826296299,
      "language_loss": 0.63964385,
      "learning_rate": 2.404674362521533e-07,
      "loss": 0.65991122,
      "num_input_tokens_seen": 152335725,
      "step": 7044,
      "time_per_iteration": 3.4396214485168457
    },
    {
      "auxiliary_loss_clip": 0.01132779,
      "auxiliary_loss_mlp": 0.01024779,
      "balance_loss_clip": 1.04236281,
      "balance_loss_mlp": 1.01707554,
      "epoch": 0.8471111645523959,
      "flos": 19280583152640.0,
      "grad_norm": 2.412005821565708,
      "language_loss": 0.74765027,
      "learning_rate": 2.4009724157866997e-07,
      "loss": 0.76922584,
      "num_input_tokens_seen": 152352785,
      "step": 7045,
      "time_per_iteration": 3.1656112670898438
    },
    {
      "auxiliary_loss_clip": 0.01068237,
      "auxiliary_loss_mlp": 0.00761503,
      "balance_loss_clip": 1.03528714,
      "balance_loss_mlp": 1.00050402,
      "epoch": 0.8472314074430349,
      "flos": 22015826893440.0,
      "grad_norm": 1.8638713718720648,
      "language_loss": 0.76528722,
      "learning_rate": 2.3972731388103564e-07,
      "loss": 0.78358459,
      "num_input_tokens_seen": 152371265,
      "step": 7046,
      "time_per_iteration": 2.8656792640686035
    },
    {
      "auxiliary_loss_clip": 0.01042166,
      "auxiliary_loss_mlp": 0.01001428,
      "balance_loss_clip": 1.01195049,
      "balance_loss_mlp": 1.00038457,
      "epoch": 0.847351650333674,
      "flos": 57882580243200.0,
      "grad_norm": 0.8061949062162568,
      "language_loss": 0.62330729,
      "learning_rate": 2.393576532153687e-07,
      "loss": 0.64374322,
      "num_input_tokens_seen": 152435050,
      "step": 7047,
      "time_per_iteration": 3.350454807281494
    },
    {
      "auxiliary_loss_clip": 0.01045542,
      "auxiliary_loss_mlp": 0.01001984,
      "balance_loss_clip": 1.01473963,
      "balance_loss_mlp": 1.00103652,
      "epoch": 0.8474718932243132,
      "flos": 41284238313600.0,
      "grad_norm": 0.9351776800333733,
      "language_loss": 0.57743627,
      "learning_rate": 2.389882596377453e-07,
      "loss": 0.59791154,
      "num_input_tokens_seen": 152489315,
      "step": 7048,
      "time_per_iteration": 3.1795406341552734
    },
    {
      "auxiliary_loss_clip": 0.01139111,
      "auxiliary_loss_mlp": 0.01024448,
      "balance_loss_clip": 1.04330206,
      "balance_loss_mlp": 1.01679802,
      "epoch": 0.8475921361149522,
      "flos": 38180906974080.0,
      "grad_norm": 2.15740849283187,
      "language_loss": 0.76335776,
      "learning_rate": 2.386191332042031e-07,
      "loss": 0.78499335,
      "num_input_tokens_seen": 152511210,
      "step": 7049,
      "time_per_iteration": 2.806727170944214
    },
    {
      "auxiliary_loss_clip": 0.01130112,
      "auxiliary_loss_mlp": 0.01031239,
      "balance_loss_clip": 1.04551184,
      "balance_loss_mlp": 1.02308476,
      "epoch": 0.8477123790055913,
      "flos": 25375054723200.0,
      "grad_norm": 2.070071145744234,
      "language_loss": 0.73104334,
      "learning_rate": 2.3825027397073794e-07,
      "loss": 0.75265682,
      "num_input_tokens_seen": 152531685,
      "step": 7050,
      "time_per_iteration": 2.77992844581604
    },
    {
      "auxiliary_loss_clip": 0.0110465,
      "auxiliary_loss_mlp": 0.01025598,
      "balance_loss_clip": 1.0393455,
      "balance_loss_mlp": 1.01741397,
      "epoch": 0.8478326218962304,
      "flos": 30225185389440.0,
      "grad_norm": 2.080465398258318,
      "language_loss": 0.66446841,
      "learning_rate": 2.3788168199330515e-07,
      "loss": 0.68577087,
      "num_input_tokens_seen": 152553245,
      "step": 7051,
      "time_per_iteration": 2.8211474418640137
    },
    {
      "auxiliary_loss_clip": 0.01106083,
      "auxiliary_loss_mlp": 0.01025121,
      "balance_loss_clip": 1.03993487,
      "balance_loss_mlp": 1.01772761,
      "epoch": 0.8479528647868695,
      "flos": 38213800853760.0,
      "grad_norm": 12.764362619479567,
      "language_loss": 0.72457731,
      "learning_rate": 2.3751335732782074e-07,
      "loss": 0.74588937,
      "num_input_tokens_seen": 152574505,
      "step": 7052,
      "time_per_iteration": 4.125114440917969
    },
    {
      "auxiliary_loss_clip": 0.01111633,
      "auxiliary_loss_mlp": 0.00762491,
      "balance_loss_clip": 1.03973794,
      "balance_loss_mlp": 1.00044763,
      "epoch": 0.8480731076775085,
      "flos": 20957790856320.0,
      "grad_norm": 1.7892131126017412,
      "language_loss": 0.79226017,
      "learning_rate": 2.371453000301582e-07,
      "loss": 0.81100142,
      "num_input_tokens_seen": 152593190,
      "step": 7053,
      "time_per_iteration": 2.8466436862945557
    },
    {
      "auxiliary_loss_clip": 0.01031038,
      "auxiliary_loss_mlp": 0.01023985,
      "balance_loss_clip": 1.02823186,
      "balance_loss_mlp": 1.01659727,
      "epoch": 0.8481933505681477,
      "flos": 32596510487040.0,
      "grad_norm": 2.7547633654253394,
      "language_loss": 0.74236822,
      "learning_rate": 2.3677751015615222e-07,
      "loss": 0.76291847,
      "num_input_tokens_seen": 152615265,
      "step": 7054,
      "time_per_iteration": 3.2606053352355957
    },
    {
      "auxiliary_loss_clip": 0.01075249,
      "auxiliary_loss_mlp": 0.01030289,
      "balance_loss_clip": 1.04310465,
      "balance_loss_mlp": 1.02161074,
      "epoch": 0.8483135934587868,
      "flos": 20741177888640.0,
      "grad_norm": 2.4771983488089933,
      "language_loss": 0.85356945,
      "learning_rate": 2.3640998776159593e-07,
      "loss": 0.87462491,
      "num_input_tokens_seen": 152632770,
      "step": 7055,
      "time_per_iteration": 3.2983033657073975
    },
    {
      "auxiliary_loss_clip": 0.01087241,
      "auxiliary_loss_mlp": 0.0102136,
      "balance_loss_clip": 1.04237664,
      "balance_loss_mlp": 1.01369512,
      "epoch": 0.8484338363494258,
      "flos": 21653057485440.0,
      "grad_norm": 1.7341276786327577,
      "language_loss": 0.81199592,
      "learning_rate": 2.3604273290224253e-07,
      "loss": 0.83308196,
      "num_input_tokens_seen": 152653485,
      "step": 7056,
      "time_per_iteration": 2.8251891136169434
    },
    {
      "auxiliary_loss_clip": 0.01146895,
      "auxiliary_loss_mlp": 0.01033385,
      "balance_loss_clip": 1.04396641,
      "balance_loss_mlp": 1.02436733,
      "epoch": 0.848554079240065,
      "flos": 15013964926080.0,
      "grad_norm": 2.0451724385534757,
      "language_loss": 0.74689758,
      "learning_rate": 2.356757456338039e-07,
      "loss": 0.76870036,
      "num_input_tokens_seen": 152670970,
      "step": 7057,
      "time_per_iteration": 2.6847076416015625
    },
    {
      "auxiliary_loss_clip": 0.0103451,
      "auxiliary_loss_mlp": 0.01013088,
      "balance_loss_clip": 1.01904488,
      "balance_loss_mlp": 1.01210427,
      "epoch": 0.848674322130704,
      "flos": 68060453742720.0,
      "grad_norm": 0.7535989256605038,
      "language_loss": 0.59009725,
      "learning_rate": 2.3530902601195147e-07,
      "loss": 0.61057323,
      "num_input_tokens_seen": 152739460,
      "step": 7058,
      "time_per_iteration": 3.3909647464752197
    },
    {
      "auxiliary_loss_clip": 0.0111631,
      "auxiliary_loss_mlp": 0.01025818,
      "balance_loss_clip": 1.04289055,
      "balance_loss_mlp": 1.01699638,
      "epoch": 0.8487945650213431,
      "flos": 18475788977280.0,
      "grad_norm": 5.967162838189481,
      "language_loss": 0.78372592,
      "learning_rate": 2.34942574092317e-07,
      "loss": 0.80514717,
      "num_input_tokens_seen": 152754710,
      "step": 7059,
      "time_per_iteration": 2.797430992126465
    },
    {
      "auxiliary_loss_clip": 0.01103098,
      "auxiliary_loss_mlp": 0.0102421,
      "balance_loss_clip": 1.04175019,
      "balance_loss_mlp": 1.0167594,
      "epoch": 0.8489148079119821,
      "flos": 23473189405440.0,
      "grad_norm": 2.1909712653740536,
      "language_loss": 0.76666343,
      "learning_rate": 2.3457638993049045e-07,
      "loss": 0.78793645,
      "num_input_tokens_seen": 152772700,
      "step": 7060,
      "time_per_iteration": 2.77506685256958
    },
    {
      "auxiliary_loss_clip": 0.01106014,
      "auxiliary_loss_mlp": 0.00762729,
      "balance_loss_clip": 1.04030848,
      "balance_loss_mlp": 1.0005486,
      "epoch": 0.8490350508026213,
      "flos": 19937604775680.0,
      "grad_norm": 2.349170820110265,
      "language_loss": 0.6431582,
      "learning_rate": 2.3421047358202252e-07,
      "loss": 0.66184556,
      "num_input_tokens_seen": 152791550,
      "step": 7061,
      "time_per_iteration": 2.7875046730041504
    },
    {
      "auxiliary_loss_clip": 0.01086432,
      "auxiliary_loss_mlp": 0.01024648,
      "balance_loss_clip": 1.0407095,
      "balance_loss_mlp": 1.01680398,
      "epoch": 0.8491552936932604,
      "flos": 24279958828800.0,
      "grad_norm": 3.0711525448234194,
      "language_loss": 0.83823854,
      "learning_rate": 2.3384482510242144e-07,
      "loss": 0.85934937,
      "num_input_tokens_seen": 152809410,
      "step": 7062,
      "time_per_iteration": 2.7558720111846924
    },
    {
      "auxiliary_loss_clip": 0.01124195,
      "auxiliary_loss_mlp": 0.01025989,
      "balance_loss_clip": 1.04256964,
      "balance_loss_mlp": 1.01810348,
      "epoch": 0.8492755365838994,
      "flos": 22522526098560.0,
      "grad_norm": 1.9961328118276123,
      "language_loss": 0.77136582,
      "learning_rate": 2.3347944454715575e-07,
      "loss": 0.7928676,
      "num_input_tokens_seen": 152825800,
      "step": 7063,
      "time_per_iteration": 2.912708282470703
    },
    {
      "auxiliary_loss_clip": 0.01144674,
      "auxiliary_loss_mlp": 0.01024928,
      "balance_loss_clip": 1.04363418,
      "balance_loss_mlp": 1.01683402,
      "epoch": 0.8493957794745386,
      "flos": 26980441182720.0,
      "grad_norm": 2.155359798493835,
      "language_loss": 0.67477185,
      "learning_rate": 2.331143319716542e-07,
      "loss": 0.69646788,
      "num_input_tokens_seen": 152845330,
      "step": 7064,
      "time_per_iteration": 2.765841007232666
    },
    {
      "auxiliary_loss_clip": 0.01128439,
      "auxiliary_loss_mlp": 0.00762855,
      "balance_loss_clip": 1.04094243,
      "balance_loss_mlp": 1.00055301,
      "epoch": 0.8495160223651776,
      "flos": 29861985018240.0,
      "grad_norm": 2.055141139674627,
      "language_loss": 0.65949106,
      "learning_rate": 2.3274948743130363e-07,
      "loss": 0.67840397,
      "num_input_tokens_seen": 152865165,
      "step": 7065,
      "time_per_iteration": 2.7552998065948486
    },
    {
      "auxiliary_loss_clip": 0.01151047,
      "auxiliary_loss_mlp": 0.01024226,
      "balance_loss_clip": 1.04344606,
      "balance_loss_mlp": 1.01683521,
      "epoch": 0.8496362652558167,
      "flos": 23075443128960.0,
      "grad_norm": 2.578816852640133,
      "language_loss": 0.79561311,
      "learning_rate": 2.3238491098145085e-07,
      "loss": 0.81736588,
      "num_input_tokens_seen": 152884695,
      "step": 7066,
      "time_per_iteration": 4.1343748569488525
    },
    {
      "auxiliary_loss_clip": 0.01130266,
      "auxiliary_loss_mlp": 0.01025062,
      "balance_loss_clip": 1.04154408,
      "balance_loss_mlp": 1.01708078,
      "epoch": 0.8497565081464559,
      "flos": 14609107756800.0,
      "grad_norm": 2.2604287131741927,
      "language_loss": 0.73211968,
      "learning_rate": 2.3202060267740141e-07,
      "loss": 0.75367296,
      "num_input_tokens_seen": 152902220,
      "step": 7067,
      "time_per_iteration": 4.801274061203003
    },
    {
      "auxiliary_loss_clip": 0.01167017,
      "auxiliary_loss_mlp": 0.01020588,
      "balance_loss_clip": 1.04515433,
      "balance_loss_mlp": 1.01283336,
      "epoch": 0.8498767510370949,
      "flos": 21136446126720.0,
      "grad_norm": 2.5316957934060778,
      "language_loss": 0.77213079,
      "learning_rate": 2.3165656257442044e-07,
      "loss": 0.79400682,
      "num_input_tokens_seen": 152920740,
      "step": 7068,
      "time_per_iteration": 2.6917028427124023
    },
    {
      "auxiliary_loss_clip": 0.01100216,
      "auxiliary_loss_mlp": 0.01026466,
      "balance_loss_clip": 1.03897476,
      "balance_loss_mlp": 1.01923597,
      "epoch": 0.849996993927734,
      "flos": 23654538195840.0,
      "grad_norm": 2.053467465514194,
      "language_loss": 0.90157551,
      "learning_rate": 2.31292790727734e-07,
      "loss": 0.92284238,
      "num_input_tokens_seen": 152938305,
      "step": 7069,
      "time_per_iteration": 2.8365049362182617
    },
    {
      "auxiliary_loss_clip": 0.01110303,
      "auxiliary_loss_mlp": 0.01018718,
      "balance_loss_clip": 1.03771901,
      "balance_loss_mlp": 1.01114857,
      "epoch": 0.8501172368183731,
      "flos": 20558069331840.0,
      "grad_norm": 2.2393381586013335,
      "language_loss": 0.8060323,
      "learning_rate": 2.3092928719252392e-07,
      "loss": 0.82732248,
      "num_input_tokens_seen": 152956705,
      "step": 7070,
      "time_per_iteration": 2.6987266540527344
    },
    {
      "auxiliary_loss_clip": 0.01146668,
      "auxiliary_loss_mlp": 0.01024602,
      "balance_loss_clip": 1.04746056,
      "balance_loss_mlp": 1.0169667,
      "epoch": 0.8502374797090122,
      "flos": 22272624201600.0,
      "grad_norm": 2.2436809350861564,
      "language_loss": 0.78740293,
      "learning_rate": 2.3056605202393475e-07,
      "loss": 0.80911565,
      "num_input_tokens_seen": 152974265,
      "step": 7071,
      "time_per_iteration": 2.694632053375244
    },
    {
      "auxiliary_loss_clip": 0.01095451,
      "auxiliary_loss_mlp": 0.0102704,
      "balance_loss_clip": 1.04147601,
      "balance_loss_mlp": 1.01880884,
      "epoch": 0.8503577225996513,
      "flos": 23659817495040.0,
      "grad_norm": 1.767913858860906,
      "language_loss": 0.666987,
      "learning_rate": 2.3020308527706888e-07,
      "loss": 0.68821192,
      "num_input_tokens_seen": 152993680,
      "step": 7072,
      "time_per_iteration": 2.7912893295288086
    },
    {
      "auxiliary_loss_clip": 0.01116426,
      "auxiliary_loss_mlp": 0.00761562,
      "balance_loss_clip": 1.04327393,
      "balance_loss_mlp": 1.00051963,
      "epoch": 0.8504779654902904,
      "flos": 26758513002240.0,
      "grad_norm": 1.8030272044538453,
      "language_loss": 0.88957274,
      "learning_rate": 2.2984038700698715e-07,
      "loss": 0.90835261,
      "num_input_tokens_seen": 153012990,
      "step": 7073,
      "time_per_iteration": 2.8337831497192383
    },
    {
      "auxiliary_loss_clip": 0.01149845,
      "auxiliary_loss_mlp": 0.01025623,
      "balance_loss_clip": 1.04749727,
      "balance_loss_mlp": 1.01771355,
      "epoch": 0.8505982083809295,
      "flos": 26468247196800.0,
      "grad_norm": 1.7508750202209968,
      "language_loss": 0.79286504,
      "learning_rate": 2.2947795726871222e-07,
      "loss": 0.81461978,
      "num_input_tokens_seen": 153034015,
      "step": 7074,
      "time_per_iteration": 2.7790212631225586
    },
    {
      "auxiliary_loss_clip": 0.01127477,
      "auxiliary_loss_mlp": 0.01027768,
      "balance_loss_clip": 1.04696822,
      "balance_loss_mlp": 1.02023959,
      "epoch": 0.8507184512715685,
      "flos": 20303390926080.0,
      "grad_norm": 1.773059250270634,
      "language_loss": 0.85733569,
      "learning_rate": 2.2911579611722253e-07,
      "loss": 0.87888813,
      "num_input_tokens_seen": 153053160,
      "step": 7075,
      "time_per_iteration": 2.70526385307312
    },
    {
      "auxiliary_loss_clip": 0.01098369,
      "auxiliary_loss_mlp": 0.01022115,
      "balance_loss_clip": 1.0378927,
      "balance_loss_mlp": 1.01496208,
      "epoch": 0.8508386941622077,
      "flos": 19025186474880.0,
      "grad_norm": 1.9484766392135675,
      "language_loss": 0.87343401,
      "learning_rate": 2.2875390360745905e-07,
      "loss": 0.89463884,
      "num_input_tokens_seen": 153072565,
      "step": 7076,
      "time_per_iteration": 2.8652865886688232
    },
    {
      "auxiliary_loss_clip": 0.01121181,
      "auxiliary_loss_mlp": 0.01028736,
      "balance_loss_clip": 1.04020083,
      "balance_loss_mlp": 1.01988506,
      "epoch": 0.8509589370528468,
      "flos": 16433405654400.0,
      "grad_norm": 1.7670154815595596,
      "language_loss": 0.77475321,
      "learning_rate": 2.2839227979432008e-07,
      "loss": 0.79625237,
      "num_input_tokens_seen": 153090215,
      "step": 7077,
      "time_per_iteration": 2.675318479537964
    },
    {
      "auxiliary_loss_clip": 0.01123089,
      "auxiliary_loss_mlp": 0.01025146,
      "balance_loss_clip": 1.04321635,
      "balance_loss_mlp": 1.01781511,
      "epoch": 0.8510791799434858,
      "flos": 18259714713600.0,
      "grad_norm": 1.7111386796326173,
      "language_loss": 0.85694528,
      "learning_rate": 2.2803092473266373e-07,
      "loss": 0.87842762,
      "num_input_tokens_seen": 153107740,
      "step": 7078,
      "time_per_iteration": 3.838226795196533
    },
    {
      "auxiliary_loss_clip": 0.01129979,
      "auxiliary_loss_mlp": 0.01023906,
      "balance_loss_clip": 1.04181862,
      "balance_loss_mlp": 1.01634789,
      "epoch": 0.851199422834125,
      "flos": 23441372933760.0,
      "grad_norm": 2.3567514229762296,
      "language_loss": 0.8714965,
      "learning_rate": 2.2766983847730724e-07,
      "loss": 0.89303535,
      "num_input_tokens_seen": 153127410,
      "step": 7079,
      "time_per_iteration": 2.7553303241729736
    },
    {
      "auxiliary_loss_clip": 0.01158166,
      "auxiliary_loss_mlp": 0.01033027,
      "balance_loss_clip": 1.04689813,
      "balance_loss_mlp": 1.02402949,
      "epoch": 0.851319665724764,
      "flos": 16289404030080.0,
      "grad_norm": 2.039079900382453,
      "language_loss": 0.66958332,
      "learning_rate": 2.2730902108302663e-07,
      "loss": 0.6914953,
      "num_input_tokens_seen": 153144325,
      "step": 7080,
      "time_per_iteration": 2.6421046257019043
    },
    {
      "auxiliary_loss_clip": 0.01145706,
      "auxiliary_loss_mlp": 0.01023659,
      "balance_loss_clip": 1.04313636,
      "balance_loss_mlp": 1.01593399,
      "epoch": 0.8514399086154031,
      "flos": 18989347680000.0,
      "grad_norm": 1.78682091224724,
      "language_loss": 0.68595135,
      "learning_rate": 2.269484726045583e-07,
      "loss": 0.707645,
      "num_input_tokens_seen": 153163240,
      "step": 7081,
      "time_per_iteration": 2.700322151184082
    },
    {
      "auxiliary_loss_clip": 0.01171532,
      "auxiliary_loss_mlp": 0.01023519,
      "balance_loss_clip": 1.0489918,
      "balance_loss_mlp": 1.01615751,
      "epoch": 0.8515601515060423,
      "flos": 24571194301440.0,
      "grad_norm": 1.919810101526117,
      "language_loss": 0.79364187,
      "learning_rate": 2.2658819309659672e-07,
      "loss": 0.81559241,
      "num_input_tokens_seen": 153183440,
      "step": 7082,
      "time_per_iteration": 2.6282083988189697
    },
    {
      "auxiliary_loss_clip": 0.01118365,
      "auxiliary_loss_mlp": 0.01024324,
      "balance_loss_clip": 1.04332089,
      "balance_loss_mlp": 1.01664984,
      "epoch": 0.8516803943966813,
      "flos": 19529443555200.0,
      "grad_norm": 3.14032251180781,
      "language_loss": 0.84903944,
      "learning_rate": 2.2622818261379706e-07,
      "loss": 0.87046635,
      "num_input_tokens_seen": 153200460,
      "step": 7083,
      "time_per_iteration": 2.7109148502349854
    },
    {
      "auxiliary_loss_clip": 0.0106377,
      "auxiliary_loss_mlp": 0.01024194,
      "balance_loss_clip": 1.03444147,
      "balance_loss_mlp": 1.01558137,
      "epoch": 0.8518006372873204,
      "flos": 20265792364800.0,
      "grad_norm": 2.0971896192355906,
      "language_loss": 0.75135732,
      "learning_rate": 2.2586844121077142e-07,
      "loss": 0.77223694,
      "num_input_tokens_seen": 153218970,
      "step": 7084,
      "time_per_iteration": 3.016590118408203
    },
    {
      "auxiliary_loss_clip": 0.01062967,
      "auxiliary_loss_mlp": 0.01025894,
      "balance_loss_clip": 1.03365731,
      "balance_loss_mlp": 1.0173223,
      "epoch": 0.8519208801779595,
      "flos": 24133227770880.0,
      "grad_norm": 1.906431673178754,
      "language_loss": 0.72073376,
      "learning_rate": 2.2550896894209215e-07,
      "loss": 0.74162239,
      "num_input_tokens_seen": 153238485,
      "step": 7085,
      "time_per_iteration": 3.143956422805786
    },
    {
      "auxiliary_loss_clip": 0.01012508,
      "auxiliary_loss_mlp": 0.01006942,
      "balance_loss_clip": 1.01647294,
      "balance_loss_mlp": 1.00560713,
      "epoch": 0.8520411230685986,
      "flos": 63035223252480.0,
      "grad_norm": 0.6822137393548169,
      "language_loss": 0.56582093,
      "learning_rate": 2.2514976586229184e-07,
      "loss": 0.58601546,
      "num_input_tokens_seen": 153306430,
      "step": 7086,
      "time_per_iteration": 3.4447431564331055
    },
    {
      "auxiliary_loss_clip": 0.01024614,
      "auxiliary_loss_mlp": 0.01000881,
      "balance_loss_clip": 1.00881934,
      "balance_loss_mlp": 0.99985546,
      "epoch": 0.8521613659592376,
      "flos": 65836865283840.0,
      "grad_norm": 0.7716365440357779,
      "language_loss": 0.54630721,
      "learning_rate": 2.247908320258609e-07,
      "loss": 0.56656218,
      "num_input_tokens_seen": 153366520,
      "step": 7087,
      "time_per_iteration": 3.324038028717041
    },
    {
      "auxiliary_loss_clip": 0.01113087,
      "auxiliary_loss_mlp": 0.00762946,
      "balance_loss_clip": 1.03929138,
      "balance_loss_mlp": 1.00053477,
      "epoch": 0.8522816088498768,
      "flos": 23112323418240.0,
      "grad_norm": 3.3628187580957176,
      "language_loss": 0.79731286,
      "learning_rate": 2.2443216748724914e-07,
      "loss": 0.81607318,
      "num_input_tokens_seen": 153387230,
      "step": 7088,
      "time_per_iteration": 2.826371192932129
    },
    {
      "auxiliary_loss_clip": 0.01157467,
      "auxiliary_loss_mlp": 0.01023653,
      "balance_loss_clip": 1.04837918,
      "balance_loss_mlp": 1.0158987,
      "epoch": 0.8524018517405159,
      "flos": 31758140073600.0,
      "grad_norm": 2.2188842879946704,
      "language_loss": 0.7476697,
      "learning_rate": 2.2407377230086588e-07,
      "loss": 0.76948094,
      "num_input_tokens_seen": 153409585,
      "step": 7089,
      "time_per_iteration": 3.1191864013671875
    },
    {
      "auxiliary_loss_clip": 0.01132645,
      "auxiliary_loss_mlp": 0.01026153,
      "balance_loss_clip": 1.04457128,
      "balance_loss_mlp": 1.01795697,
      "epoch": 0.8525220946311549,
      "flos": 18690318956160.0,
      "grad_norm": 1.976361033438866,
      "language_loss": 0.83928227,
      "learning_rate": 2.23715646521079e-07,
      "loss": 0.86087024,
      "num_input_tokens_seen": 153427105,
      "step": 7090,
      "time_per_iteration": 3.060793876647949
    },
    {
      "auxiliary_loss_clip": 0.01106897,
      "auxiliary_loss_mlp": 0.01026203,
      "balance_loss_clip": 1.03779471,
      "balance_loss_mlp": 1.01750624,
      "epoch": 0.852642337521794,
      "flos": 21793216354560.0,
      "grad_norm": 1.8874719198446146,
      "language_loss": 0.84104621,
      "learning_rate": 2.2335779020221724e-07,
      "loss": 0.86237723,
      "num_input_tokens_seen": 153443725,
      "step": 7091,
      "time_per_iteration": 2.8691844940185547
    },
    {
      "auxiliary_loss_clip": 0.01045383,
      "auxiliary_loss_mlp": 0.0100005,
      "balance_loss_clip": 1.00957942,
      "balance_loss_mlp": 0.99901932,
      "epoch": 0.8527625804124331,
      "flos": 69040132260480.0,
      "grad_norm": 0.7983970588529856,
      "language_loss": 0.56428677,
      "learning_rate": 2.2300020339856497e-07,
      "loss": 0.58474112,
      "num_input_tokens_seen": 153506410,
      "step": 7092,
      "time_per_iteration": 4.170027017593384
    },
    {
      "auxiliary_loss_clip": 0.01142992,
      "auxiliary_loss_mlp": 0.00762276,
      "balance_loss_clip": 1.04195428,
      "balance_loss_mlp": 1.00045598,
      "epoch": 0.8528828233030722,
      "flos": 26979399688320.0,
      "grad_norm": 2.295955110638782,
      "language_loss": 0.78197187,
      "learning_rate": 2.2264288616436966e-07,
      "loss": 0.80102456,
      "num_input_tokens_seen": 153526665,
      "step": 7093,
      "time_per_iteration": 3.624124050140381
    },
    {
      "auxiliary_loss_clip": 0.01125093,
      "auxiliary_loss_mlp": 0.00762239,
      "balance_loss_clip": 1.04284549,
      "balance_loss_mlp": 1.00042391,
      "epoch": 0.8530030661937112,
      "flos": 17487598936320.0,
      "grad_norm": 2.28728684710825,
      "language_loss": 0.7267316,
      "learning_rate": 2.222858385538351e-07,
      "loss": 0.74560493,
      "num_input_tokens_seen": 153543465,
      "step": 7094,
      "time_per_iteration": 3.6206722259521484
    },
    {
      "auxiliary_loss_clip": 0.01169157,
      "auxiliary_loss_mlp": 0.01021155,
      "balance_loss_clip": 1.04824686,
      "balance_loss_mlp": 1.01461029,
      "epoch": 0.8531233090843504,
      "flos": 22160798184960.0,
      "grad_norm": 1.7513283256646555,
      "language_loss": 0.68304062,
      "learning_rate": 2.2192906062112527e-07,
      "loss": 0.70494378,
      "num_input_tokens_seen": 153563340,
      "step": 7095,
      "time_per_iteration": 2.6508443355560303
    },
    {
      "auxiliary_loss_clip": 0.01113991,
      "auxiliary_loss_mlp": 0.00762326,
      "balance_loss_clip": 1.03877568,
      "balance_loss_mlp": 1.00053549,
      "epoch": 0.8532435519749895,
      "flos": 37635388145280.0,
      "grad_norm": 1.7729849175531214,
      "language_loss": 0.70650268,
      "learning_rate": 2.2157255242036377e-07,
      "loss": 0.72526586,
      "num_input_tokens_seen": 153587005,
      "step": 7096,
      "time_per_iteration": 2.9368255138397217
    },
    {
      "auxiliary_loss_clip": 0.01165329,
      "auxiliary_loss_mlp": 0.01022782,
      "balance_loss_clip": 1.04606462,
      "balance_loss_mlp": 1.01511991,
      "epoch": 0.8533637948656285,
      "flos": 21398163598080.0,
      "grad_norm": 3.4702194680996166,
      "language_loss": 0.7428804,
      "learning_rate": 2.2121631400563135e-07,
      "loss": 0.76476157,
      "num_input_tokens_seen": 153606835,
      "step": 7097,
      "time_per_iteration": 2.719686985015869
    },
    {
      "auxiliary_loss_clip": 0.01043455,
      "auxiliary_loss_mlp": 0.01000058,
      "balance_loss_clip": 1.00966883,
      "balance_loss_mlp": 0.99895537,
      "epoch": 0.8534840377562677,
      "flos": 53345122490880.0,
      "grad_norm": 0.7675832346783921,
      "language_loss": 0.52952909,
      "learning_rate": 2.208603454309701e-07,
      "loss": 0.54996425,
      "num_input_tokens_seen": 153664925,
      "step": 7098,
      "time_per_iteration": 3.237088203430176
    },
    {
      "auxiliary_loss_clip": 0.01159788,
      "auxiliary_loss_mlp": 0.01022007,
      "balance_loss_clip": 1.04911828,
      "balance_loss_mlp": 1.01390052,
      "epoch": 0.8536042806469067,
      "flos": 20814148368000.0,
      "grad_norm": 1.9500695222637514,
      "language_loss": 0.71163332,
      "learning_rate": 2.2050464675037994e-07,
      "loss": 0.73345131,
      "num_input_tokens_seen": 153683550,
      "step": 7099,
      "time_per_iteration": 2.6496262550354004
    },
    {
      "auxiliary_loss_clip": 0.01100038,
      "auxiliary_loss_mlp": 0.01030841,
      "balance_loss_clip": 1.04432058,
      "balance_loss_mlp": 1.02265096,
      "epoch": 0.8537245235375458,
      "flos": 24681368292480.0,
      "grad_norm": 2.083214880802895,
      "language_loss": 0.72995371,
      "learning_rate": 2.2014921801782016e-07,
      "loss": 0.75126249,
      "num_input_tokens_seen": 153703040,
      "step": 7100,
      "time_per_iteration": 2.832451105117798
    },
    {
      "auxiliary_loss_clip": 0.01119249,
      "auxiliary_loss_mlp": 0.01026388,
      "balance_loss_clip": 1.04222512,
      "balance_loss_mlp": 1.01831126,
      "epoch": 0.853844766428185,
      "flos": 24384817607040.0,
      "grad_norm": 1.8859116627048278,
      "language_loss": 0.74063963,
      "learning_rate": 2.1979405928720872e-07,
      "loss": 0.76209593,
      "num_input_tokens_seen": 153722695,
      "step": 7101,
      "time_per_iteration": 2.7667384147644043
    },
    {
      "auxiliary_loss_clip": 0.01139341,
      "auxiliary_loss_mlp": 0.01023983,
      "balance_loss_clip": 1.04663682,
      "balance_loss_mlp": 1.01683044,
      "epoch": 0.853965009318824,
      "flos": 20955707867520.0,
      "grad_norm": 1.493946304153918,
      "language_loss": 0.79558158,
      "learning_rate": 2.1943917061242257e-07,
      "loss": 0.81721479,
      "num_input_tokens_seen": 153742550,
      "step": 7102,
      "time_per_iteration": 2.7236738204956055
    },
    {
      "auxiliary_loss_clip": 0.01159475,
      "auxiliary_loss_mlp": 0.01025261,
      "balance_loss_clip": 1.04850149,
      "balance_loss_mlp": 1.01623678,
      "epoch": 0.8540852522094631,
      "flos": 24201816791040.0,
      "grad_norm": 1.9930028092119834,
      "language_loss": 0.66291434,
      "learning_rate": 2.1908455204729903e-07,
      "loss": 0.68476164,
      "num_input_tokens_seen": 153761700,
      "step": 7103,
      "time_per_iteration": 2.7193763256073
    },
    {
      "auxiliary_loss_clip": 0.01110974,
      "auxiliary_loss_mlp": 0.01022255,
      "balance_loss_clip": 1.04078591,
      "balance_loss_mlp": 1.01471519,
      "epoch": 0.8542054951001022,
      "flos": 25082921410560.0,
      "grad_norm": 2.352566867150913,
      "language_loss": 0.78146708,
      "learning_rate": 2.1873020364563265e-07,
      "loss": 0.80279934,
      "num_input_tokens_seen": 153780765,
      "step": 7104,
      "time_per_iteration": 3.8398690223693848
    },
    {
      "auxiliary_loss_clip": 0.01080194,
      "auxiliary_loss_mlp": 0.01022817,
      "balance_loss_clip": 1.03637958,
      "balance_loss_mlp": 1.01525271,
      "epoch": 0.8543257379907413,
      "flos": 24316551809280.0,
      "grad_norm": 2.273342248647202,
      "language_loss": 0.76281917,
      "learning_rate": 2.183761254611789e-07,
      "loss": 0.78384924,
      "num_input_tokens_seen": 153801090,
      "step": 7105,
      "time_per_iteration": 3.009967565536499
    },
    {
      "auxiliary_loss_clip": 0.01119021,
      "auxiliary_loss_mlp": 0.01024808,
      "balance_loss_clip": 1.04475272,
      "balance_loss_mlp": 1.01757193,
      "epoch": 0.8544459808813804,
      "flos": 55286630467200.0,
      "grad_norm": 5.2770869344369675,
      "language_loss": 0.70340097,
      "learning_rate": 2.1802231754764987e-07,
      "loss": 0.72483933,
      "num_input_tokens_seen": 153826530,
      "step": 7106,
      "time_per_iteration": 3.1273915767669678
    },
    {
      "auxiliary_loss_clip": 0.01122902,
      "auxiliary_loss_mlp": 0.01021872,
      "balance_loss_clip": 1.0429914,
      "balance_loss_mlp": 1.01343179,
      "epoch": 0.8545662237720195,
      "flos": 25776248705280.0,
      "grad_norm": 2.117036812409559,
      "language_loss": 0.76543736,
      "learning_rate": 2.17668779958718e-07,
      "loss": 0.78688508,
      "num_input_tokens_seen": 153849110,
      "step": 7107,
      "time_per_iteration": 2.799201488494873
    },
    {
      "auxiliary_loss_clip": 0.01134885,
      "auxiliary_loss_mlp": 0.01024033,
      "balance_loss_clip": 1.04266584,
      "balance_loss_mlp": 1.0162276,
      "epoch": 0.8546864666626586,
      "flos": 11108320427520.0,
      "grad_norm": 2.3494213024802613,
      "language_loss": 0.8059603,
      "learning_rate": 2.1731551274801553e-07,
      "loss": 0.82754946,
      "num_input_tokens_seen": 153865550,
      "step": 7108,
      "time_per_iteration": 2.76147723197937
    },
    {
      "auxiliary_loss_clip": 0.01111434,
      "auxiliary_loss_mlp": 0.01028982,
      "balance_loss_clip": 1.03891277,
      "balance_loss_mlp": 1.02176094,
      "epoch": 0.8548067095532976,
      "flos": 25520169669120.0,
      "grad_norm": 2.0340582412071933,
      "language_loss": 0.61432016,
      "learning_rate": 2.169625159691324e-07,
      "loss": 0.63572437,
      "num_input_tokens_seen": 153885425,
      "step": 7109,
      "time_per_iteration": 2.847628593444824
    },
    {
      "auxiliary_loss_clip": 0.01124873,
      "auxiliary_loss_mlp": 0.01027734,
      "balance_loss_clip": 1.04292369,
      "balance_loss_mlp": 1.01966393,
      "epoch": 0.8549269524439368,
      "flos": 24717853532160.0,
      "grad_norm": 2.1821210636558632,
      "language_loss": 0.74317145,
      "learning_rate": 2.1660978967561784e-07,
      "loss": 0.76469755,
      "num_input_tokens_seen": 153904760,
      "step": 7110,
      "time_per_iteration": 2.7316439151763916
    },
    {
      "auxiliary_loss_clip": 0.01092393,
      "auxiliary_loss_mlp": 0.01020971,
      "balance_loss_clip": 1.03600764,
      "balance_loss_mlp": 1.01338363,
      "epoch": 0.8550471953345758,
      "flos": 19825599191040.0,
      "grad_norm": 3.812741014235008,
      "language_loss": 0.79031205,
      "learning_rate": 2.1625733392098035e-07,
      "loss": 0.81144571,
      "num_input_tokens_seen": 153920370,
      "step": 7111,
      "time_per_iteration": 2.823763370513916
    },
    {
      "auxiliary_loss_clip": 0.0116571,
      "auxiliary_loss_mlp": 0.01020243,
      "balance_loss_clip": 1.04472423,
      "balance_loss_mlp": 1.01364517,
      "epoch": 0.8551674382252149,
      "flos": 22820441500800.0,
      "grad_norm": 2.6618854029379504,
      "language_loss": 0.79685152,
      "learning_rate": 2.159051487586867e-07,
      "loss": 0.81871104,
      "num_input_tokens_seen": 153940500,
      "step": 7112,
      "time_per_iteration": 2.7061028480529785
    },
    {
      "auxiliary_loss_clip": 0.01129416,
      "auxiliary_loss_mlp": 0.01025508,
      "balance_loss_clip": 1.04584062,
      "balance_loss_mlp": 1.0176909,
      "epoch": 0.8552876811158541,
      "flos": 20631255292800.0,
      "grad_norm": 3.1305381626562228,
      "language_loss": 0.72452778,
      "learning_rate": 2.155532342421642e-07,
      "loss": 0.74607706,
      "num_input_tokens_seen": 153958500,
      "step": 7113,
      "time_per_iteration": 2.7762112617492676
    },
    {
      "auxiliary_loss_clip": 0.0112516,
      "auxiliary_loss_mlp": 0.010331,
      "balance_loss_clip": 1.04273129,
      "balance_loss_mlp": 1.02494025,
      "epoch": 0.8554079240064931,
      "flos": 23112359331840.0,
      "grad_norm": 2.052415358824265,
      "language_loss": 0.78372395,
      "learning_rate": 2.1520159042479636e-07,
      "loss": 0.80530649,
      "num_input_tokens_seen": 153976790,
      "step": 7114,
      "time_per_iteration": 2.704119920730591
    },
    {
      "auxiliary_loss_clip": 0.01102455,
      "auxiliary_loss_mlp": 0.00761798,
      "balance_loss_clip": 1.03901637,
      "balance_loss_mlp": 1.00049043,
      "epoch": 0.8555281668971322,
      "flos": 22128047959680.0,
      "grad_norm": 2.0795638527143168,
      "language_loss": 0.70942152,
      "learning_rate": 2.148502173599287e-07,
      "loss": 0.72806406,
      "num_input_tokens_seen": 153994930,
      "step": 7115,
      "time_per_iteration": 2.720125198364258
    },
    {
      "auxiliary_loss_clip": 0.01090303,
      "auxiliary_loss_mlp": 0.00762374,
      "balance_loss_clip": 1.03832436,
      "balance_loss_mlp": 1.00045204,
      "epoch": 0.8556484097877713,
      "flos": 31139040234240.0,
      "grad_norm": 2.0330839160427794,
      "language_loss": 0.65831244,
      "learning_rate": 2.1449911510086372e-07,
      "loss": 0.67683923,
      "num_input_tokens_seen": 154014400,
      "step": 7116,
      "time_per_iteration": 2.90228533744812
    },
    {
      "auxiliary_loss_clip": 0.01086724,
      "auxiliary_loss_mlp": 0.01022617,
      "balance_loss_clip": 1.03430295,
      "balance_loss_mlp": 1.01485038,
      "epoch": 0.8557686526784104,
      "flos": 24316551809280.0,
      "grad_norm": 4.255309675331106,
      "language_loss": 0.76878315,
      "learning_rate": 2.141482837008628e-07,
      "loss": 0.78987658,
      "num_input_tokens_seen": 154034940,
      "step": 7117,
      "time_per_iteration": 3.8788859844207764
    },
    {
      "auxiliary_loss_clip": 0.01124818,
      "auxiliary_loss_mlp": 0.01028844,
      "balance_loss_clip": 1.04168248,
      "balance_loss_mlp": 1.02111626,
      "epoch": 0.8558888955690495,
      "flos": 17712723427200.0,
      "grad_norm": 2.0368257484632855,
      "language_loss": 0.72166628,
      "learning_rate": 2.1379772321314826e-07,
      "loss": 0.74320292,
      "num_input_tokens_seen": 154052985,
      "step": 7118,
      "time_per_iteration": 2.7757513523101807
    },
    {
      "auxiliary_loss_clip": 0.01113486,
      "auxiliary_loss_mlp": 0.01022573,
      "balance_loss_clip": 1.04418898,
      "balance_loss_mlp": 1.01507449,
      "epoch": 0.8560091384596886,
      "flos": 19171702051200.0,
      "grad_norm": 2.7815207779676774,
      "language_loss": 0.81775713,
      "learning_rate": 2.1344743369089802e-07,
      "loss": 0.83911765,
      "num_input_tokens_seen": 154068765,
      "step": 7119,
      "time_per_iteration": 4.7143235206604
    },
    {
      "auxiliary_loss_clip": 0.01111967,
      "auxiliary_loss_mlp": 0.01024077,
      "balance_loss_clip": 1.0432272,
      "balance_loss_mlp": 1.01679945,
      "epoch": 0.8561293813503277,
      "flos": 23914855036800.0,
      "grad_norm": 1.7265624276639726,
      "language_loss": 0.82217139,
      "learning_rate": 2.130974151872522e-07,
      "loss": 0.84353185,
      "num_input_tokens_seen": 154089100,
      "step": 7120,
      "time_per_iteration": 2.776426315307617
    },
    {
      "auxiliary_loss_clip": 0.01130731,
      "auxiliary_loss_mlp": 0.01023974,
      "balance_loss_clip": 1.04012179,
      "balance_loss_mlp": 1.01604366,
      "epoch": 0.8562496242409667,
      "flos": 22529206028160.0,
      "grad_norm": 4.022585100911627,
      "language_loss": 0.78521299,
      "learning_rate": 2.1274766775530773e-07,
      "loss": 0.80676001,
      "num_input_tokens_seen": 154108965,
      "step": 7121,
      "time_per_iteration": 2.751807451248169
    },
    {
      "auxiliary_loss_clip": 0.0113517,
      "auxiliary_loss_mlp": 0.01026024,
      "balance_loss_clip": 1.04161024,
      "balance_loss_mlp": 1.0182755,
      "epoch": 0.8563698671316058,
      "flos": 14712745472640.0,
      "grad_norm": 2.2185028170437313,
      "language_loss": 0.79394865,
      "learning_rate": 2.1239819144812077e-07,
      "loss": 0.81556058,
      "num_input_tokens_seen": 154123425,
      "step": 7122,
      "time_per_iteration": 2.635356903076172
    },
    {
      "auxiliary_loss_clip": 0.01107688,
      "auxiliary_loss_mlp": 0.01026695,
      "balance_loss_clip": 1.0409435,
      "balance_loss_mlp": 1.01887751,
      "epoch": 0.856490110022245,
      "flos": 39167768211840.0,
      "grad_norm": 1.7774753401699073,
      "language_loss": 0.70097649,
      "learning_rate": 2.1204898631870716e-07,
      "loss": 0.72232032,
      "num_input_tokens_seen": 154148315,
      "step": 7123,
      "time_per_iteration": 2.943915367126465
    },
    {
      "auxiliary_loss_clip": 0.01111835,
      "auxiliary_loss_mlp": 0.01026149,
      "balance_loss_clip": 1.04405761,
      "balance_loss_mlp": 1.01846611,
      "epoch": 0.856610352912884,
      "flos": 29059345658880.0,
      "grad_norm": 1.8567839636764578,
      "language_loss": 0.76220936,
      "learning_rate": 2.1170005242004006e-07,
      "loss": 0.78358924,
      "num_input_tokens_seen": 154169665,
      "step": 7124,
      "time_per_iteration": 2.895507335662842
    },
    {
      "auxiliary_loss_clip": 0.01110265,
      "auxiliary_loss_mlp": 0.01025976,
      "balance_loss_clip": 1.03948915,
      "balance_loss_mlp": 1.01835823,
      "epoch": 0.8567305958035231,
      "flos": 23878333883520.0,
      "grad_norm": 2.608051991612531,
      "language_loss": 0.78043467,
      "learning_rate": 2.1135138980505384e-07,
      "loss": 0.80179703,
      "num_input_tokens_seen": 154190335,
      "step": 7125,
      "time_per_iteration": 2.796104907989502
    },
    {
      "auxiliary_loss_clip": 0.01114654,
      "auxiliary_loss_mlp": 0.01023531,
      "balance_loss_clip": 1.0402801,
      "balance_loss_mlp": 1.01613402,
      "epoch": 0.8568508386941622,
      "flos": 22200120599040.0,
      "grad_norm": 2.1228648285495595,
      "language_loss": 0.72426319,
      "learning_rate": 2.110029985266395e-07,
      "loss": 0.74564499,
      "num_input_tokens_seen": 154210040,
      "step": 7126,
      "time_per_iteration": 2.7887232303619385
    },
    {
      "auxiliary_loss_clip": 0.01157608,
      "auxiliary_loss_mlp": 0.01024105,
      "balance_loss_clip": 1.0463469,
      "balance_loss_mlp": 1.01681519,
      "epoch": 0.8569710815848013,
      "flos": 17307507121920.0,
      "grad_norm": 1.9094385784219499,
      "language_loss": 0.73859006,
      "learning_rate": 2.1065487863764787e-07,
      "loss": 0.76040721,
      "num_input_tokens_seen": 154228385,
      "step": 7127,
      "time_per_iteration": 2.6924426555633545
    },
    {
      "auxiliary_loss_clip": 0.01121566,
      "auxiliary_loss_mlp": 0.01025916,
      "balance_loss_clip": 1.04143918,
      "balance_loss_mlp": 1.01848316,
      "epoch": 0.8570913244754403,
      "flos": 23732285184000.0,
      "grad_norm": 1.602278991228722,
      "language_loss": 0.85614967,
      "learning_rate": 2.1030703019088846e-07,
      "loss": 0.87762445,
      "num_input_tokens_seen": 154249015,
      "step": 7128,
      "time_per_iteration": 2.7615880966186523
    },
    {
      "auxiliary_loss_clip": 0.01141442,
      "auxiliary_loss_mlp": 0.00760833,
      "balance_loss_clip": 1.04210663,
      "balance_loss_mlp": 1.00042617,
      "epoch": 0.8572115673660795,
      "flos": 20048748433920.0,
      "grad_norm": 2.540804838448528,
      "language_loss": 0.70756954,
      "learning_rate": 2.099594532391291e-07,
      "loss": 0.7265923,
      "num_input_tokens_seen": 154267700,
      "step": 7129,
      "time_per_iteration": 3.665386915206909
    },
    {
      "auxiliary_loss_clip": 0.01097926,
      "auxiliary_loss_mlp": 0.01025431,
      "balance_loss_clip": 1.03740096,
      "balance_loss_mlp": 1.01819539,
      "epoch": 0.8573318102567186,
      "flos": 27160389342720.0,
      "grad_norm": 2.814175959120464,
      "language_loss": 0.78828514,
      "learning_rate": 2.0961214783509806e-07,
      "loss": 0.80951875,
      "num_input_tokens_seen": 154290580,
      "step": 7130,
      "time_per_iteration": 2.8651297092437744
    },
    {
      "auxiliary_loss_clip": 0.01126537,
      "auxiliary_loss_mlp": 0.00762011,
      "balance_loss_clip": 1.04004586,
      "balance_loss_mlp": 1.0004071,
      "epoch": 0.8574520531473576,
      "flos": 24936585402240.0,
      "grad_norm": 2.7462749790046153,
      "language_loss": 0.74459785,
      "learning_rate": 2.0926511403148051e-07,
      "loss": 0.76348335,
      "num_input_tokens_seen": 154309545,
      "step": 7131,
      "time_per_iteration": 2.7840471267700195
    },
    {
      "auxiliary_loss_clip": 0.01154756,
      "auxiliary_loss_mlp": 0.0102326,
      "balance_loss_clip": 1.04882383,
      "balance_loss_mlp": 1.01569295,
      "epoch": 0.8575722960379968,
      "flos": 18771154513920.0,
      "grad_norm": 2.478293991050913,
      "language_loss": 0.76111847,
      "learning_rate": 2.0891835188092143e-07,
      "loss": 0.7828986,
      "num_input_tokens_seen": 154326545,
      "step": 7132,
      "time_per_iteration": 2.678924798965454
    },
    {
      "auxiliary_loss_clip": 0.01071551,
      "auxiliary_loss_mlp": 0.01027074,
      "balance_loss_clip": 1.03292203,
      "balance_loss_mlp": 1.01983464,
      "epoch": 0.8576925389286358,
      "flos": 22200300167040.0,
      "grad_norm": 2.584502449453496,
      "language_loss": 0.81684279,
      "learning_rate": 2.0857186143602434e-07,
      "loss": 0.83782899,
      "num_input_tokens_seen": 154345190,
      "step": 7133,
      "time_per_iteration": 3.1406192779541016
    },
    {
      "auxiliary_loss_clip": 0.01111363,
      "auxiliary_loss_mlp": 0.00762195,
      "balance_loss_clip": 1.04213154,
      "balance_loss_mlp": 1.00046766,
      "epoch": 0.8578127818192749,
      "flos": 22894345733760.0,
      "grad_norm": 1.676762702772385,
      "language_loss": 0.67866242,
      "learning_rate": 2.0822564274935094e-07,
      "loss": 0.69739795,
      "num_input_tokens_seen": 154364615,
      "step": 7134,
      "time_per_iteration": 3.5599515438079834
    },
    {
      "auxiliary_loss_clip": 0.01129252,
      "auxiliary_loss_mlp": 0.01026513,
      "balance_loss_clip": 1.04264307,
      "balance_loss_mlp": 1.01830816,
      "epoch": 0.8579330247099141,
      "flos": 34824839541120.0,
      "grad_norm": 3.668676812474857,
      "language_loss": 0.67112803,
      "learning_rate": 2.078796958734239e-07,
      "loss": 0.69268572,
      "num_input_tokens_seen": 154387335,
      "step": 7135,
      "time_per_iteration": 2.8697092533111572
    },
    {
      "auxiliary_loss_clip": 0.01153936,
      "auxiliary_loss_mlp": 0.01022411,
      "balance_loss_clip": 1.04703367,
      "balance_loss_mlp": 1.01468062,
      "epoch": 0.8580532676005531,
      "flos": 19755681367680.0,
      "grad_norm": 2.083816000299754,
      "language_loss": 0.75375879,
      "learning_rate": 2.0753402086072124e-07,
      "loss": 0.77552223,
      "num_input_tokens_seen": 154405965,
      "step": 7136,
      "time_per_iteration": 2.684081792831421
    },
    {
      "auxiliary_loss_clip": 0.01098838,
      "auxiliary_loss_mlp": 0.01027198,
      "balance_loss_clip": 1.03641391,
      "balance_loss_mlp": 1.01968753,
      "epoch": 0.8581735104911922,
      "flos": 22739318634240.0,
      "grad_norm": 3.1613307292915285,
      "language_loss": 0.75050867,
      "learning_rate": 2.071886177636828e-07,
      "loss": 0.77176911,
      "num_input_tokens_seen": 154422750,
      "step": 7137,
      "time_per_iteration": 2.7439308166503906
    },
    {
      "auxiliary_loss_clip": 0.0112387,
      "auxiliary_loss_mlp": 0.01024169,
      "balance_loss_clip": 1.04118156,
      "balance_loss_mlp": 1.01633072,
      "epoch": 0.8582937533818313,
      "flos": 23149131880320.0,
      "grad_norm": 2.194167305164457,
      "language_loss": 0.83455551,
      "learning_rate": 2.0684348663470575e-07,
      "loss": 0.85603595,
      "num_input_tokens_seen": 154442930,
      "step": 7138,
      "time_per_iteration": 2.760470390319824
    },
    {
      "auxiliary_loss_clip": 0.01144217,
      "auxiliary_loss_mlp": 0.01021271,
      "balance_loss_clip": 1.04691577,
      "balance_loss_mlp": 1.01278925,
      "epoch": 0.8584139962724704,
      "flos": 19498668577920.0,
      "grad_norm": 2.2975667961698805,
      "language_loss": 0.61500436,
      "learning_rate": 2.0649862752614555e-07,
      "loss": 0.63665926,
      "num_input_tokens_seen": 154461640,
      "step": 7139,
      "time_per_iteration": 2.6514241695404053
    },
    {
      "auxiliary_loss_clip": 0.01047117,
      "auxiliary_loss_mlp": 0.01001692,
      "balance_loss_clip": 1.01411796,
      "balance_loss_mlp": 1.00073242,
      "epoch": 0.8585342391631094,
      "flos": 71276577788160.0,
      "grad_norm": 0.7547017712528158,
      "language_loss": 0.57010961,
      "learning_rate": 2.0615404049031838e-07,
      "loss": 0.59059769,
      "num_input_tokens_seen": 154518610,
      "step": 7140,
      "time_per_iteration": 3.268239974975586
    },
    {
      "auxiliary_loss_clip": 0.01080381,
      "auxiliary_loss_mlp": 0.01029574,
      "balance_loss_clip": 1.03906918,
      "balance_loss_mlp": 1.02121782,
      "epoch": 0.8586544820537486,
      "flos": 10815432929280.0,
      "grad_norm": 2.333992169319431,
      "language_loss": 0.78311414,
      "learning_rate": 2.0580972557949616e-07,
      "loss": 0.80421376,
      "num_input_tokens_seen": 154533700,
      "step": 7141,
      "time_per_iteration": 2.8156728744506836
    },
    {
      "auxiliary_loss_clip": 0.01024812,
      "auxiliary_loss_mlp": 0.00999643,
      "balance_loss_clip": 1.00909233,
      "balance_loss_mlp": 0.99861735,
      "epoch": 0.8587747249443877,
      "flos": 64811184422400.0,
      "grad_norm": 0.8095790274417952,
      "language_loss": 0.5421772,
      "learning_rate": 2.054656828459125e-07,
      "loss": 0.56242174,
      "num_input_tokens_seen": 154597810,
      "step": 7142,
      "time_per_iteration": 3.273862361907959
    },
    {
      "auxiliary_loss_clip": 0.01148825,
      "auxiliary_loss_mlp": 0.01024692,
      "balance_loss_clip": 1.04523253,
      "balance_loss_mlp": 1.0162164,
      "epoch": 0.8588949678350267,
      "flos": 26834607964800.0,
      "grad_norm": 3.226649543054765,
      "language_loss": 0.77379119,
      "learning_rate": 2.051219123417578e-07,
      "loss": 0.79552639,
      "num_input_tokens_seen": 154617870,
      "step": 7143,
      "time_per_iteration": 4.038362979888916
    },
    {
      "auxiliary_loss_clip": 0.01148403,
      "auxiliary_loss_mlp": 0.01025739,
      "balance_loss_clip": 1.04336095,
      "balance_loss_mlp": 1.01739419,
      "epoch": 0.8590152107256659,
      "flos": 26104256726400.0,
      "grad_norm": 9.430255582927739,
      "language_loss": 0.60255527,
      "learning_rate": 2.0477841411918196e-07,
      "loss": 0.62429667,
      "num_input_tokens_seen": 154637395,
      "step": 7144,
      "time_per_iteration": 3.66383695602417
    },
    {
      "auxiliary_loss_clip": 0.01119572,
      "auxiliary_loss_mlp": 0.01020038,
      "balance_loss_clip": 1.03890264,
      "balance_loss_mlp": 1.01290965,
      "epoch": 0.859135453616305,
      "flos": 26140885620480.0,
      "grad_norm": 2.0370722295820602,
      "language_loss": 0.74547428,
      "learning_rate": 2.0443518823029326e-07,
      "loss": 0.76687044,
      "num_input_tokens_seen": 154657935,
      "step": 7145,
      "time_per_iteration": 3.693016290664673
    },
    {
      "auxiliary_loss_clip": 0.01109759,
      "auxiliary_loss_mlp": 0.01024124,
      "balance_loss_clip": 1.04050779,
      "balance_loss_mlp": 1.0170368,
      "epoch": 0.859255696506944,
      "flos": 12969319046400.0,
      "grad_norm": 2.725525209244872,
      "language_loss": 0.76802742,
      "learning_rate": 2.0409223472715854e-07,
      "loss": 0.78936625,
      "num_input_tokens_seen": 154675080,
      "step": 7146,
      "time_per_iteration": 2.738231658935547
    },
    {
      "auxiliary_loss_clip": 0.01123504,
      "auxiliary_loss_mlp": 0.00762059,
      "balance_loss_clip": 1.04788518,
      "balance_loss_mlp": 1.00047398,
      "epoch": 0.8593759393975832,
      "flos": 18475753063680.0,
      "grad_norm": 2.088783061021281,
      "language_loss": 0.74838555,
      "learning_rate": 2.0374955366180434e-07,
      "loss": 0.76724124,
      "num_input_tokens_seen": 154692720,
      "step": 7147,
      "time_per_iteration": 2.7297253608703613
    },
    {
      "auxiliary_loss_clip": 0.010973,
      "auxiliary_loss_mlp": 0.0102254,
      "balance_loss_clip": 1.04287529,
      "balance_loss_mlp": 1.01460671,
      "epoch": 0.8594961822882222,
      "flos": 22200156512640.0,
      "grad_norm": 2.425231880380815,
      "language_loss": 0.72724819,
      "learning_rate": 2.034071450862147e-07,
      "loss": 0.74844652,
      "num_input_tokens_seen": 154710190,
      "step": 7148,
      "time_per_iteration": 2.804382562637329
    },
    {
      "auxiliary_loss_clip": 0.01146426,
      "auxiliary_loss_mlp": 0.01027866,
      "balance_loss_clip": 1.04815698,
      "balance_loss_mlp": 1.02001572,
      "epoch": 0.8596164251788613,
      "flos": 23294749616640.0,
      "grad_norm": 2.8088802090552774,
      "language_loss": 0.76929367,
      "learning_rate": 2.030650090523327e-07,
      "loss": 0.79103655,
      "num_input_tokens_seen": 154729380,
      "step": 7149,
      "time_per_iteration": 2.7219316959381104
    },
    {
      "auxiliary_loss_clip": 0.01073538,
      "auxiliary_loss_mlp": 0.01031146,
      "balance_loss_clip": 1.0380193,
      "balance_loss_mlp": 1.02304542,
      "epoch": 0.8597366680695004,
      "flos": 31649905416960.0,
      "grad_norm": 1.679529139395361,
      "language_loss": 0.59482992,
      "learning_rate": 2.0272314561205995e-07,
      "loss": 0.61587673,
      "num_input_tokens_seen": 154749775,
      "step": 7150,
      "time_per_iteration": 3.051276206970215
    },
    {
      "auxiliary_loss_clip": 0.01114105,
      "auxiliary_loss_mlp": 0.01025144,
      "balance_loss_clip": 1.03703392,
      "balance_loss_mlp": 1.0176692,
      "epoch": 0.8598569109601395,
      "flos": 21287738211840.0,
      "grad_norm": 2.215410146998637,
      "language_loss": 0.7266255,
      "learning_rate": 2.023815548172567e-07,
      "loss": 0.74801803,
      "num_input_tokens_seen": 154769845,
      "step": 7151,
      "time_per_iteration": 2.807668685913086
    },
    {
      "auxiliary_loss_clip": 0.01124833,
      "auxiliary_loss_mlp": 0.01025219,
      "balance_loss_clip": 1.04338312,
      "balance_loss_mlp": 1.01760185,
      "epoch": 0.8599771538507786,
      "flos": 25447809720960.0,
      "grad_norm": 1.9901092934103557,
      "language_loss": 0.66398883,
      "learning_rate": 2.0204023671974267e-07,
      "loss": 0.6854893,
      "num_input_tokens_seen": 154789230,
      "step": 7152,
      "time_per_iteration": 2.7484264373779297
    },
    {
      "auxiliary_loss_clip": 0.01151242,
      "auxiliary_loss_mlp": 0.01020474,
      "balance_loss_clip": 1.04332399,
      "balance_loss_mlp": 1.01321971,
      "epoch": 0.8600973967414177,
      "flos": 16723958768640.0,
      "grad_norm": 2.31703940383353,
      "language_loss": 0.81021726,
      "learning_rate": 2.0169919137129532e-07,
      "loss": 0.83193445,
      "num_input_tokens_seen": 154807670,
      "step": 7153,
      "time_per_iteration": 2.7259535789489746
    },
    {
      "auxiliary_loss_clip": 0.01130129,
      "auxiliary_loss_mlp": 0.01026066,
      "balance_loss_clip": 1.04785705,
      "balance_loss_mlp": 1.01762581,
      "epoch": 0.8602176396320568,
      "flos": 25227928615680.0,
      "grad_norm": 2.2690325035603434,
      "language_loss": 0.70745498,
      "learning_rate": 2.013584188236508e-07,
      "loss": 0.72901702,
      "num_input_tokens_seen": 154825575,
      "step": 7154,
      "time_per_iteration": 2.7807772159576416
    },
    {
      "auxiliary_loss_clip": 0.01113585,
      "auxiliary_loss_mlp": 0.01022822,
      "balance_loss_clip": 1.04218733,
      "balance_loss_mlp": 1.01544917,
      "epoch": 0.8603378825226958,
      "flos": 20412236113920.0,
      "grad_norm": 1.6364192579990469,
      "language_loss": 0.79931313,
      "learning_rate": 2.0101791912850396e-07,
      "loss": 0.82067716,
      "num_input_tokens_seen": 154845115,
      "step": 7155,
      "time_per_iteration": 2.770967483520508
    },
    {
      "auxiliary_loss_clip": 0.01138529,
      "auxiliary_loss_mlp": 0.00762337,
      "balance_loss_clip": 1.04356074,
      "balance_loss_mlp": 1.00042713,
      "epoch": 0.8604581254133349,
      "flos": 34930201109760.0,
      "grad_norm": 2.0864445410621455,
      "language_loss": 0.64201617,
      "learning_rate": 2.006776923375082e-07,
      "loss": 0.66102481,
      "num_input_tokens_seen": 154866770,
      "step": 7156,
      "time_per_iteration": 3.8370018005371094
    },
    {
      "auxiliary_loss_clip": 0.0112689,
      "auxiliary_loss_mlp": 0.01025441,
      "balance_loss_clip": 1.04379153,
      "balance_loss_mlp": 1.01759672,
      "epoch": 0.860578368303974,
      "flos": 22596538072320.0,
      "grad_norm": 2.3036104700513556,
      "language_loss": 0.71256018,
      "learning_rate": 2.003377385022764e-07,
      "loss": 0.73408347,
      "num_input_tokens_seen": 154885595,
      "step": 7157,
      "time_per_iteration": 2.680867910385132
    },
    {
      "auxiliary_loss_clip": 0.01113994,
      "auxiliary_loss_mlp": 0.01025083,
      "balance_loss_clip": 1.0385778,
      "balance_loss_mlp": 1.0178678,
      "epoch": 0.8606986111946131,
      "flos": 21324331192320.0,
      "grad_norm": 3.475504109002881,
      "language_loss": 0.77661496,
      "learning_rate": 1.9999805767437826e-07,
      "loss": 0.7980057,
      "num_input_tokens_seen": 154904485,
      "step": 7158,
      "time_per_iteration": 2.7900938987731934
    },
    {
      "auxiliary_loss_clip": 0.01141885,
      "auxiliary_loss_mlp": 0.01020695,
      "balance_loss_clip": 1.04406261,
      "balance_loss_mlp": 1.01325965,
      "epoch": 0.8608188540852522,
      "flos": 28877206769280.0,
      "grad_norm": 1.848152183556251,
      "language_loss": 0.71819139,
      "learning_rate": 1.9965864990534386e-07,
      "loss": 0.73981714,
      "num_input_tokens_seen": 154925010,
      "step": 7159,
      "time_per_iteration": 2.7729616165161133
    },
    {
      "auxiliary_loss_clip": 0.01141128,
      "auxiliary_loss_mlp": 0.0102377,
      "balance_loss_clip": 1.04201102,
      "balance_loss_mlp": 1.01665902,
      "epoch": 0.8609390969758913,
      "flos": 29716187713920.0,
      "grad_norm": 1.8153028691985562,
      "language_loss": 0.77810884,
      "learning_rate": 1.9931951524666092e-07,
      "loss": 0.79975778,
      "num_input_tokens_seen": 154946100,
      "step": 7160,
      "time_per_iteration": 2.724322557449341
    },
    {
      "auxiliary_loss_clip": 0.01081728,
      "auxiliary_loss_mlp": 0.00761506,
      "balance_loss_clip": 1.03745651,
      "balance_loss_mlp": 1.00053322,
      "epoch": 0.8610593398665304,
      "flos": 21249349551360.0,
      "grad_norm": 1.6771601306894413,
      "language_loss": 0.80962777,
      "learning_rate": 1.9898065374977534e-07,
      "loss": 0.82806015,
      "num_input_tokens_seen": 154966305,
      "step": 7161,
      "time_per_iteration": 2.943890333175659
    },
    {
      "auxiliary_loss_clip": 0.01117927,
      "auxiliary_loss_mlp": 0.00760728,
      "balance_loss_clip": 1.04281294,
      "balance_loss_mlp": 1.00041413,
      "epoch": 0.8611795827571694,
      "flos": 14830102183680.0,
      "grad_norm": 2.0526264489456385,
      "language_loss": 0.73221439,
      "learning_rate": 1.9864206546609342e-07,
      "loss": 0.751001,
      "num_input_tokens_seen": 154985145,
      "step": 7162,
      "time_per_iteration": 2.684375047683716
    },
    {
      "auxiliary_loss_clip": 0.01109375,
      "auxiliary_loss_mlp": 0.01025795,
      "balance_loss_clip": 1.04035819,
      "balance_loss_mlp": 1.01851702,
      "epoch": 0.8612998256478086,
      "flos": 24243258107520.0,
      "grad_norm": 1.743916203145559,
      "language_loss": 0.84563214,
      "learning_rate": 1.983037504469771e-07,
      "loss": 0.86698383,
      "num_input_tokens_seen": 155003855,
      "step": 7163,
      "time_per_iteration": 2.7800722122192383
    },
    {
      "auxiliary_loss_clip": 0.01140126,
      "auxiliary_loss_mlp": 0.01029787,
      "balance_loss_clip": 1.0486474,
      "balance_loss_mlp": 1.02156711,
      "epoch": 0.8614200685384477,
      "flos": 21252653602560.0,
      "grad_norm": 1.7946568499100162,
      "language_loss": 0.66825873,
      "learning_rate": 1.9796570874374984e-07,
      "loss": 0.68995786,
      "num_input_tokens_seen": 155023960,
      "step": 7164,
      "time_per_iteration": 2.681689500808716
    },
    {
      "auxiliary_loss_clip": 0.01143332,
      "auxiliary_loss_mlp": 0.01032679,
      "balance_loss_clip": 1.04452479,
      "balance_loss_mlp": 1.02493024,
      "epoch": 0.8615403114290867,
      "flos": 20007738080640.0,
      "grad_norm": 2.2036034951737977,
      "language_loss": 0.77591753,
      "learning_rate": 1.976279404076917e-07,
      "loss": 0.79767764,
      "num_input_tokens_seen": 155043360,
      "step": 7165,
      "time_per_iteration": 2.7324376106262207
    },
    {
      "auxiliary_loss_clip": 0.01143666,
      "auxiliary_loss_mlp": 0.01030899,
      "balance_loss_clip": 1.04518282,
      "balance_loss_mlp": 1.02316809,
      "epoch": 0.8616605543197259,
      "flos": 29789373674880.0,
      "grad_norm": 2.011253417596588,
      "language_loss": 0.76052791,
      "learning_rate": 1.9729044549004193e-07,
      "loss": 0.78227359,
      "num_input_tokens_seen": 155064745,
      "step": 7166,
      "time_per_iteration": 2.7010557651519775
    },
    {
      "auxiliary_loss_clip": 0.01107009,
      "auxiliary_loss_mlp": 0.01024272,
      "balance_loss_clip": 1.04025698,
      "balance_loss_mlp": 1.01688135,
      "epoch": 0.8617807972103649,
      "flos": 28911609020160.0,
      "grad_norm": 2.2870197408402704,
      "language_loss": 0.70335871,
      "learning_rate": 1.9695322404199822e-07,
      "loss": 0.72467154,
      "num_input_tokens_seen": 155086790,
      "step": 7167,
      "time_per_iteration": 2.8647000789642334
    },
    {
      "auxiliary_loss_clip": 0.01120513,
      "auxiliary_loss_mlp": 0.01019588,
      "balance_loss_clip": 1.03988552,
      "balance_loss_mlp": 1.01203871,
      "epoch": 0.861901040101004,
      "flos": 27673804391040.0,
      "grad_norm": 2.7605502966724504,
      "language_loss": 0.82449037,
      "learning_rate": 1.9661627611471654e-07,
      "loss": 0.84589136,
      "num_input_tokens_seen": 155106585,
      "step": 7168,
      "time_per_iteration": 2.779496908187866
    },
    {
      "auxiliary_loss_clip": 0.01114549,
      "auxiliary_loss_mlp": 0.0102617,
      "balance_loss_clip": 1.03909957,
      "balance_loss_mlp": 1.01796269,
      "epoch": 0.8620212829916432,
      "flos": 49748056755840.0,
      "grad_norm": 2.386258953941684,
      "language_loss": 0.70334387,
      "learning_rate": 1.9627960175931246e-07,
      "loss": 0.72475106,
      "num_input_tokens_seen": 155131285,
      "step": 7169,
      "time_per_iteration": 3.8691232204437256
    },
    {
      "auxiliary_loss_clip": 0.0109311,
      "auxiliary_loss_mlp": 0.01027046,
      "balance_loss_clip": 1.03732133,
      "balance_loss_mlp": 1.01920831,
      "epoch": 0.8621415258822822,
      "flos": 21138672769920.0,
      "grad_norm": 1.9902817013436196,
      "language_loss": 0.74450976,
      "learning_rate": 1.9594320102685847e-07,
      "loss": 0.76571131,
      "num_input_tokens_seen": 155150555,
      "step": 7170,
      "time_per_iteration": 3.711411952972412
    },
    {
      "auxiliary_loss_clip": 0.01152649,
      "auxiliary_loss_mlp": 0.01021654,
      "balance_loss_clip": 1.04755044,
      "balance_loss_mlp": 1.01470149,
      "epoch": 0.8622617687729213,
      "flos": 21689039934720.0,
      "grad_norm": 2.9347127479668487,
      "language_loss": 0.64015156,
      "learning_rate": 1.956070739683864e-07,
      "loss": 0.66189456,
      "num_input_tokens_seen": 155169890,
      "step": 7171,
      "time_per_iteration": 3.627922773361206
    },
    {
      "auxiliary_loss_clip": 0.01122993,
      "auxiliary_loss_mlp": 0.01029661,
      "balance_loss_clip": 1.0444752,
      "balance_loss_mlp": 1.02280021,
      "epoch": 0.8623820116635604,
      "flos": 26250592734720.0,
      "grad_norm": 1.572408508279045,
      "language_loss": 0.74112594,
      "learning_rate": 1.9527122063488678e-07,
      "loss": 0.76265252,
      "num_input_tokens_seen": 155191005,
      "step": 7172,
      "time_per_iteration": 2.8032066822052
    },
    {
      "auxiliary_loss_clip": 0.01115309,
      "auxiliary_loss_mlp": 0.01021634,
      "balance_loss_clip": 1.03868353,
      "balance_loss_mlp": 1.01390338,
      "epoch": 0.8625022545541995,
      "flos": 19647554451840.0,
      "grad_norm": 1.9739099955273032,
      "language_loss": 0.80188447,
      "learning_rate": 1.9493564107730755e-07,
      "loss": 0.82325393,
      "num_input_tokens_seen": 155211005,
      "step": 7173,
      "time_per_iteration": 2.8154380321502686
    },
    {
      "auxiliary_loss_clip": 0.01151723,
      "auxiliary_loss_mlp": 0.01022971,
      "balance_loss_clip": 1.04410839,
      "balance_loss_mlp": 1.0156635,
      "epoch": 0.8626224974448385,
      "flos": 21908382336000.0,
      "grad_norm": 1.9130036932496692,
      "language_loss": 0.61094129,
      "learning_rate": 1.9460033534655684e-07,
      "loss": 0.63268816,
      "num_input_tokens_seen": 155230365,
      "step": 7174,
      "time_per_iteration": 2.7485361099243164
    },
    {
      "auxiliary_loss_clip": 0.01101444,
      "auxiliary_loss_mlp": 0.01024245,
      "balance_loss_clip": 1.03795171,
      "balance_loss_mlp": 1.01611531,
      "epoch": 0.8627427403354777,
      "flos": 23331198942720.0,
      "grad_norm": 1.7075365519446335,
      "language_loss": 0.8392294,
      "learning_rate": 1.9426530349349978e-07,
      "loss": 0.86048627,
      "num_input_tokens_seen": 155250815,
      "step": 7175,
      "time_per_iteration": 2.8989648818969727
    },
    {
      "auxiliary_loss_clip": 0.01146615,
      "auxiliary_loss_mlp": 0.0102768,
      "balance_loss_clip": 1.04390574,
      "balance_loss_mlp": 1.02077794,
      "epoch": 0.8628629832261168,
      "flos": 16362877299840.0,
      "grad_norm": 2.4231119515778534,
      "language_loss": 0.64858794,
      "learning_rate": 1.9393054556896038e-07,
      "loss": 0.67033088,
      "num_input_tokens_seen": 155268515,
      "step": 7176,
      "time_per_iteration": 2.6491105556488037
    },
    {
      "auxiliary_loss_clip": 0.01130431,
      "auxiliary_loss_mlp": 0.01026316,
      "balance_loss_clip": 1.04035068,
      "balance_loss_mlp": 1.01810575,
      "epoch": 0.8629832261167558,
      "flos": 28103941756800.0,
      "grad_norm": 3.181132006048911,
      "language_loss": 0.69194567,
      "learning_rate": 1.9359606162372133e-07,
      "loss": 0.71351314,
      "num_input_tokens_seen": 155290120,
      "step": 7177,
      "time_per_iteration": 2.7649455070495605
    },
    {
      "auxiliary_loss_clip": 0.01133146,
      "auxiliary_loss_mlp": 0.01026122,
      "balance_loss_clip": 1.0424248,
      "balance_loss_mlp": 1.01886821,
      "epoch": 0.863103469007395,
      "flos": 20230061310720.0,
      "grad_norm": 1.6483475595780435,
      "language_loss": 0.70541537,
      "learning_rate": 1.9326185170852293e-07,
      "loss": 0.72700799,
      "num_input_tokens_seen": 155309085,
      "step": 7178,
      "time_per_iteration": 2.7517178058624268
    },
    {
      "auxiliary_loss_clip": 0.01101451,
      "auxiliary_loss_mlp": 0.00762509,
      "balance_loss_clip": 1.04034281,
      "balance_loss_mlp": 1.00045967,
      "epoch": 0.863223711898034,
      "flos": 24498547044480.0,
      "grad_norm": 1.7760085722658985,
      "language_loss": 0.72247303,
      "learning_rate": 1.9292791587406598e-07,
      "loss": 0.74111265,
      "num_input_tokens_seen": 155327945,
      "step": 7179,
      "time_per_iteration": 2.861860752105713
    },
    {
      "auxiliary_loss_clip": 0.01090522,
      "auxiliary_loss_mlp": 0.01027423,
      "balance_loss_clip": 1.03760791,
      "balance_loss_mlp": 1.01979089,
      "epoch": 0.8633439547886731,
      "flos": 17675376261120.0,
      "grad_norm": 2.055262421171125,
      "language_loss": 0.86695898,
      "learning_rate": 1.9259425417100661e-07,
      "loss": 0.88813841,
      "num_input_tokens_seen": 155344060,
      "step": 7180,
      "time_per_iteration": 2.8034727573394775
    },
    {
      "auxiliary_loss_clip": 0.01087248,
      "auxiliary_loss_mlp": 0.01026386,
      "balance_loss_clip": 1.03611636,
      "balance_loss_mlp": 1.01883399,
      "epoch": 0.8634641976793123,
      "flos": 12895055677440.0,
      "grad_norm": 4.159457888900711,
      "language_loss": 0.74872875,
      "learning_rate": 1.9226086664996234e-07,
      "loss": 0.76986504,
      "num_input_tokens_seen": 155362305,
      "step": 7181,
      "time_per_iteration": 3.751936674118042
    },
    {
      "auxiliary_loss_clip": 0.01095991,
      "auxiliary_loss_mlp": 0.01024253,
      "balance_loss_clip": 1.04237342,
      "balance_loss_mlp": 1.01726198,
      "epoch": 0.8635844405699513,
      "flos": 23878980328320.0,
      "grad_norm": 1.8698449410536389,
      "language_loss": 0.7435571,
      "learning_rate": 1.9192775336150712e-07,
      "loss": 0.76475948,
      "num_input_tokens_seen": 155382605,
      "step": 7182,
      "time_per_iteration": 2.897447347640991
    },
    {
      "auxiliary_loss_clip": 0.01043119,
      "auxiliary_loss_mlp": 0.01000822,
      "balance_loss_clip": 1.00859165,
      "balance_loss_mlp": 0.99973691,
      "epoch": 0.8637046834605904,
      "flos": 60453387521280.0,
      "grad_norm": 0.7602591637230915,
      "language_loss": 0.56272811,
      "learning_rate": 1.915949143561739e-07,
      "loss": 0.58316749,
      "num_input_tokens_seen": 155437280,
      "step": 7183,
      "time_per_iteration": 3.231801748275757
    },
    {
      "auxiliary_loss_clip": 0.01138868,
      "auxiliary_loss_mlp": 0.01025639,
      "balance_loss_clip": 1.04323709,
      "balance_loss_mlp": 1.01712179,
      "epoch": 0.8638249263512295,
      "flos": 20558751690240.0,
      "grad_norm": 1.75018326929471,
      "language_loss": 0.77957809,
      "learning_rate": 1.9126234968445498e-07,
      "loss": 0.80122316,
      "num_input_tokens_seen": 155456970,
      "step": 7184,
      "time_per_iteration": 2.684251308441162
    },
    {
      "auxiliary_loss_clip": 0.01138922,
      "auxiliary_loss_mlp": 0.01023207,
      "balance_loss_clip": 1.04609919,
      "balance_loss_mlp": 1.01563764,
      "epoch": 0.8639451692418686,
      "flos": 26615768353920.0,
      "grad_norm": 1.5710882376438786,
      "language_loss": 0.67714596,
      "learning_rate": 1.9093005939679884e-07,
      "loss": 0.69876724,
      "num_input_tokens_seen": 155478925,
      "step": 7185,
      "time_per_iteration": 2.7416982650756836
    },
    {
      "auxiliary_loss_clip": 0.01142193,
      "auxiliary_loss_mlp": 0.01021557,
      "balance_loss_clip": 1.04390144,
      "balance_loss_mlp": 1.01374865,
      "epoch": 0.8640654121325076,
      "flos": 15122450977920.0,
      "grad_norm": 2.226743874933661,
      "language_loss": 0.76782799,
      "learning_rate": 1.9059804354361452e-07,
      "loss": 0.78946549,
      "num_input_tokens_seen": 155496700,
      "step": 7186,
      "time_per_iteration": 2.641932964324951
    },
    {
      "auxiliary_loss_clip": 0.01142022,
      "auxiliary_loss_mlp": 0.01026067,
      "balance_loss_clip": 1.04568338,
      "balance_loss_mlp": 1.01856315,
      "epoch": 0.8641856550231467,
      "flos": 31869068250240.0,
      "grad_norm": 1.67484792879823,
      "language_loss": 0.70385647,
      "learning_rate": 1.902663021752684e-07,
      "loss": 0.7255373,
      "num_input_tokens_seen": 155518130,
      "step": 7187,
      "time_per_iteration": 2.8087239265441895
    },
    {
      "auxiliary_loss_clip": 0.01096022,
      "auxiliary_loss_mlp": 0.01032715,
      "balance_loss_clip": 1.03831077,
      "balance_loss_mlp": 1.02526784,
      "epoch": 0.8643058979137859,
      "flos": 14976545932800.0,
      "grad_norm": 2.692526826183163,
      "language_loss": 0.82645291,
      "learning_rate": 1.8993483534208556e-07,
      "loss": 0.84774029,
      "num_input_tokens_seen": 155537040,
      "step": 7188,
      "time_per_iteration": 2.7716026306152344
    },
    {
      "auxiliary_loss_clip": 0.01141802,
      "auxiliary_loss_mlp": 0.01020984,
      "balance_loss_clip": 1.04481483,
      "balance_loss_mlp": 1.01350403,
      "epoch": 0.8644261408044249,
      "flos": 13115726881920.0,
      "grad_norm": 3.1827548934731382,
      "language_loss": 0.74670255,
      "learning_rate": 1.8960364309434884e-07,
      "loss": 0.76833045,
      "num_input_tokens_seen": 155554535,
      "step": 7189,
      "time_per_iteration": 2.6816396713256836
    },
    {
      "auxiliary_loss_clip": 0.01081532,
      "auxiliary_loss_mlp": 0.01023158,
      "balance_loss_clip": 1.03806961,
      "balance_loss_mlp": 1.01559997,
      "epoch": 0.864546383695064,
      "flos": 20850920916480.0,
      "grad_norm": 2.1189057478662052,
      "language_loss": 0.78330743,
      "learning_rate": 1.8927272548229967e-07,
      "loss": 0.80435431,
      "num_input_tokens_seen": 155574225,
      "step": 7190,
      "time_per_iteration": 2.797236442565918
    },
    {
      "auxiliary_loss_clip": 0.01102013,
      "auxiliary_loss_mlp": 0.01033306,
      "balance_loss_clip": 1.0418669,
      "balance_loss_mlp": 1.0249728,
      "epoch": 0.8646666265857031,
      "flos": 21324582587520.0,
      "grad_norm": 1.8134845612497508,
      "language_loss": 0.82912123,
      "learning_rate": 1.8894208255613876e-07,
      "loss": 0.85047442,
      "num_input_tokens_seen": 155593540,
      "step": 7191,
      "time_per_iteration": 2.772678852081299
    },
    {
      "auxiliary_loss_clip": 0.01092621,
      "auxiliary_loss_mlp": 0.01027953,
      "balance_loss_clip": 1.03456974,
      "balance_loss_mlp": 1.02061844,
      "epoch": 0.8647868694763422,
      "flos": 19750833031680.0,
      "grad_norm": 2.485920145358389,
      "language_loss": 0.78218389,
      "learning_rate": 1.8861171436602397e-07,
      "loss": 0.80338967,
      "num_input_tokens_seen": 155610655,
      "step": 7192,
      "time_per_iteration": 2.8109829425811768
    },
    {
      "auxiliary_loss_clip": 0.01120174,
      "auxiliary_loss_mlp": 0.01028681,
      "balance_loss_clip": 1.04714036,
      "balance_loss_mlp": 1.02020538,
      "epoch": 0.8649071123669813,
      "flos": 26176760328960.0,
      "grad_norm": 2.151934116141033,
      "language_loss": 0.80517715,
      "learning_rate": 1.882816209620719e-07,
      "loss": 0.82666576,
      "num_input_tokens_seen": 155627365,
      "step": 7193,
      "time_per_iteration": 2.7427332401275635
    },
    {
      "auxiliary_loss_clip": 0.01172151,
      "auxiliary_loss_mlp": 0.00762069,
      "balance_loss_clip": 1.04991651,
      "balance_loss_mlp": 1.00051808,
      "epoch": 0.8650273552576204,
      "flos": 20302888135680.0,
      "grad_norm": 1.8680586244058328,
      "language_loss": 0.76705438,
      "learning_rate": 1.8795180239435738e-07,
      "loss": 0.78639662,
      "num_input_tokens_seen": 155646220,
      "step": 7194,
      "time_per_iteration": 2.7283952236175537
    },
    {
      "auxiliary_loss_clip": 0.01135528,
      "auxiliary_loss_mlp": 0.01025316,
      "balance_loss_clip": 1.04305863,
      "balance_loss_mlp": 1.01743579,
      "epoch": 0.8651475981482595,
      "flos": 23951088881280.0,
      "grad_norm": 3.080372412665803,
      "language_loss": 0.75210989,
      "learning_rate": 1.8762225871291348e-07,
      "loss": 0.77371824,
      "num_input_tokens_seen": 155662095,
      "step": 7195,
      "time_per_iteration": 4.616030931472778
    },
    {
      "auxiliary_loss_clip": 0.01037651,
      "auxiliary_loss_mlp": 0.01019069,
      "balance_loss_clip": 1.03390527,
      "balance_loss_mlp": 1.01116586,
      "epoch": 0.8652678410388985,
      "flos": 21684622561920.0,
      "grad_norm": 2.9947901921535385,
      "language_loss": 0.80977166,
      "learning_rate": 1.8729298996773201e-07,
      "loss": 0.83033895,
      "num_input_tokens_seen": 155680845,
      "step": 7196,
      "time_per_iteration": 4.26993465423584
    },
    {
      "auxiliary_loss_clip": 0.01017766,
      "auxiliary_loss_mlp": 0.01000295,
      "balance_loss_clip": 1.01060271,
      "balance_loss_mlp": 0.99928153,
      "epoch": 0.8653880839295377,
      "flos": 65224660855680.0,
      "grad_norm": 0.8323667169089112,
      "language_loss": 0.60957336,
      "learning_rate": 1.8696399620876301e-07,
      "loss": 0.62975395,
      "num_input_tokens_seen": 155737875,
      "step": 7197,
      "time_per_iteration": 3.6877501010894775
    },
    {
      "auxiliary_loss_clip": 0.01105165,
      "auxiliary_loss_mlp": 0.01027597,
      "balance_loss_clip": 1.04118705,
      "balance_loss_mlp": 1.02025056,
      "epoch": 0.8655083268201768,
      "flos": 17749172753280.0,
      "grad_norm": 2.698592094779672,
      "language_loss": 0.79744864,
      "learning_rate": 1.866352774859141e-07,
      "loss": 0.81877625,
      "num_input_tokens_seen": 155753100,
      "step": 7198,
      "time_per_iteration": 2.802412509918213
    },
    {
      "auxiliary_loss_clip": 0.01145581,
      "auxiliary_loss_mlp": 0.01019889,
      "balance_loss_clip": 1.04271162,
      "balance_loss_mlp": 1.01315951,
      "epoch": 0.8656285697108158,
      "flos": 20703974376960.0,
      "grad_norm": 2.6496483150055816,
      "language_loss": 0.69558197,
      "learning_rate": 1.8630683384905188e-07,
      "loss": 0.71723664,
      "num_input_tokens_seen": 155772430,
      "step": 7199,
      "time_per_iteration": 2.6352221965789795
    },
    {
      "auxiliary_loss_clip": 0.01125663,
      "auxiliary_loss_mlp": 0.01026598,
      "balance_loss_clip": 1.04702663,
      "balance_loss_mlp": 1.01883709,
      "epoch": 0.865748812601455,
      "flos": 18653833716480.0,
      "grad_norm": 1.9126614546321172,
      "language_loss": 0.88622582,
      "learning_rate": 1.8597866534800045e-07,
      "loss": 0.90774846,
      "num_input_tokens_seen": 155787545,
      "step": 7200,
      "time_per_iteration": 2.7286086082458496
    },
    {
      "auxiliary_loss_clip": 0.01157963,
      "auxiliary_loss_mlp": 0.01021857,
      "balance_loss_clip": 1.04927111,
      "balance_loss_mlp": 1.01455534,
      "epoch": 0.865869055492094,
      "flos": 70652554807680.0,
      "grad_norm": 3.6549410137599248,
      "language_loss": 0.74668527,
      "learning_rate": 1.8565077203254398e-07,
      "loss": 0.76848346,
      "num_input_tokens_seen": 155813005,
      "step": 7201,
      "time_per_iteration": 3.017326593399048
    },
    {
      "auxiliary_loss_clip": 0.01073825,
      "auxiliary_loss_mlp": 0.01026372,
      "balance_loss_clip": 1.03725708,
      "balance_loss_mlp": 1.01863837,
      "epoch": 0.8659892983827331,
      "flos": 17383961220480.0,
      "grad_norm": 4.2341774635907665,
      "language_loss": 0.72696996,
      "learning_rate": 1.8532315395242203e-07,
      "loss": 0.74797195,
      "num_input_tokens_seen": 155829455,
      "step": 7202,
      "time_per_iteration": 2.8201181888580322
    },
    {
      "auxiliary_loss_clip": 0.01070978,
      "auxiliary_loss_mlp": 0.01020796,
      "balance_loss_clip": 1.03581011,
      "balance_loss_mlp": 1.01356864,
      "epoch": 0.8661095412733723,
      "flos": 17895221452800.0,
      "grad_norm": 1.9628653426222296,
      "language_loss": 0.72364521,
      "learning_rate": 1.849958111573353e-07,
      "loss": 0.74456298,
      "num_input_tokens_seen": 155848060,
      "step": 7203,
      "time_per_iteration": 3.082996368408203
    },
    {
      "auxiliary_loss_clip": 0.01147841,
      "auxiliary_loss_mlp": 0.01022976,
      "balance_loss_clip": 1.04373622,
      "balance_loss_mlp": 1.01573658,
      "epoch": 0.8662297841640113,
      "flos": 18224163227520.0,
      "grad_norm": 1.9980031943596082,
      "language_loss": 0.63994342,
      "learning_rate": 1.8466874369694074e-07,
      "loss": 0.66165161,
      "num_input_tokens_seen": 155865755,
      "step": 7204,
      "time_per_iteration": 2.6278269290924072
    },
    {
      "auxiliary_loss_clip": 0.01140887,
      "auxiliary_loss_mlp": 0.01027999,
      "balance_loss_clip": 1.04687142,
      "balance_loss_mlp": 1.02038467,
      "epoch": 0.8663500270546504,
      "flos": 16362159027840.0,
      "grad_norm": 5.208878399629637,
      "language_loss": 0.70278943,
      "learning_rate": 1.843419516208542e-07,
      "loss": 0.72447836,
      "num_input_tokens_seen": 155882680,
      "step": 7205,
      "time_per_iteration": 2.640977621078491
    },
    {
      "auxiliary_loss_clip": 0.01129945,
      "auxiliary_loss_mlp": 0.01027694,
      "balance_loss_clip": 1.04372752,
      "balance_loss_mlp": 1.01920044,
      "epoch": 0.8664702699452895,
      "flos": 17894431353600.0,
      "grad_norm": 2.5894551299009234,
      "language_loss": 0.79451019,
      "learning_rate": 1.8401543497865047e-07,
      "loss": 0.81608653,
      "num_input_tokens_seen": 155900680,
      "step": 7206,
      "time_per_iteration": 4.18142032623291
    },
    {
      "auxiliary_loss_clip": 0.01158591,
      "auxiliary_loss_mlp": 0.01026462,
      "balance_loss_clip": 1.04577923,
      "balance_loss_mlp": 1.01831436,
      "epoch": 0.8665905128359286,
      "flos": 30736373794560.0,
      "grad_norm": 2.3336620381518025,
      "language_loss": 0.64105844,
      "learning_rate": 1.836891938198608e-07,
      "loss": 0.66290903,
      "num_input_tokens_seen": 155921105,
      "step": 7207,
      "time_per_iteration": 2.7403626441955566
    },
    {
      "auxiliary_loss_clip": 0.01129287,
      "auxiliary_loss_mlp": 0.01026422,
      "balance_loss_clip": 1.04673982,
      "balance_loss_mlp": 1.01897168,
      "epoch": 0.8667107557265676,
      "flos": 18656419495680.0,
      "grad_norm": 3.4483545619013696,
      "language_loss": 0.71254712,
      "learning_rate": 1.8336322819397677e-07,
      "loss": 0.73410428,
      "num_input_tokens_seen": 155938640,
      "step": 7208,
      "time_per_iteration": 2.8164873123168945
    },
    {
      "auxiliary_loss_clip": 0.01128257,
      "auxiliary_loss_mlp": 0.01028447,
      "balance_loss_clip": 1.04375339,
      "balance_loss_mlp": 1.02024829,
      "epoch": 0.8668309986172068,
      "flos": 20083725302400.0,
      "grad_norm": 2.229739571989627,
      "language_loss": 0.62832701,
      "learning_rate": 1.8303753815044654e-07,
      "loss": 0.64989406,
      "num_input_tokens_seen": 155957945,
      "step": 7209,
      "time_per_iteration": 2.6877450942993164
    },
    {
      "auxiliary_loss_clip": 0.01175768,
      "auxiliary_loss_mlp": 0.01026801,
      "balance_loss_clip": 1.050107,
      "balance_loss_mlp": 1.01756179,
      "epoch": 0.8669512415078459,
      "flos": 21615099788160.0,
      "grad_norm": 2.949927809359586,
      "language_loss": 0.70713806,
      "learning_rate": 1.827121237386773e-07,
      "loss": 0.72916377,
      "num_input_tokens_seen": 155975390,
      "step": 7210,
      "time_per_iteration": 2.5893020629882812
    },
    {
      "auxiliary_loss_clip": 0.01106179,
      "auxiliary_loss_mlp": 0.00762898,
      "balance_loss_clip": 1.04205191,
      "balance_loss_mlp": 1.00048208,
      "epoch": 0.8670714843984849,
      "flos": 17703601372800.0,
      "grad_norm": 2.734645631080283,
      "language_loss": 0.75204325,
      "learning_rate": 1.8238698500803374e-07,
      "loss": 0.77073395,
      "num_input_tokens_seen": 155988155,
      "step": 7211,
      "time_per_iteration": 2.7360517978668213
    },
    {
      "auxiliary_loss_clip": 0.01043372,
      "auxiliary_loss_mlp": 0.00999524,
      "balance_loss_clip": 1.00947881,
      "balance_loss_mlp": 0.99848098,
      "epoch": 0.8671917272891241,
      "flos": 60705483125760.0,
      "grad_norm": 0.7190838935644042,
      "language_loss": 0.5623821,
      "learning_rate": 1.820621220078391e-07,
      "loss": 0.58281106,
      "num_input_tokens_seen": 156052065,
      "step": 7212,
      "time_per_iteration": 3.3295209407806396
    },
    {
      "auxiliary_loss_clip": 0.01143403,
      "auxiliary_loss_mlp": 0.01021272,
      "balance_loss_clip": 1.04454243,
      "balance_loss_mlp": 1.01312423,
      "epoch": 0.8673119701797631,
      "flos": 20451881750400.0,
      "grad_norm": 1.8864019045692022,
      "language_loss": 0.68069249,
      "learning_rate": 1.8173753478737553e-07,
      "loss": 0.70233923,
      "num_input_tokens_seen": 156072500,
      "step": 7213,
      "time_per_iteration": 2.694474458694458
    },
    {
      "auxiliary_loss_clip": 0.01144108,
      "auxiliary_loss_mlp": 0.01025696,
      "balance_loss_clip": 1.0446074,
      "balance_loss_mlp": 1.0173986,
      "epoch": 0.8674322130704022,
      "flos": 19647410797440.0,
      "grad_norm": 2.3717210403338127,
      "language_loss": 0.80180734,
      "learning_rate": 1.8141322339588205e-07,
      "loss": 0.8235054,
      "num_input_tokens_seen": 156089840,
      "step": 7214,
      "time_per_iteration": 2.835750102996826
    },
    {
      "auxiliary_loss_clip": 0.01123823,
      "auxiliary_loss_mlp": 0.01028477,
      "balance_loss_clip": 1.04534578,
      "balance_loss_mlp": 1.0208838,
      "epoch": 0.8675524559610414,
      "flos": 26025001367040.0,
      "grad_norm": 1.9571058571752131,
      "language_loss": 0.70542872,
      "learning_rate": 1.810891878825569e-07,
      "loss": 0.72695166,
      "num_input_tokens_seen": 156109815,
      "step": 7215,
      "time_per_iteration": 2.773607015609741
    },
    {
      "auxiliary_loss_clip": 0.01113012,
      "auxiliary_loss_mlp": 0.01025344,
      "balance_loss_clip": 1.04036498,
      "balance_loss_mlp": 1.01760149,
      "epoch": 0.8676726988516804,
      "flos": 15049444584960.0,
      "grad_norm": 2.6544881324378715,
      "language_loss": 0.71936214,
      "learning_rate": 1.8076542829655561e-07,
      "loss": 0.74074566,
      "num_input_tokens_seen": 156128620,
      "step": 7216,
      "time_per_iteration": 2.7013866901397705
    },
    {
      "auxiliary_loss_clip": 0.0117012,
      "auxiliary_loss_mlp": 0.01029903,
      "balance_loss_clip": 1.04693091,
      "balance_loss_mlp": 1.02143276,
      "epoch": 0.8677929417423195,
      "flos": 16288111140480.0,
      "grad_norm": 2.245175928371307,
      "language_loss": 0.79456031,
      "learning_rate": 1.8044194468699203e-07,
      "loss": 0.81656063,
      "num_input_tokens_seen": 156145930,
      "step": 7217,
      "time_per_iteration": 2.6228926181793213
    },
    {
      "auxiliary_loss_clip": 0.01142006,
      "auxiliary_loss_mlp": 0.01023919,
      "balance_loss_clip": 1.04638886,
      "balance_loss_mlp": 1.01596189,
      "epoch": 0.8679131846329585,
      "flos": 18844160906880.0,
      "grad_norm": 2.433122118489287,
      "language_loss": 0.7616744,
      "learning_rate": 1.8011873710293912e-07,
      "loss": 0.78333366,
      "num_input_tokens_seen": 156164435,
      "step": 7218,
      "time_per_iteration": 2.653536319732666
    },
    {
      "auxiliary_loss_clip": 0.01145648,
      "auxiliary_loss_mlp": 0.01020564,
      "balance_loss_clip": 1.04723907,
      "balance_loss_mlp": 1.01302707,
      "epoch": 0.8680334275235977,
      "flos": 33620718890880.0,
      "grad_norm": 1.8302584071372145,
      "language_loss": 0.69673276,
      "learning_rate": 1.7979580559342677e-07,
      "loss": 0.71839494,
      "num_input_tokens_seen": 156185165,
      "step": 7219,
      "time_per_iteration": 2.802844285964966
    },
    {
      "auxiliary_loss_clip": 0.01141933,
      "auxiliary_loss_mlp": 0.01027916,
      "balance_loss_clip": 1.04403281,
      "balance_loss_mlp": 1.02041817,
      "epoch": 0.8681536704142367,
      "flos": 24681152810880.0,
      "grad_norm": 1.7446021296888214,
      "language_loss": 0.66894072,
      "learning_rate": 1.7947315020744358e-07,
      "loss": 0.69063926,
      "num_input_tokens_seen": 156206260,
      "step": 7220,
      "time_per_iteration": 2.707606315612793
    },
    {
      "auxiliary_loss_clip": 0.0112813,
      "auxiliary_loss_mlp": 0.01019828,
      "balance_loss_clip": 1.04250669,
      "balance_loss_mlp": 1.01229358,
      "epoch": 0.8682739133048758,
      "flos": 20011042131840.0,
      "grad_norm": 2.1616682903863347,
      "language_loss": 0.80524409,
      "learning_rate": 1.7915077099393594e-07,
      "loss": 0.82672364,
      "num_input_tokens_seen": 156222860,
      "step": 7221,
      "time_per_iteration": 4.495007514953613
    },
    {
      "auxiliary_loss_clip": 0.01123536,
      "auxiliary_loss_mlp": 0.00762017,
      "balance_loss_clip": 1.04339337,
      "balance_loss_mlp": 1.0004766,
      "epoch": 0.868394156195515,
      "flos": 16654759217280.0,
      "grad_norm": 8.585736607159582,
      "language_loss": 0.73386389,
      "learning_rate": 1.788286680018083e-07,
      "loss": 0.7527194,
      "num_input_tokens_seen": 156241570,
      "step": 7222,
      "time_per_iteration": 3.5765323638916016
    },
    {
      "auxiliary_loss_clip": 0.01107079,
      "auxiliary_loss_mlp": 0.01021765,
      "balance_loss_clip": 1.04293752,
      "balance_loss_mlp": 1.01434422,
      "epoch": 0.868514399086154,
      "flos": 28001381448960.0,
      "grad_norm": 1.7025420288354558,
      "language_loss": 0.72829324,
      "learning_rate": 1.7850684127992443e-07,
      "loss": 0.74958163,
      "num_input_tokens_seen": 156261315,
      "step": 7223,
      "time_per_iteration": 2.8025712966918945
    },
    {
      "auxiliary_loss_clip": 0.0114075,
      "auxiliary_loss_mlp": 0.01026896,
      "balance_loss_clip": 1.04784083,
      "balance_loss_mlp": 1.01938629,
      "epoch": 0.8686346419767931,
      "flos": 20084587228800.0,
      "grad_norm": 1.5520750732033408,
      "language_loss": 0.70613718,
      "learning_rate": 1.7818529087710378e-07,
      "loss": 0.7278136,
      "num_input_tokens_seen": 156281670,
      "step": 7224,
      "time_per_iteration": 2.998013496398926
    },
    {
      "auxiliary_loss_clip": 0.01090885,
      "auxiliary_loss_mlp": 0.01022905,
      "balance_loss_clip": 1.03774357,
      "balance_loss_mlp": 1.01526403,
      "epoch": 0.8687548848674322,
      "flos": 18223516782720.0,
      "grad_norm": 1.7543159014597005,
      "language_loss": 0.8408283,
      "learning_rate": 1.7786401684212637e-07,
      "loss": 0.86196625,
      "num_input_tokens_seen": 156300500,
      "step": 7225,
      "time_per_iteration": 2.993438720703125
    },
    {
      "auxiliary_loss_clip": 0.01011981,
      "auxiliary_loss_mlp": 0.0075406,
      "balance_loss_clip": 1.01067185,
      "balance_loss_mlp": 1.00087285,
      "epoch": 0.8688751277580713,
      "flos": 70457885049600.0,
      "grad_norm": 0.730852854859007,
      "language_loss": 0.55885053,
      "learning_rate": 1.7754301922372883e-07,
      "loss": 0.57651091,
      "num_input_tokens_seen": 156350145,
      "step": 7226,
      "time_per_iteration": 3.8472585678100586
    },
    {
      "auxiliary_loss_clip": 0.01141281,
      "auxiliary_loss_mlp": 0.01026102,
      "balance_loss_clip": 1.04221153,
      "balance_loss_mlp": 1.0180912,
      "epoch": 0.8689953706487104,
      "flos": 26906788344960.0,
      "grad_norm": 2.3215861248499747,
      "language_loss": 0.81149924,
      "learning_rate": 1.7722229807060617e-07,
      "loss": 0.83317316,
      "num_input_tokens_seen": 156368725,
      "step": 7227,
      "time_per_iteration": 2.9589951038360596
    },
    {
      "auxiliary_loss_clip": 0.01102092,
      "auxiliary_loss_mlp": 0.0102576,
      "balance_loss_clip": 1.0425241,
      "balance_loss_mlp": 1.01812172,
      "epoch": 0.8691156135393495,
      "flos": 34637385438720.0,
      "grad_norm": 3.1489609943976076,
      "language_loss": 0.81650746,
      "learning_rate": 1.7690185343141172e-07,
      "loss": 0.83778596,
      "num_input_tokens_seen": 156388640,
      "step": 7228,
      "time_per_iteration": 2.9889941215515137
    },
    {
      "auxiliary_loss_clip": 0.01145375,
      "auxiliary_loss_mlp": 0.01019501,
      "balance_loss_clip": 1.04409766,
      "balance_loss_mlp": 1.01242256,
      "epoch": 0.8692358564299886,
      "flos": 18989814556800.0,
      "grad_norm": 2.092000770903263,
      "language_loss": 0.70090008,
      "learning_rate": 1.7658168535475615e-07,
      "loss": 0.72254884,
      "num_input_tokens_seen": 156406425,
      "step": 7229,
      "time_per_iteration": 2.680588483810425
    },
    {
      "auxiliary_loss_clip": 0.01169254,
      "auxiliary_loss_mlp": 0.01026423,
      "balance_loss_clip": 1.04807973,
      "balance_loss_mlp": 1.01839423,
      "epoch": 0.8693560993206276,
      "flos": 30370839039360.0,
      "grad_norm": 1.838963730565214,
      "language_loss": 0.64706963,
      "learning_rate": 1.7626179388920948e-07,
      "loss": 0.66902637,
      "num_input_tokens_seen": 156427705,
      "step": 7230,
      "time_per_iteration": 2.693765640258789
    },
    {
      "auxiliary_loss_clip": 0.01153926,
      "auxiliary_loss_mlp": 0.01026275,
      "balance_loss_clip": 1.04375207,
      "balance_loss_mlp": 1.01875925,
      "epoch": 0.8694763422112668,
      "flos": 27200430028800.0,
      "grad_norm": 1.9331387099899608,
      "language_loss": 0.80786479,
      "learning_rate": 1.7594217908329866e-07,
      "loss": 0.82966685,
      "num_input_tokens_seen": 156449890,
      "step": 7231,
      "time_per_iteration": 2.6841061115264893
    },
    {
      "auxiliary_loss_clip": 0.01138969,
      "auxiliary_loss_mlp": 0.00761745,
      "balance_loss_clip": 1.04533172,
      "balance_loss_mlp": 1.00050116,
      "epoch": 0.8695965851019059,
      "flos": 26139161767680.0,
      "grad_norm": 1.9225099755031438,
      "language_loss": 0.73927593,
      "learning_rate": 1.7562284098550895e-07,
      "loss": 0.75828308,
      "num_input_tokens_seen": 156469600,
      "step": 7232,
      "time_per_iteration": 3.6294045448303223
    },
    {
      "auxiliary_loss_clip": 0.01027973,
      "auxiliary_loss_mlp": 0.01000705,
      "balance_loss_clip": 1.01012886,
      "balance_loss_mlp": 0.99970323,
      "epoch": 0.8697168279925449,
      "flos": 67332616456320.0,
      "grad_norm": 0.8324334957170239,
      "language_loss": 0.62173152,
      "learning_rate": 1.753037796442838e-07,
      "loss": 0.64201826,
      "num_input_tokens_seen": 156529040,
      "step": 7233,
      "time_per_iteration": 3.266505002975464
    },
    {
      "auxiliary_loss_clip": 0.0111914,
      "auxiliary_loss_mlp": 0.01026331,
      "balance_loss_clip": 1.0388608,
      "balance_loss_mlp": 1.01898754,
      "epoch": 0.8698370708831841,
      "flos": 19718693337600.0,
      "grad_norm": 2.7518355420985454,
      "language_loss": 0.75164855,
      "learning_rate": 1.74984995108024e-07,
      "loss": 0.77310324,
      "num_input_tokens_seen": 156546970,
      "step": 7234,
      "time_per_iteration": 2.6913154125213623
    },
    {
      "auxiliary_loss_clip": 0.01152967,
      "auxiliary_loss_mlp": 0.01025054,
      "balance_loss_clip": 1.04567242,
      "balance_loss_mlp": 1.01768351,
      "epoch": 0.8699573137738231,
      "flos": 12859971068160.0,
      "grad_norm": 2.156289194653164,
      "language_loss": 0.83451569,
      "learning_rate": 1.7466648742508981e-07,
      "loss": 0.85629594,
      "num_input_tokens_seen": 156563155,
      "step": 7235,
      "time_per_iteration": 2.6789438724517822
    },
    {
      "auxiliary_loss_clip": 0.01136151,
      "auxiliary_loss_mlp": 0.01023358,
      "balance_loss_clip": 1.04355216,
      "balance_loss_mlp": 1.01567459,
      "epoch": 0.8700775566644622,
      "flos": 17420733768960.0,
      "grad_norm": 2.2167681741311434,
      "language_loss": 0.84523487,
      "learning_rate": 1.7434825664379837e-07,
      "loss": 0.86682993,
      "num_input_tokens_seen": 156581660,
      "step": 7236,
      "time_per_iteration": 2.7064096927642822
    },
    {
      "auxiliary_loss_clip": 0.01126495,
      "auxiliary_loss_mlp": 0.01027853,
      "balance_loss_clip": 1.04161859,
      "balance_loss_mlp": 1.01972938,
      "epoch": 0.8701977995551013,
      "flos": 13735221770880.0,
      "grad_norm": 3.088786035717831,
      "language_loss": 0.86125958,
      "learning_rate": 1.740303028124246e-07,
      "loss": 0.88280302,
      "num_input_tokens_seen": 156597720,
      "step": 7237,
      "time_per_iteration": 2.6889588832855225
    },
    {
      "auxiliary_loss_clip": 0.01165609,
      "auxiliary_loss_mlp": 0.01026274,
      "balance_loss_clip": 1.04579413,
      "balance_loss_mlp": 1.01887989,
      "epoch": 0.8703180424457404,
      "flos": 30555707362560.0,
      "grad_norm": 1.9886339657826184,
      "language_loss": 0.75734377,
      "learning_rate": 1.7371262597920212e-07,
      "loss": 0.77926266,
      "num_input_tokens_seen": 156619780,
      "step": 7238,
      "time_per_iteration": 2.7349700927734375
    },
    {
      "auxiliary_loss_clip": 0.01120722,
      "auxiliary_loss_mlp": 0.01024658,
      "balance_loss_clip": 1.04249525,
      "balance_loss_mlp": 1.01727915,
      "epoch": 0.8704382853363795,
      "flos": 19608986223360.0,
      "grad_norm": 1.5695122070761849,
      "language_loss": 0.76221871,
      "learning_rate": 1.7339522619232195e-07,
      "loss": 0.78367245,
      "num_input_tokens_seen": 156638160,
      "step": 7239,
      "time_per_iteration": 2.6915223598480225
    },
    {
      "auxiliary_loss_clip": 0.01140046,
      "auxiliary_loss_mlp": 0.01022814,
      "balance_loss_clip": 1.0420903,
      "balance_loss_mlp": 1.01510167,
      "epoch": 0.8705585282270186,
      "flos": 26613900846720.0,
      "grad_norm": 4.0904711099482345,
      "language_loss": 0.75618583,
      "learning_rate": 1.730781034999338e-07,
      "loss": 0.77781445,
      "num_input_tokens_seen": 156659740,
      "step": 7240,
      "time_per_iteration": 2.6930747032165527
    },
    {
      "auxiliary_loss_clip": 0.01100167,
      "auxiliary_loss_mlp": 0.0102552,
      "balance_loss_clip": 1.04025984,
      "balance_loss_mlp": 1.01826048,
      "epoch": 0.8706787711176577,
      "flos": 34090465979520.0,
      "grad_norm": 7.509444548536622,
      "language_loss": 0.73460257,
      "learning_rate": 1.7276125795014497e-07,
      "loss": 0.75585938,
      "num_input_tokens_seen": 156678190,
      "step": 7241,
      "time_per_iteration": 2.882622003555298
    },
    {
      "auxiliary_loss_clip": 0.01173006,
      "auxiliary_loss_mlp": 0.01024078,
      "balance_loss_clip": 1.04817867,
      "balance_loss_mlp": 1.0159632,
      "epoch": 0.8707990140082967,
      "flos": 14611513968000.0,
      "grad_norm": 2.249707685984987,
      "language_loss": 0.67452925,
      "learning_rate": 1.7244468959102054e-07,
      "loss": 0.69650006,
      "num_input_tokens_seen": 156695245,
      "step": 7242,
      "time_per_iteration": 2.590306282043457
    },
    {
      "auxiliary_loss_clip": 0.01097773,
      "auxiliary_loss_mlp": 0.01029828,
      "balance_loss_clip": 1.04415905,
      "balance_loss_mlp": 1.0219903,
      "epoch": 0.8709192568989359,
      "flos": 20084156265600.0,
      "grad_norm": 2.050875249092203,
      "language_loss": 0.8514576,
      "learning_rate": 1.7212839847058348e-07,
      "loss": 0.87273359,
      "num_input_tokens_seen": 156710375,
      "step": 7243,
      "time_per_iteration": 2.7854092121124268
    },
    {
      "auxiliary_loss_clip": 0.01067446,
      "auxiliary_loss_mlp": 0.01023709,
      "balance_loss_clip": 1.0376668,
      "balance_loss_mlp": 1.01568031,
      "epoch": 0.871039499789575,
      "flos": 16727083251840.0,
      "grad_norm": 1.8493509601593114,
      "language_loss": 0.73646259,
      "learning_rate": 1.718123846368147e-07,
      "loss": 0.75737411,
      "num_input_tokens_seen": 156729420,
      "step": 7244,
      "time_per_iteration": 2.9370787143707275
    },
    {
      "auxiliary_loss_clip": 0.01139032,
      "auxiliary_loss_mlp": 0.01017699,
      "balance_loss_clip": 1.04294729,
      "balance_loss_mlp": 1.01085031,
      "epoch": 0.871159742680214,
      "flos": 21068790860160.0,
      "grad_norm": 2.062124106845365,
      "language_loss": 0.71549946,
      "learning_rate": 1.714966481376543e-07,
      "loss": 0.73706675,
      "num_input_tokens_seen": 156746100,
      "step": 7245,
      "time_per_iteration": 3.3572351932525635
    },
    {
      "auxiliary_loss_clip": 0.01127578,
      "auxiliary_loss_mlp": 0.01023769,
      "balance_loss_clip": 1.04286051,
      "balance_loss_mlp": 1.01620471,
      "epoch": 0.8712799855708532,
      "flos": 28256526731520.0,
      "grad_norm": 2.0757486383875907,
      "language_loss": 0.83112383,
      "learning_rate": 1.7118118902099797e-07,
      "loss": 0.85263729,
      "num_input_tokens_seen": 156764185,
      "step": 7246,
      "time_per_iteration": 4.743350505828857
    },
    {
      "auxiliary_loss_clip": 0.01141874,
      "auxiliary_loss_mlp": 0.01026338,
      "balance_loss_clip": 1.04666495,
      "balance_loss_mlp": 1.01869988,
      "epoch": 0.8714002284614922,
      "flos": 22236677665920.0,
      "grad_norm": 1.66785400116762,
      "language_loss": 0.80654055,
      "learning_rate": 1.7086600733470146e-07,
      "loss": 0.82822263,
      "num_input_tokens_seen": 156784855,
      "step": 7247,
      "time_per_iteration": 2.773808002471924
    },
    {
      "auxiliary_loss_clip": 0.01104628,
      "auxiliary_loss_mlp": 0.00761821,
      "balance_loss_clip": 1.03982902,
      "balance_loss_mlp": 1.00049746,
      "epoch": 0.8715204713521313,
      "flos": 21431919404160.0,
      "grad_norm": 7.832093146231902,
      "language_loss": 0.77088332,
      "learning_rate": 1.7055110312657738e-07,
      "loss": 0.7895478,
      "num_input_tokens_seen": 156804350,
      "step": 7248,
      "time_per_iteration": 3.718684434890747
    },
    {
      "auxiliary_loss_clip": 0.01114536,
      "auxiliary_loss_mlp": 0.0102598,
      "balance_loss_clip": 1.04485369,
      "balance_loss_mlp": 1.01774228,
      "epoch": 0.8716407142427703,
      "flos": 23440439180160.0,
      "grad_norm": 2.1040019702512245,
      "language_loss": 0.74121827,
      "learning_rate": 1.702364764443962e-07,
      "loss": 0.76262343,
      "num_input_tokens_seen": 156823425,
      "step": 7249,
      "time_per_iteration": 2.723656415939331
    },
    {
      "auxiliary_loss_clip": 0.01037919,
      "auxiliary_loss_mlp": 0.01027053,
      "balance_loss_clip": 1.03265476,
      "balance_loss_mlp": 1.01933169,
      "epoch": 0.8717609571334095,
      "flos": 27958683156480.0,
      "grad_norm": 2.2664333753099464,
      "language_loss": 0.72049016,
      "learning_rate": 1.6992212733588685e-07,
      "loss": 0.74113989,
      "num_input_tokens_seen": 156843090,
      "step": 7250,
      "time_per_iteration": 3.326286554336548
    },
    {
      "auxiliary_loss_clip": 0.01090996,
      "auxiliary_loss_mlp": 0.01024788,
      "balance_loss_clip": 1.03565836,
      "balance_loss_mlp": 1.01745629,
      "epoch": 0.8718812000240486,
      "flos": 25479482538240.0,
      "grad_norm": 2.136881372354235,
      "language_loss": 0.75040168,
      "learning_rate": 1.6960805584873538e-07,
      "loss": 0.77155954,
      "num_input_tokens_seen": 156861090,
      "step": 7251,
      "time_per_iteration": 3.0239524841308594
    },
    {
      "auxiliary_loss_clip": 0.01083414,
      "auxiliary_loss_mlp": 0.01023454,
      "balance_loss_clip": 1.03471482,
      "balance_loss_mlp": 1.01531208,
      "epoch": 0.8720014429146876,
      "flos": 23403056100480.0,
      "grad_norm": 1.5841678253222988,
      "language_loss": 0.78317976,
      "learning_rate": 1.6929426203058684e-07,
      "loss": 0.80424839,
      "num_input_tokens_seen": 156881515,
      "step": 7252,
      "time_per_iteration": 2.771381378173828
    },
    {
      "auxiliary_loss_clip": 0.01110242,
      "auxiliary_loss_mlp": 0.01025885,
      "balance_loss_clip": 1.03723967,
      "balance_loss_mlp": 1.01776648,
      "epoch": 0.8721216858053268,
      "flos": 24352821567360.0,
      "grad_norm": 2.4036263226119012,
      "language_loss": 0.79880661,
      "learning_rate": 1.689807459290431e-07,
      "loss": 0.8201679,
      "num_input_tokens_seen": 156900170,
      "step": 7253,
      "time_per_iteration": 2.7555882930755615
    },
    {
      "auxiliary_loss_clip": 0.01145225,
      "auxiliary_loss_mlp": 0.01023786,
      "balance_loss_clip": 1.04608035,
      "balance_loss_mlp": 1.01596618,
      "epoch": 0.8722419286959658,
      "flos": 33869687034240.0,
      "grad_norm": 2.0968224630014554,
      "language_loss": 0.71497756,
      "learning_rate": 1.6866750759166437e-07,
      "loss": 0.73666763,
      "num_input_tokens_seen": 156920150,
      "step": 7254,
      "time_per_iteration": 2.8235180377960205
    },
    {
      "auxiliary_loss_clip": 0.01087257,
      "auxiliary_loss_mlp": 0.01024629,
      "balance_loss_clip": 1.03701401,
      "balance_loss_mlp": 1.01731539,
      "epoch": 0.8723621715866049,
      "flos": 18369385914240.0,
      "grad_norm": 2.3934911534101713,
      "language_loss": 0.77111322,
      "learning_rate": 1.6835454706596865e-07,
      "loss": 0.79223204,
      "num_input_tokens_seen": 156937980,
      "step": 7255,
      "time_per_iteration": 2.7982428073883057
    },
    {
      "auxiliary_loss_clip": 0.01101797,
      "auxiliary_loss_mlp": 0.0102299,
      "balance_loss_clip": 1.04173625,
      "balance_loss_mlp": 1.01507473,
      "epoch": 0.8724824144772441,
      "flos": 22013348855040.0,
      "grad_norm": 1.6713286889247942,
      "language_loss": 0.73802352,
      "learning_rate": 1.680418643994317e-07,
      "loss": 0.75927138,
      "num_input_tokens_seen": 156956550,
      "step": 7256,
      "time_per_iteration": 2.7575371265411377
    },
    {
      "auxiliary_loss_clip": 0.01048928,
      "auxiliary_loss_mlp": 0.01001055,
      "balance_loss_clip": 1.01035893,
      "balance_loss_mlp": 0.99996436,
      "epoch": 0.8726026573678831,
      "flos": 66698720213760.0,
      "grad_norm": 0.8833683612697615,
      "language_loss": 0.64482772,
      "learning_rate": 1.6772945963948738e-07,
      "loss": 0.66532755,
      "num_input_tokens_seen": 157014715,
      "step": 7257,
      "time_per_iteration": 4.195022344589233
    },
    {
      "auxiliary_loss_clip": 0.01091682,
      "auxiliary_loss_mlp": 0.01023064,
      "balance_loss_clip": 1.03782678,
      "balance_loss_mlp": 1.01502025,
      "epoch": 0.8727229002585222,
      "flos": 13370908078080.0,
      "grad_norm": 2.513592002559813,
      "language_loss": 0.77223814,
      "learning_rate": 1.6741733283352733e-07,
      "loss": 0.79338557,
      "num_input_tokens_seen": 157032320,
      "step": 7258,
      "time_per_iteration": 2.806201219558716
    },
    {
      "auxiliary_loss_clip": 0.0107097,
      "auxiliary_loss_mlp": 0.01023562,
      "balance_loss_clip": 1.03392434,
      "balance_loss_mlp": 1.01595044,
      "epoch": 0.8728431431491613,
      "flos": 21796987282560.0,
      "grad_norm": 1.6708719520076556,
      "language_loss": 0.83915555,
      "learning_rate": 1.6710548402890102e-07,
      "loss": 0.86010087,
      "num_input_tokens_seen": 157052845,
      "step": 7259,
      "time_per_iteration": 2.8398051261901855
    },
    {
      "auxiliary_loss_clip": 0.01132151,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.04521275,
      "balance_loss_mlp": 1.02071214,
      "epoch": 0.8729633860398004,
      "flos": 36173823742080.0,
      "grad_norm": 2.410179284079536,
      "language_loss": 0.66768312,
      "learning_rate": 1.6679391327291527e-07,
      "loss": 0.68928975,
      "num_input_tokens_seen": 157074050,
      "step": 7260,
      "time_per_iteration": 2.851431131362915
    },
    {
      "auxiliary_loss_clip": 0.01126505,
      "auxiliary_loss_mlp": 0.01026061,
      "balance_loss_clip": 1.04315662,
      "balance_loss_mlp": 1.01812768,
      "epoch": 0.8730836289304394,
      "flos": 16359680989440.0,
      "grad_norm": 3.225240307644639,
      "language_loss": 0.68419242,
      "learning_rate": 1.6648262061283492e-07,
      "loss": 0.7057181,
      "num_input_tokens_seen": 157089350,
      "step": 7261,
      "time_per_iteration": 2.6889865398406982
    },
    {
      "auxiliary_loss_clip": 0.01135525,
      "auxiliary_loss_mlp": 0.01024544,
      "balance_loss_clip": 1.04192615,
      "balance_loss_mlp": 1.01718247,
      "epoch": 0.8732038718210786,
      "flos": 21215126868480.0,
      "grad_norm": 2.0844140276741054,
      "language_loss": 0.73850673,
      "learning_rate": 1.6617160609588353e-07,
      "loss": 0.7601074,
      "num_input_tokens_seen": 157108525,
      "step": 7262,
      "time_per_iteration": 2.734233856201172
    },
    {
      "auxiliary_loss_clip": 0.0115192,
      "auxiliary_loss_mlp": 0.01028172,
      "balance_loss_clip": 1.04354513,
      "balance_loss_mlp": 1.02000046,
      "epoch": 0.8733241147117177,
      "flos": 16610696208000.0,
      "grad_norm": 2.308918238962825,
      "language_loss": 0.72199762,
      "learning_rate": 1.6586086976924163e-07,
      "loss": 0.74379849,
      "num_input_tokens_seen": 157124025,
      "step": 7263,
      "time_per_iteration": 2.622313976287842
    },
    {
      "auxiliary_loss_clip": 0.01119534,
      "auxiliary_loss_mlp": 0.00761843,
      "balance_loss_clip": 1.04141879,
      "balance_loss_mlp": 1.00048852,
      "epoch": 0.8734443576023567,
      "flos": 20193935207040.0,
      "grad_norm": 1.8216865863784293,
      "language_loss": 0.78424656,
      "learning_rate": 1.6555041168004747e-07,
      "loss": 0.80306035,
      "num_input_tokens_seen": 157143345,
      "step": 7264,
      "time_per_iteration": 2.759918689727783
    },
    {
      "auxiliary_loss_clip": 0.01102219,
      "auxiliary_loss_mlp": 0.01026772,
      "balance_loss_clip": 1.04227805,
      "balance_loss_mlp": 1.01928544,
      "epoch": 0.8735646004929959,
      "flos": 18041162411520.0,
      "grad_norm": 2.964073268756133,
      "language_loss": 0.69476438,
      "learning_rate": 1.6524023187539715e-07,
      "loss": 0.71605432,
      "num_input_tokens_seen": 157161630,
      "step": 7265,
      "time_per_iteration": 2.7679309844970703
    },
    {
      "auxiliary_loss_clip": 0.01157356,
      "auxiliary_loss_mlp": 0.01022657,
      "balance_loss_clip": 1.047894,
      "balance_loss_mlp": 1.01521873,
      "epoch": 0.873684843383635,
      "flos": 20262344659200.0,
      "grad_norm": 1.8353761477381134,
      "language_loss": 0.74898994,
      "learning_rate": 1.649303304023446e-07,
      "loss": 0.77079016,
      "num_input_tokens_seen": 157181385,
      "step": 7266,
      "time_per_iteration": 2.6454544067382812
    },
    {
      "auxiliary_loss_clip": 0.01109577,
      "auxiliary_loss_mlp": 0.01024558,
      "balance_loss_clip": 1.04034138,
      "balance_loss_mlp": 1.0175457,
      "epoch": 0.873805086274274,
      "flos": 16947287579520.0,
      "grad_norm": 1.898189251451628,
      "language_loss": 0.78380609,
      "learning_rate": 1.6462070730790246e-07,
      "loss": 0.80514741,
      "num_input_tokens_seen": 157200545,
      "step": 7267,
      "time_per_iteration": 2.80759859085083
    },
    {
      "auxiliary_loss_clip": 0.01139432,
      "auxiliary_loss_mlp": 0.01028345,
      "balance_loss_clip": 1.04268324,
      "balance_loss_mlp": 1.02098942,
      "epoch": 0.8739253291649132,
      "flos": 18041270152320.0,
      "grad_norm": 2.864293653214884,
      "language_loss": 0.78085369,
      "learning_rate": 1.6431136263903912e-07,
      "loss": 0.80253142,
      "num_input_tokens_seen": 157219545,
      "step": 7268,
      "time_per_iteration": 2.711085319519043
    },
    {
      "auxiliary_loss_clip": 0.01153778,
      "auxiliary_loss_mlp": 0.01019856,
      "balance_loss_clip": 1.04517508,
      "balance_loss_mlp": 1.01216125,
      "epoch": 0.8740455720555522,
      "flos": 21325085377920.0,
      "grad_norm": 2.6073916093568874,
      "language_loss": 0.73984611,
      "learning_rate": 1.6400229644268282e-07,
      "loss": 0.76158243,
      "num_input_tokens_seen": 157237900,
      "step": 7269,
      "time_per_iteration": 2.6995654106140137
    },
    {
      "auxiliary_loss_clip": 0.01102117,
      "auxiliary_loss_mlp": 0.01025862,
      "balance_loss_clip": 1.04082191,
      "balance_loss_mlp": 1.01855469,
      "epoch": 0.8741658149461913,
      "flos": 15158684822400.0,
      "grad_norm": 2.13139428132183,
      "language_loss": 0.81245625,
      "learning_rate": 1.6369350876571852e-07,
      "loss": 0.83373606,
      "num_input_tokens_seen": 157256055,
      "step": 7270,
      "time_per_iteration": 2.726041316986084
    },
    {
      "auxiliary_loss_clip": 0.01168376,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 1.04674768,
      "balance_loss_mlp": 1.01715684,
      "epoch": 0.8742860578368304,
      "flos": 23039855729280.0,
      "grad_norm": 2.4695500657923377,
      "language_loss": 0.81519985,
      "learning_rate": 1.6338499965498874e-07,
      "loss": 0.8371287,
      "num_input_tokens_seen": 157274785,
      "step": 7271,
      "time_per_iteration": 2.6517975330352783
    },
    {
      "auxiliary_loss_clip": 0.01158513,
      "auxiliary_loss_mlp": 0.01026175,
      "balance_loss_clip": 1.04685056,
      "balance_loss_mlp": 1.01748478,
      "epoch": 0.8744063007274695,
      "flos": 28145347159680.0,
      "grad_norm": 1.5498920350067478,
      "language_loss": 0.77647889,
      "learning_rate": 1.630767691572943e-07,
      "loss": 0.79832572,
      "num_input_tokens_seen": 157294805,
      "step": 7272,
      "time_per_iteration": 5.239895820617676
    },
    {
      "auxiliary_loss_clip": 0.01066037,
      "auxiliary_loss_mlp": 0.01000659,
      "balance_loss_clip": 1.01009715,
      "balance_loss_mlp": 0.99965811,
      "epoch": 0.8745265436181086,
      "flos": 64034076654720.0,
      "grad_norm": 0.7390514130209409,
      "language_loss": 0.53468621,
      "learning_rate": 1.6276881731939306e-07,
      "loss": 0.55535322,
      "num_input_tokens_seen": 157356695,
      "step": 7273,
      "time_per_iteration": 4.320552825927734
    },
    {
      "auxiliary_loss_clip": 0.01136696,
      "auxiliary_loss_mlp": 0.0102105,
      "balance_loss_clip": 1.04360855,
      "balance_loss_mlp": 1.01419234,
      "epoch": 0.8746467865087477,
      "flos": 28658618553600.0,
      "grad_norm": 2.386568415781939,
      "language_loss": 0.75442201,
      "learning_rate": 1.6246114418800193e-07,
      "loss": 0.77599943,
      "num_input_tokens_seen": 157376975,
      "step": 7274,
      "time_per_iteration": 2.7379255294799805
    },
    {
      "auxiliary_loss_clip": 0.01141493,
      "auxiliary_loss_mlp": 0.01026537,
      "balance_loss_clip": 1.04385281,
      "balance_loss_mlp": 1.01884842,
      "epoch": 0.8747670293993868,
      "flos": 23985850268160.0,
      "grad_norm": 42.070719420652836,
      "language_loss": 0.76584899,
      "learning_rate": 1.6215374980979423e-07,
      "loss": 0.78752941,
      "num_input_tokens_seen": 157397385,
      "step": 7275,
      "time_per_iteration": 2.837275981903076
    },
    {
      "auxiliary_loss_clip": 0.01133181,
      "auxiliary_loss_mlp": 0.01026226,
      "balance_loss_clip": 1.04266429,
      "balance_loss_mlp": 1.01898694,
      "epoch": 0.8748872722900258,
      "flos": 45221624478720.0,
      "grad_norm": 2.7098850123892286,
      "language_loss": 0.69241214,
      "learning_rate": 1.6184663423140133e-07,
      "loss": 0.71400619,
      "num_input_tokens_seen": 157417685,
      "step": 7276,
      "time_per_iteration": 2.8444337844848633
    },
    {
      "auxiliary_loss_clip": 0.0112304,
      "auxiliary_loss_mlp": 0.01024041,
      "balance_loss_clip": 1.04429877,
      "balance_loss_mlp": 1.01644158,
      "epoch": 0.875007515180665,
      "flos": 19754280737280.0,
      "grad_norm": 2.181183036779118,
      "language_loss": 0.64052975,
      "learning_rate": 1.615397974994126e-07,
      "loss": 0.66200054,
      "num_input_tokens_seen": 157435490,
      "step": 7277,
      "time_per_iteration": 2.720634937286377
    },
    {
      "auxiliary_loss_clip": 0.01094093,
      "auxiliary_loss_mlp": 0.01023312,
      "balance_loss_clip": 1.03859532,
      "balance_loss_mlp": 1.01654053,
      "epoch": 0.875127758071304,
      "flos": 22710734386560.0,
      "grad_norm": 1.5075090145289671,
      "language_loss": 0.80915624,
      "learning_rate": 1.6123323966037438e-07,
      "loss": 0.83033025,
      "num_input_tokens_seen": 157454010,
      "step": 7278,
      "time_per_iteration": 2.8722586631774902
    },
    {
      "auxiliary_loss_clip": 0.01084329,
      "auxiliary_loss_mlp": 0.01027295,
      "balance_loss_clip": 1.03795409,
      "balance_loss_mlp": 1.01887918,
      "epoch": 0.8752480009619431,
      "flos": 23403846199680.0,
      "grad_norm": 2.2514088460278954,
      "language_loss": 0.78924406,
      "learning_rate": 1.6092696076079216e-07,
      "loss": 0.81036031,
      "num_input_tokens_seen": 157472385,
      "step": 7279,
      "time_per_iteration": 2.8390023708343506
    },
    {
      "auxiliary_loss_clip": 0.0112438,
      "auxiliary_loss_mlp": 0.01020589,
      "balance_loss_clip": 1.03950572,
      "balance_loss_mlp": 1.01280463,
      "epoch": 0.8753682438525822,
      "flos": 26213101914240.0,
      "grad_norm": 2.207035121509743,
      "language_loss": 0.73666298,
      "learning_rate": 1.6062096084712785e-07,
      "loss": 0.75811267,
      "num_input_tokens_seen": 157493735,
      "step": 7280,
      "time_per_iteration": 2.7305099964141846
    },
    {
      "auxiliary_loss_clip": 0.01119529,
      "auxiliary_loss_mlp": 0.01025271,
      "balance_loss_clip": 1.04368711,
      "balance_loss_mlp": 1.01733804,
      "epoch": 0.8754884867432213,
      "flos": 23326745656320.0,
      "grad_norm": 2.0691366363966663,
      "language_loss": 0.70653576,
      "learning_rate": 1.6031523996580098e-07,
      "loss": 0.72798383,
      "num_input_tokens_seen": 157511295,
      "step": 7281,
      "time_per_iteration": 2.7583022117614746
    },
    {
      "auxiliary_loss_clip": 0.01058666,
      "auxiliary_loss_mlp": 0.01027725,
      "balance_loss_clip": 1.03369379,
      "balance_loss_mlp": 1.01932025,
      "epoch": 0.8756087296338604,
      "flos": 12495226412160.0,
      "grad_norm": 2.474381082707362,
      "language_loss": 0.66888469,
      "learning_rate": 1.6000979816318981e-07,
      "loss": 0.68974859,
      "num_input_tokens_seen": 157529760,
      "step": 7282,
      "time_per_iteration": 2.9247190952301025
    },
    {
      "auxiliary_loss_clip": 0.01093975,
      "auxiliary_loss_mlp": 0.01025347,
      "balance_loss_clip": 1.03835893,
      "balance_loss_mlp": 1.01783109,
      "epoch": 0.8757289725244994,
      "flos": 18952898353920.0,
      "grad_norm": 2.3214887646214595,
      "language_loss": 0.74997151,
      "learning_rate": 1.5970463548562886e-07,
      "loss": 0.77116472,
      "num_input_tokens_seen": 157548915,
      "step": 7283,
      "time_per_iteration": 4.275784730911255
    },
    {
      "auxiliary_loss_clip": 0.01086263,
      "auxiliary_loss_mlp": 0.01023264,
      "balance_loss_clip": 1.03863287,
      "balance_loss_mlp": 1.01527715,
      "epoch": 0.8758492154151386,
      "flos": 25265958140160.0,
      "grad_norm": 1.7313984816761134,
      "language_loss": 0.7106213,
      "learning_rate": 1.5939975197941192e-07,
      "loss": 0.73171657,
      "num_input_tokens_seen": 157570570,
      "step": 7284,
      "time_per_iteration": 2.8391706943511963
    },
    {
      "auxiliary_loss_clip": 0.01032377,
      "auxiliary_loss_mlp": 0.01002146,
      "balance_loss_clip": 1.01017451,
      "balance_loss_mlp": 1.00116861,
      "epoch": 0.8759694583057777,
      "flos": 65571664193280.0,
      "grad_norm": 0.8066048531750625,
      "language_loss": 0.53318834,
      "learning_rate": 1.5909514769078892e-07,
      "loss": 0.55353355,
      "num_input_tokens_seen": 157635675,
      "step": 7285,
      "time_per_iteration": 3.3697102069854736
    },
    {
      "auxiliary_loss_clip": 0.01127441,
      "auxiliary_loss_mlp": 0.01021699,
      "balance_loss_clip": 1.04406452,
      "balance_loss_mlp": 1.01418614,
      "epoch": 0.8760897011964167,
      "flos": 25446193608960.0,
      "grad_norm": 1.7471402884298641,
      "language_loss": 0.77754474,
      "learning_rate": 1.5879082266596867e-07,
      "loss": 0.79903615,
      "num_input_tokens_seen": 157657015,
      "step": 7286,
      "time_per_iteration": 2.7627763748168945
    },
    {
      "auxiliary_loss_clip": 0.01103108,
      "auxiliary_loss_mlp": 0.01026889,
      "balance_loss_clip": 1.04076076,
      "balance_loss_mlp": 1.01937246,
      "epoch": 0.8762099440870559,
      "flos": 28984830894720.0,
      "grad_norm": 2.1461797194391363,
      "language_loss": 0.71803916,
      "learning_rate": 1.5848677695111645e-07,
      "loss": 0.73933911,
      "num_input_tokens_seen": 157678615,
      "step": 7287,
      "time_per_iteration": 2.8352551460266113
    },
    {
      "auxiliary_loss_clip": 0.01147452,
      "auxiliary_loss_mlp": 0.01027146,
      "balance_loss_clip": 1.04853594,
      "balance_loss_mlp": 1.0193851,
      "epoch": 0.8763301869776949,
      "flos": 21609461352960.0,
      "grad_norm": 2.550777734075566,
      "language_loss": 0.70044488,
      "learning_rate": 1.5818301059235562e-07,
      "loss": 0.72219086,
      "num_input_tokens_seen": 157693790,
      "step": 7288,
      "time_per_iteration": 2.7330989837646484
    },
    {
      "auxiliary_loss_clip": 0.01122634,
      "auxiliary_loss_mlp": 0.01025566,
      "balance_loss_clip": 1.0394032,
      "balance_loss_mlp": 1.01782382,
      "epoch": 0.876450429868334,
      "flos": 24644416176000.0,
      "grad_norm": 3.053042413921692,
      "language_loss": 0.81522679,
      "learning_rate": 1.578795236357684e-07,
      "loss": 0.83670878,
      "num_input_tokens_seen": 157715255,
      "step": 7289,
      "time_per_iteration": 2.889014482498169
    },
    {
      "auxiliary_loss_clip": 0.01121247,
      "auxiliary_loss_mlp": 0.00761931,
      "balance_loss_clip": 1.03983545,
      "balance_loss_mlp": 1.00043893,
      "epoch": 0.8765706727589732,
      "flos": 20260046188800.0,
      "grad_norm": 2.152356343739113,
      "language_loss": 0.85654306,
      "learning_rate": 1.5757631612739218e-07,
      "loss": 0.87537485,
      "num_input_tokens_seen": 157728800,
      "step": 7290,
      "time_per_iteration": 2.700352430343628
    },
    {
      "auxiliary_loss_clip": 0.01069339,
      "auxiliary_loss_mlp": 0.01016178,
      "balance_loss_clip": 1.02160382,
      "balance_loss_mlp": 1.01512277,
      "epoch": 0.8766909156496122,
      "flos": 71371165276800.0,
      "grad_norm": 0.7931752704828687,
      "language_loss": 0.6138097,
      "learning_rate": 1.572733881132242e-07,
      "loss": 0.63466489,
      "num_input_tokens_seen": 157789445,
      "step": 7291,
      "time_per_iteration": 3.2660322189331055
    },
    {
      "auxiliary_loss_clip": 0.01034889,
      "auxiliary_loss_mlp": 0.01001545,
      "balance_loss_clip": 1.00955856,
      "balance_loss_mlp": 1.00053179,
      "epoch": 0.8768111585402513,
      "flos": 69523490603520.0,
      "grad_norm": 0.7785225631342163,
      "language_loss": 0.58468139,
      "learning_rate": 1.5697073963921814e-07,
      "loss": 0.60504568,
      "num_input_tokens_seen": 157848685,
      "step": 7292,
      "time_per_iteration": 3.2136096954345703
    },
    {
      "auxiliary_loss_clip": 0.01137544,
      "auxiliary_loss_mlp": 0.01024567,
      "balance_loss_clip": 1.04591417,
      "balance_loss_mlp": 1.01607358,
      "epoch": 0.8769314014308904,
      "flos": 18838558385280.0,
      "grad_norm": 2.1713921516180372,
      "language_loss": 0.85122192,
      "learning_rate": 1.566683707512857e-07,
      "loss": 0.87284303,
      "num_input_tokens_seen": 157866360,
      "step": 7293,
      "time_per_iteration": 2.6553595066070557
    },
    {
      "auxiliary_loss_clip": 0.01118581,
      "auxiliary_loss_mlp": 0.01025333,
      "balance_loss_clip": 1.04351747,
      "balance_loss_mlp": 1.01723301,
      "epoch": 0.8770516443215295,
      "flos": 14976402278400.0,
      "grad_norm": 3.3524985646932666,
      "language_loss": 0.79456532,
      "learning_rate": 1.5636628149529553e-07,
      "loss": 0.81600446,
      "num_input_tokens_seen": 157884150,
      "step": 7294,
      "time_per_iteration": 2.794374465942383
    },
    {
      "auxiliary_loss_clip": 0.01127384,
      "auxiliary_loss_mlp": 0.01023394,
      "balance_loss_clip": 1.04218614,
      "balance_loss_mlp": 1.01619697,
      "epoch": 0.8771718872121685,
      "flos": 31649654021760.0,
      "grad_norm": 2.2857297983054035,
      "language_loss": 0.80000484,
      "learning_rate": 1.560644719170743e-07,
      "loss": 0.82151264,
      "num_input_tokens_seen": 157905020,
      "step": 7295,
      "time_per_iteration": 2.810645341873169
    },
    {
      "auxiliary_loss_clip": 0.01103414,
      "auxiliary_loss_mlp": 0.01025652,
      "balance_loss_clip": 1.03979516,
      "balance_loss_mlp": 1.01760507,
      "epoch": 0.8772921301028077,
      "flos": 36095466222720.0,
      "grad_norm": 2.320965703970583,
      "language_loss": 0.7202698,
      "learning_rate": 1.5576294206240692e-07,
      "loss": 0.74156046,
      "num_input_tokens_seen": 157924545,
      "step": 7296,
      "time_per_iteration": 2.9020519256591797
    },
    {
      "auxiliary_loss_clip": 0.01108076,
      "auxiliary_loss_mlp": 0.01023689,
      "balance_loss_clip": 1.03908634,
      "balance_loss_mlp": 1.01616108,
      "epoch": 0.8774123729934468,
      "flos": 57116961849600.0,
      "grad_norm": 1.6498390834374455,
      "language_loss": 0.67681366,
      "learning_rate": 1.5546169197703507e-07,
      "loss": 0.69813132,
      "num_input_tokens_seen": 157950820,
      "step": 7297,
      "time_per_iteration": 3.0889909267425537
    },
    {
      "auxiliary_loss_clip": 0.01112722,
      "auxiliary_loss_mlp": 0.01023229,
      "balance_loss_clip": 1.04074907,
      "balance_loss_mlp": 1.01596916,
      "epoch": 0.8775326158840858,
      "flos": 23914495900800.0,
      "grad_norm": 2.6910743892707605,
      "language_loss": 0.76909971,
      "learning_rate": 1.5516072170665774e-07,
      "loss": 0.79045928,
      "num_input_tokens_seen": 157968790,
      "step": 7298,
      "time_per_iteration": 4.6702680587768555
    },
    {
      "auxiliary_loss_clip": 0.01051397,
      "auxiliary_loss_mlp": 0.0076149,
      "balance_loss_clip": 1.03434074,
      "balance_loss_mlp": 1.00050116,
      "epoch": 0.877652858774725,
      "flos": 17123285243520.0,
      "grad_norm": 3.3497323183722174,
      "language_loss": 0.87165457,
      "learning_rate": 1.5486003129693214e-07,
      "loss": 0.8897835,
      "num_input_tokens_seen": 157986155,
      "step": 7299,
      "time_per_iteration": 4.026042699813843
    },
    {
      "auxiliary_loss_clip": 0.01126009,
      "auxiliary_loss_mlp": 0.007627,
      "balance_loss_clip": 1.04347634,
      "balance_loss_mlp": 1.00043106,
      "epoch": 0.877773101665364,
      "flos": 16508961912960.0,
      "grad_norm": 2.7992746212041664,
      "language_loss": 0.78291059,
      "learning_rate": 1.545596207934725e-07,
      "loss": 0.80179769,
      "num_input_tokens_seen": 158004640,
      "step": 7300,
      "time_per_iteration": 3.066917896270752
    },
    {
      "auxiliary_loss_clip": 0.01167168,
      "auxiliary_loss_mlp": 0.01024134,
      "balance_loss_clip": 1.04676878,
      "balance_loss_mlp": 1.01704729,
      "epoch": 0.8778933445560031,
      "flos": 22053209973120.0,
      "grad_norm": 1.6975650728725362,
      "language_loss": 0.78108907,
      "learning_rate": 1.5425949024185147e-07,
      "loss": 0.80300212,
      "num_input_tokens_seen": 158024665,
      "step": 7301,
      "time_per_iteration": 3.14042329788208
    },
    {
      "auxiliary_loss_clip": 0.01128555,
      "auxiliary_loss_mlp": 0.01027426,
      "balance_loss_clip": 1.04540753,
      "balance_loss_mlp": 1.02017212,
      "epoch": 0.8780135874466423,
      "flos": 22564757514240.0,
      "grad_norm": 1.7740900556268717,
      "language_loss": 0.67729545,
      "learning_rate": 1.5395963968759818e-07,
      "loss": 0.69885528,
      "num_input_tokens_seen": 158044940,
      "step": 7302,
      "time_per_iteration": 2.828589677810669
    },
    {
      "auxiliary_loss_clip": 0.01102092,
      "auxiliary_loss_mlp": 0.01023411,
      "balance_loss_clip": 1.03936601,
      "balance_loss_mlp": 1.01630628,
      "epoch": 0.8781338303372813,
      "flos": 61531999073280.0,
      "grad_norm": 1.5938601739545684,
      "language_loss": 0.64723539,
      "learning_rate": 1.536600691761998e-07,
      "loss": 0.66849041,
      "num_input_tokens_seen": 158070770,
      "step": 7303,
      "time_per_iteration": 3.1976444721221924
    },
    {
      "auxiliary_loss_clip": 0.01114733,
      "auxiliary_loss_mlp": 0.0102669,
      "balance_loss_clip": 1.03964043,
      "balance_loss_mlp": 1.01900744,
      "epoch": 0.8782540732279204,
      "flos": 22674751937280.0,
      "grad_norm": 2.255392834717727,
      "language_loss": 0.72104895,
      "learning_rate": 1.5336077875310084e-07,
      "loss": 0.74246317,
      "num_input_tokens_seen": 158089995,
      "step": 7304,
      "time_per_iteration": 2.7564687728881836
    },
    {
      "auxiliary_loss_clip": 0.01140664,
      "auxiliary_loss_mlp": 0.01026545,
      "balance_loss_clip": 1.04401374,
      "balance_loss_mlp": 1.01877236,
      "epoch": 0.8783743161185595,
      "flos": 16070348937600.0,
      "grad_norm": 1.9771480514083413,
      "language_loss": 0.74047685,
      "learning_rate": 1.5306176846370321e-07,
      "loss": 0.76214892,
      "num_input_tokens_seen": 158108140,
      "step": 7305,
      "time_per_iteration": 2.6905429363250732
    },
    {
      "auxiliary_loss_clip": 0.0109381,
      "auxiliary_loss_mlp": 0.0102655,
      "balance_loss_clip": 1.03556311,
      "balance_loss_mlp": 1.01866484,
      "epoch": 0.8784945590091986,
      "flos": 26067879227520.0,
      "grad_norm": 2.147175715730362,
      "language_loss": 0.74152476,
      "learning_rate": 1.5276303835336712e-07,
      "loss": 0.76272839,
      "num_input_tokens_seen": 158128680,
      "step": 7306,
      "time_per_iteration": 2.82975697517395
    },
    {
      "auxiliary_loss_clip": 0.01002521,
      "auxiliary_loss_mlp": 0.01008762,
      "balance_loss_clip": 1.01824057,
      "balance_loss_mlp": 1.00773084,
      "epoch": 0.8786148018998376,
      "flos": 62720643939840.0,
      "grad_norm": 0.7571573565140893,
      "language_loss": 0.5349251,
      "learning_rate": 1.524645884674094e-07,
      "loss": 0.55503798,
      "num_input_tokens_seen": 158185610,
      "step": 7307,
      "time_per_iteration": 3.5199248790740967
    },
    {
      "auxiliary_loss_clip": 0.01142796,
      "auxiliary_loss_mlp": 0.01024219,
      "balance_loss_clip": 1.04278529,
      "balance_loss_mlp": 1.01584435,
      "epoch": 0.8787350447904768,
      "flos": 21652734263040.0,
      "grad_norm": 3.96330363544101,
      "language_loss": 0.79065418,
      "learning_rate": 1.521664188511047e-07,
      "loss": 0.8123244,
      "num_input_tokens_seen": 158205635,
      "step": 7308,
      "time_per_iteration": 3.913912773132324
    },
    {
      "auxiliary_loss_clip": 0.01153025,
      "auxiliary_loss_mlp": 0.01021226,
      "balance_loss_clip": 1.04742777,
      "balance_loss_mlp": 1.01378751,
      "epoch": 0.8788552876811159,
      "flos": 25478476957440.0,
      "grad_norm": 2.5162999957898213,
      "language_loss": 0.80189955,
      "learning_rate": 1.518685295496851e-07,
      "loss": 0.82364202,
      "num_input_tokens_seen": 158223495,
      "step": 7309,
      "time_per_iteration": 3.718212604522705
    },
    {
      "auxiliary_loss_clip": 0.01121517,
      "auxiliary_loss_mlp": 0.00761889,
      "balance_loss_clip": 1.04069328,
      "balance_loss_mlp": 1.00050962,
      "epoch": 0.8789755305717549,
      "flos": 22310222762880.0,
      "grad_norm": 2.0863195997332378,
      "language_loss": 0.85671604,
      "learning_rate": 1.5157092060833975e-07,
      "loss": 0.87555015,
      "num_input_tokens_seen": 158243145,
      "step": 7310,
      "time_per_iteration": 2.7783613204956055
    },
    {
      "auxiliary_loss_clip": 0.01138754,
      "auxiliary_loss_mlp": 0.01023422,
      "balance_loss_clip": 1.04641104,
      "balance_loss_mlp": 1.01593852,
      "epoch": 0.879095773462394,
      "flos": 29310971408640.0,
      "grad_norm": 1.8793636961018683,
      "language_loss": 0.66260785,
      "learning_rate": 1.5127359207221658e-07,
      "loss": 0.68422961,
      "num_input_tokens_seen": 158262625,
      "step": 7311,
      "time_per_iteration": 2.7264697551727295
    },
    {
      "auxiliary_loss_clip": 0.01064583,
      "auxiliary_loss_mlp": 0.01027594,
      "balance_loss_clip": 1.03612804,
      "balance_loss_mlp": 1.01924336,
      "epoch": 0.8792160163530331,
      "flos": 16690023394560.0,
      "grad_norm": 2.642365954101208,
      "language_loss": 0.73613048,
      "learning_rate": 1.5097654398641923e-07,
      "loss": 0.7570523,
      "num_input_tokens_seen": 158280530,
      "step": 7312,
      "time_per_iteration": 2.883402109146118
    },
    {
      "auxiliary_loss_clip": 0.01086846,
      "auxiliary_loss_mlp": 0.00761915,
      "balance_loss_clip": 1.03818774,
      "balance_loss_mlp": 1.00050962,
      "epoch": 0.8793362592436722,
      "flos": 24499301230080.0,
      "grad_norm": 1.6360717376221705,
      "language_loss": 0.73234856,
      "learning_rate": 1.5067977639601014e-07,
      "loss": 0.75083619,
      "num_input_tokens_seen": 158303290,
      "step": 7313,
      "time_per_iteration": 2.8162953853607178
    },
    {
      "auxiliary_loss_clip": 0.01109083,
      "auxiliary_loss_mlp": 0.01020727,
      "balance_loss_clip": 1.0396713,
      "balance_loss_mlp": 1.01391375,
      "epoch": 0.8794565021343113,
      "flos": 14538399834240.0,
      "grad_norm": 5.191906819799206,
      "language_loss": 0.71131963,
      "learning_rate": 1.5038328934600864e-07,
      "loss": 0.73261774,
      "num_input_tokens_seen": 158319925,
      "step": 7314,
      "time_per_iteration": 2.7369322776794434
    },
    {
      "auxiliary_loss_clip": 0.01118192,
      "auxiliary_loss_mlp": 0.01022732,
      "balance_loss_clip": 1.03892875,
      "balance_loss_mlp": 1.01541245,
      "epoch": 0.8795767450249504,
      "flos": 39530286224640.0,
      "grad_norm": 3.2352226298584577,
      "language_loss": 0.69757926,
      "learning_rate": 1.5008708288139161e-07,
      "loss": 0.71898854,
      "num_input_tokens_seen": 158342285,
      "step": 7315,
      "time_per_iteration": 2.876281976699829
    },
    {
      "auxiliary_loss_clip": 0.01146274,
      "auxiliary_loss_mlp": 0.01026282,
      "balance_loss_clip": 1.04457402,
      "balance_loss_mlp": 1.01851559,
      "epoch": 0.8796969879155895,
      "flos": 22960672197120.0,
      "grad_norm": 2.34826670818617,
      "language_loss": 0.73637903,
      "learning_rate": 1.497911570470931e-07,
      "loss": 0.75810456,
      "num_input_tokens_seen": 158362290,
      "step": 7316,
      "time_per_iteration": 2.759413957595825
    },
    {
      "auxiliary_loss_clip": 0.01122612,
      "auxiliary_loss_mlp": 0.0101894,
      "balance_loss_clip": 1.04051042,
      "balance_loss_mlp": 1.01157594,
      "epoch": 0.8798172308062285,
      "flos": 28362427004160.0,
      "grad_norm": 3.673925370406684,
      "language_loss": 0.8578617,
      "learning_rate": 1.494955118880048e-07,
      "loss": 0.87927717,
      "num_input_tokens_seen": 158383275,
      "step": 7317,
      "time_per_iteration": 2.7730860710144043
    },
    {
      "auxiliary_loss_clip": 0.01104111,
      "auxiliary_loss_mlp": 0.00761956,
      "balance_loss_clip": 1.03751886,
      "balance_loss_mlp": 1.00048065,
      "epoch": 0.8799374736968677,
      "flos": 23988974751360.0,
      "grad_norm": 1.8283165082899484,
      "language_loss": 0.73022753,
      "learning_rate": 1.4920014744897634e-07,
      "loss": 0.74888825,
      "num_input_tokens_seen": 158402690,
      "step": 7318,
      "time_per_iteration": 2.7758243083953857
    },
    {
      "auxiliary_loss_clip": 0.01122691,
      "auxiliary_loss_mlp": 0.01028331,
      "balance_loss_clip": 1.04306769,
      "balance_loss_mlp": 1.02079403,
      "epoch": 0.8800577165875068,
      "flos": 25630271832960.0,
      "grad_norm": 2.025427628649321,
      "language_loss": 0.86642098,
      "learning_rate": 1.4890506377481392e-07,
      "loss": 0.88793123,
      "num_input_tokens_seen": 158421780,
      "step": 7319,
      "time_per_iteration": 2.7488210201263428
    },
    {
      "auxiliary_loss_clip": 0.01123408,
      "auxiliary_loss_mlp": 0.01024759,
      "balance_loss_clip": 1.04160774,
      "balance_loss_mlp": 1.01734149,
      "epoch": 0.8801779594781458,
      "flos": 23440331439360.0,
      "grad_norm": 2.0922612140276393,
      "language_loss": 0.64024383,
      "learning_rate": 1.486102609102815e-07,
      "loss": 0.66172552,
      "num_input_tokens_seen": 158442330,
      "step": 7320,
      "time_per_iteration": 2.7986972332000732
    },
    {
      "auxiliary_loss_clip": 0.01150971,
      "auxiliary_loss_mlp": 0.01021421,
      "balance_loss_clip": 1.04297662,
      "balance_loss_mlp": 1.01383948,
      "epoch": 0.880298202368785,
      "flos": 11508580656000.0,
      "grad_norm": 2.8343237779249573,
      "language_loss": 0.85751438,
      "learning_rate": 1.483157389001004e-07,
      "loss": 0.87923825,
      "num_input_tokens_seen": 158459890,
      "step": 7321,
      "time_per_iteration": 2.677917003631592
    },
    {
      "auxiliary_loss_clip": 0.01137391,
      "auxiliary_loss_mlp": 0.01030469,
      "balance_loss_clip": 1.04614949,
      "balance_loss_mlp": 1.02195716,
      "epoch": 0.880418445259424,
      "flos": 22671447886080.0,
      "grad_norm": 2.3823176617416437,
      "language_loss": 0.79228485,
      "learning_rate": 1.4802149778894933e-07,
      "loss": 0.81396341,
      "num_input_tokens_seen": 158478680,
      "step": 7322,
      "time_per_iteration": 2.6804003715515137
    },
    {
      "auxiliary_loss_clip": 0.01147383,
      "auxiliary_loss_mlp": 0.01021637,
      "balance_loss_clip": 1.04272699,
      "balance_loss_mlp": 1.01454425,
      "epoch": 0.8805386881500631,
      "flos": 20522158709760.0,
      "grad_norm": 1.8280863971380197,
      "language_loss": 0.87680036,
      "learning_rate": 1.4772753762146484e-07,
      "loss": 0.89849061,
      "num_input_tokens_seen": 158497935,
      "step": 7323,
      "time_per_iteration": 3.7170193195343018
    },
    {
      "auxiliary_loss_clip": 0.01090402,
      "auxiliary_loss_mlp": 0.00761759,
      "balance_loss_clip": 1.03742838,
      "balance_loss_mlp": 1.00044382,
      "epoch": 0.8806589310407023,
      "flos": 36538891620480.0,
      "grad_norm": 2.1632587858683485,
      "language_loss": 0.7067793,
      "learning_rate": 1.474338584422401e-07,
      "loss": 0.72530091,
      "num_input_tokens_seen": 158523145,
      "step": 7324,
      "time_per_iteration": 2.9364585876464844
    },
    {
      "auxiliary_loss_clip": 0.01123201,
      "auxiliary_loss_mlp": 0.01024364,
      "balance_loss_clip": 1.04387259,
      "balance_loss_mlp": 1.0169189,
      "epoch": 0.8807791739313413,
      "flos": 23440187784960.0,
      "grad_norm": 2.04039291276566,
      "language_loss": 0.76170188,
      "learning_rate": 1.4714046029582595e-07,
      "loss": 0.78317755,
      "num_input_tokens_seen": 158542210,
      "step": 7325,
      "time_per_iteration": 3.8129265308380127
    },
    {
      "auxiliary_loss_clip": 0.01128495,
      "auxiliary_loss_mlp": 0.01023338,
      "balance_loss_clip": 1.04052472,
      "balance_loss_mlp": 1.01544666,
      "epoch": 0.8808994168219804,
      "flos": 25956843310080.0,
      "grad_norm": 2.857244398806521,
      "language_loss": 0.75619578,
      "learning_rate": 1.46847343226731e-07,
      "loss": 0.77771413,
      "num_input_tokens_seen": 158563250,
      "step": 7326,
      "time_per_iteration": 2.7926061153411865
    },
    {
      "auxiliary_loss_clip": 0.0115396,
      "auxiliary_loss_mlp": 0.01024691,
      "balance_loss_clip": 1.04624367,
      "balance_loss_mlp": 1.01699638,
      "epoch": 0.8810196597126195,
      "flos": 17092079303040.0,
      "grad_norm": 2.0012065523711624,
      "language_loss": 0.69103873,
      "learning_rate": 1.465545072794203e-07,
      "loss": 0.71282518,
      "num_input_tokens_seen": 158581125,
      "step": 7327,
      "time_per_iteration": 2.647535800933838
    },
    {
      "auxiliary_loss_clip": 0.01056049,
      "auxiliary_loss_mlp": 0.01024563,
      "balance_loss_clip": 1.0362699,
      "balance_loss_mlp": 1.01702631,
      "epoch": 0.8811399026032586,
      "flos": 23002831785600.0,
      "grad_norm": 1.850753715249664,
      "language_loss": 0.75890571,
      "learning_rate": 1.4626195249831774e-07,
      "loss": 0.77971184,
      "num_input_tokens_seen": 158602025,
      "step": 7328,
      "time_per_iteration": 2.965169668197632
    },
    {
      "auxiliary_loss_clip": 0.01075466,
      "auxiliary_loss_mlp": 0.01027619,
      "balance_loss_clip": 1.04013896,
      "balance_loss_mlp": 1.01982856,
      "epoch": 0.8812601454938976,
      "flos": 14463813242880.0,
      "grad_norm": 2.1507853001916306,
      "language_loss": 0.71721399,
      "learning_rate": 1.4596967892780244e-07,
      "loss": 0.73824489,
      "num_input_tokens_seen": 158618355,
      "step": 7329,
      "time_per_iteration": 3.2437191009521484
    },
    {
      "auxiliary_loss_clip": 0.01142518,
      "auxiliary_loss_mlp": 0.01022073,
      "balance_loss_clip": 1.0432086,
      "balance_loss_mlp": 1.01480973,
      "epoch": 0.8813803883845368,
      "flos": 22493223578880.0,
      "grad_norm": 1.7486553037190011,
      "language_loss": 0.7459662,
      "learning_rate": 1.4567768661221314e-07,
      "loss": 0.7676121,
      "num_input_tokens_seen": 158638925,
      "step": 7330,
      "time_per_iteration": 3.277554750442505
    },
    {
      "auxiliary_loss_clip": 0.01073075,
      "auxiliary_loss_mlp": 0.01024302,
      "balance_loss_clip": 1.04022551,
      "balance_loss_mlp": 1.01664829,
      "epoch": 0.8815006312751759,
      "flos": 21506901045120.0,
      "grad_norm": 2.2926666081653515,
      "language_loss": 0.74311239,
      "learning_rate": 1.4538597559584442e-07,
      "loss": 0.76408613,
      "num_input_tokens_seen": 158656715,
      "step": 7331,
      "time_per_iteration": 3.249119281768799
    },
    {
      "auxiliary_loss_clip": 0.01088256,
      "auxiliary_loss_mlp": 0.01025784,
      "balance_loss_clip": 1.04281354,
      "balance_loss_mlp": 1.01795757,
      "epoch": 0.8816208741658149,
      "flos": 22784566792320.0,
      "grad_norm": 2.0196266676143013,
      "language_loss": 0.7890203,
      "learning_rate": 1.4509454592294823e-07,
      "loss": 0.81016064,
      "num_input_tokens_seen": 158677200,
      "step": 7332,
      "time_per_iteration": 3.08647084236145
    },
    {
      "auxiliary_loss_clip": 0.01159401,
      "auxiliary_loss_mlp": 0.01024998,
      "balance_loss_clip": 1.04569626,
      "balance_loss_mlp": 1.01807201,
      "epoch": 0.8817411170564541,
      "flos": 17779409026560.0,
      "grad_norm": 2.8136674456314994,
      "language_loss": 0.79532802,
      "learning_rate": 1.448033976377354e-07,
      "loss": 0.81717199,
      "num_input_tokens_seen": 158692185,
      "step": 7333,
      "time_per_iteration": 2.7145955562591553
    },
    {
      "auxiliary_loss_clip": 0.01154454,
      "auxiliary_loss_mlp": 0.00761605,
      "balance_loss_clip": 1.04322469,
      "balance_loss_mlp": 1.00049627,
      "epoch": 0.8818613599470931,
      "flos": 18551812112640.0,
      "grad_norm": 2.824258961024925,
      "language_loss": 0.73998445,
      "learning_rate": 1.445125307843713e-07,
      "loss": 0.75914508,
      "num_input_tokens_seen": 158710410,
      "step": 7334,
      "time_per_iteration": 2.7140450477600098
    },
    {
      "auxiliary_loss_clip": 0.01140463,
      "auxiliary_loss_mlp": 0.01022328,
      "balance_loss_clip": 1.04500914,
      "balance_loss_mlp": 1.01507366,
      "epoch": 0.8819816028377322,
      "flos": 27599792417280.0,
      "grad_norm": 1.900874133107769,
      "language_loss": 0.75649035,
      "learning_rate": 1.442219454069813e-07,
      "loss": 0.77811825,
      "num_input_tokens_seen": 158731435,
      "step": 7335,
      "time_per_iteration": 2.722414016723633
    },
    {
      "auxiliary_loss_clip": 0.01125874,
      "auxiliary_loss_mlp": 0.01027698,
      "balance_loss_clip": 1.04079604,
      "balance_loss_mlp": 1.01992536,
      "epoch": 0.8821018457283714,
      "flos": 23404600385280.0,
      "grad_norm": 1.9386018721496117,
      "language_loss": 0.66708428,
      "learning_rate": 1.4393164154964676e-07,
      "loss": 0.68861997,
      "num_input_tokens_seen": 158750965,
      "step": 7336,
      "time_per_iteration": 3.834986925125122
    },
    {
      "auxiliary_loss_clip": 0.01122189,
      "auxiliary_loss_mlp": 0.00762117,
      "balance_loss_clip": 1.04200065,
      "balance_loss_mlp": 1.00045216,
      "epoch": 0.8822220886190104,
      "flos": 29132459792640.0,
      "grad_norm": 2.9034253814138955,
      "language_loss": 0.939587,
      "learning_rate": 1.4364161925640649e-07,
      "loss": 0.95843005,
      "num_input_tokens_seen": 158772365,
      "step": 7337,
      "time_per_iteration": 2.753399610519409
    },
    {
      "auxiliary_loss_clip": 0.01145617,
      "auxiliary_loss_mlp": 0.0102431,
      "balance_loss_clip": 1.04330134,
      "balance_loss_mlp": 1.01728559,
      "epoch": 0.8823423315096495,
      "flos": 20485422074880.0,
      "grad_norm": 1.9209663029936523,
      "language_loss": 0.85233891,
      "learning_rate": 1.4335187857125663e-07,
      "loss": 0.8740381,
      "num_input_tokens_seen": 158791065,
      "step": 7338,
      "time_per_iteration": 2.618121862411499
    },
    {
      "auxiliary_loss_clip": 0.01059205,
      "auxiliary_loss_mlp": 0.01022449,
      "balance_loss_clip": 1.03463423,
      "balance_loss_mlp": 1.01508439,
      "epoch": 0.8824625744002886,
      "flos": 24206377818240.0,
      "grad_norm": 1.7563685858988278,
      "language_loss": 0.75389218,
      "learning_rate": 1.4306241953815023e-07,
      "loss": 0.77470875,
      "num_input_tokens_seen": 158812125,
      "step": 7339,
      "time_per_iteration": 2.9938852787017822
    },
    {
      "auxiliary_loss_clip": 0.01121131,
      "auxiliary_loss_mlp": 0.01021775,
      "balance_loss_clip": 1.0389483,
      "balance_loss_mlp": 1.01407385,
      "epoch": 0.8825828172909277,
      "flos": 24679500785280.0,
      "grad_norm": 10.641003366089935,
      "language_loss": 0.70936751,
      "learning_rate": 1.4277324220099862e-07,
      "loss": 0.73079658,
      "num_input_tokens_seen": 158834035,
      "step": 7340,
      "time_per_iteration": 3.092163562774658
    },
    {
      "auxiliary_loss_clip": 0.01146793,
      "auxiliary_loss_mlp": 0.01023558,
      "balance_loss_clip": 1.04396892,
      "balance_loss_mlp": 1.01654887,
      "epoch": 0.8827030601815667,
      "flos": 22456163721600.0,
      "grad_norm": 1.816676206535271,
      "language_loss": 0.74121821,
      "learning_rate": 1.4248434660366938e-07,
      "loss": 0.76292169,
      "num_input_tokens_seen": 158853510,
      "step": 7341,
      "time_per_iteration": 2.7133846282958984
    },
    {
      "auxiliary_loss_clip": 0.01155608,
      "auxiliary_loss_mlp": 0.01027171,
      "balance_loss_clip": 1.04602754,
      "balance_loss_mlp": 1.01992285,
      "epoch": 0.8828233030722058,
      "flos": 19865639877120.0,
      "grad_norm": 2.519388193603723,
      "language_loss": 0.70412087,
      "learning_rate": 1.4219573278998808e-07,
      "loss": 0.72594869,
      "num_input_tokens_seen": 158871970,
      "step": 7342,
      "time_per_iteration": 2.652998447418213
    },
    {
      "auxiliary_loss_clip": 0.01138114,
      "auxiliary_loss_mlp": 0.01025087,
      "balance_loss_clip": 1.0465095,
      "balance_loss_mlp": 1.01727247,
      "epoch": 0.882943545962845,
      "flos": 39347213581440.0,
      "grad_norm": 1.9496468604035377,
      "language_loss": 0.64864719,
      "learning_rate": 1.4190740080373685e-07,
      "loss": 0.67027915,
      "num_input_tokens_seen": 158892250,
      "step": 7343,
      "time_per_iteration": 2.873800754547119
    },
    {
      "auxiliary_loss_clip": 0.01145332,
      "auxiliary_loss_mlp": 0.01029779,
      "balance_loss_clip": 1.04930794,
      "balance_loss_mlp": 1.02151203,
      "epoch": 0.883063788853484,
      "flos": 19054524908160.0,
      "grad_norm": 2.106530303527905,
      "language_loss": 0.84497696,
      "learning_rate": 1.4161935068865538e-07,
      "loss": 0.86672807,
      "num_input_tokens_seen": 158907395,
      "step": 7344,
      "time_per_iteration": 2.6747405529022217
    },
    {
      "auxiliary_loss_clip": 0.01155587,
      "auxiliary_loss_mlp": 0.0102347,
      "balance_loss_clip": 1.04378462,
      "balance_loss_mlp": 1.01589406,
      "epoch": 0.8831840317441231,
      "flos": 18733196816640.0,
      "grad_norm": 2.0668381767689175,
      "language_loss": 0.7568478,
      "learning_rate": 1.4133158248844113e-07,
      "loss": 0.77863836,
      "num_input_tokens_seen": 158926300,
      "step": 7345,
      "time_per_iteration": 2.688143253326416
    },
    {
      "auxiliary_loss_clip": 0.0111732,
      "auxiliary_loss_mlp": 0.01024085,
      "balance_loss_clip": 1.04203928,
      "balance_loss_mlp": 1.01592207,
      "epoch": 0.8833042746347622,
      "flos": 26827712553600.0,
      "grad_norm": 1.7230267692199523,
      "language_loss": 0.73318285,
      "learning_rate": 1.4104409624674785e-07,
      "loss": 0.75459689,
      "num_input_tokens_seen": 158946085,
      "step": 7346,
      "time_per_iteration": 2.7132630348205566
    },
    {
      "auxiliary_loss_clip": 0.01094321,
      "auxiliary_loss_mlp": 0.01029905,
      "balance_loss_clip": 1.04316211,
      "balance_loss_mlp": 1.02172756,
      "epoch": 0.8834245175254013,
      "flos": 26104077158400.0,
      "grad_norm": 1.81333051293187,
      "language_loss": 0.7860204,
      "learning_rate": 1.407568920071873e-07,
      "loss": 0.80726266,
      "num_input_tokens_seen": 158964950,
      "step": 7347,
      "time_per_iteration": 2.834001064300537
    },
    {
      "auxiliary_loss_clip": 0.01145362,
      "auxiliary_loss_mlp": 0.01027941,
      "balance_loss_clip": 1.04863667,
      "balance_loss_mlp": 1.02026355,
      "epoch": 0.8835447604160404,
      "flos": 30629036977920.0,
      "grad_norm": 1.9658461638820173,
      "language_loss": 0.68040752,
      "learning_rate": 1.4046996981332782e-07,
      "loss": 0.70214057,
      "num_input_tokens_seen": 158984835,
      "step": 7348,
      "time_per_iteration": 2.744741439819336
    },
    {
      "auxiliary_loss_clip": 0.01098771,
      "auxiliary_loss_mlp": 0.0076261,
      "balance_loss_clip": 1.03920472,
      "balance_loss_mlp": 1.00045371,
      "epoch": 0.8836650033066795,
      "flos": 24718356322560.0,
      "grad_norm": 2.0151488505952773,
      "language_loss": 0.78427112,
      "learning_rate": 1.4018332970869516e-07,
      "loss": 0.80288494,
      "num_input_tokens_seen": 159002775,
      "step": 7349,
      "time_per_iteration": 4.970644474029541
    },
    {
      "auxiliary_loss_clip": 0.01108787,
      "auxiliary_loss_mlp": 0.01028984,
      "balance_loss_clip": 1.03853822,
      "balance_loss_mlp": 1.0208782,
      "epoch": 0.8837852461973186,
      "flos": 25413371556480.0,
      "grad_norm": 1.8281586179709544,
      "language_loss": 0.8538087,
      "learning_rate": 1.398969717367733e-07,
      "loss": 0.87518644,
      "num_input_tokens_seen": 159024100,
      "step": 7350,
      "time_per_iteration": 3.741780996322632
    },
    {
      "auxiliary_loss_clip": 0.0111374,
      "auxiliary_loss_mlp": 0.01026585,
      "balance_loss_clip": 1.0441699,
      "balance_loss_mlp": 1.01979303,
      "epoch": 0.8839054890879576,
      "flos": 17822574195840.0,
      "grad_norm": 2.3190546865720965,
      "language_loss": 0.76768827,
      "learning_rate": 1.396108959410014e-07,
      "loss": 0.78909147,
      "num_input_tokens_seen": 159043315,
      "step": 7351,
      "time_per_iteration": 2.720242738723755
    },
    {
      "auxiliary_loss_clip": 0.01089974,
      "auxiliary_loss_mlp": 0.01024017,
      "balance_loss_clip": 1.04011822,
      "balance_loss_mlp": 1.01640534,
      "epoch": 0.8840257319785968,
      "flos": 23769021818880.0,
      "grad_norm": 1.565417504367074,
      "language_loss": 0.81341088,
      "learning_rate": 1.3932510236477745e-07,
      "loss": 0.83455086,
      "num_input_tokens_seen": 159063985,
      "step": 7352,
      "time_per_iteration": 2.838434934616089
    },
    {
      "auxiliary_loss_clip": 0.0109238,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.04077005,
      "balance_loss_mlp": 1.01575708,
      "epoch": 0.8841459748692359,
      "flos": 29059776622080.0,
      "grad_norm": 3.044656461885769,
      "language_loss": 0.55832112,
      "learning_rate": 1.3903959105145636e-07,
      "loss": 0.57948852,
      "num_input_tokens_seen": 159084475,
      "step": 7353,
      "time_per_iteration": 2.9071013927459717
    },
    {
      "auxiliary_loss_clip": 0.01115045,
      "auxiliary_loss_mlp": 0.01021773,
      "balance_loss_clip": 1.04138553,
      "balance_loss_mlp": 1.01456118,
      "epoch": 0.8842662177598749,
      "flos": 24311523905280.0,
      "grad_norm": 2.519701142193514,
      "language_loss": 0.83198762,
      "learning_rate": 1.387543620443492e-07,
      "loss": 0.85335577,
      "num_input_tokens_seen": 159101320,
      "step": 7354,
      "time_per_iteration": 2.690755844116211
    },
    {
      "auxiliary_loss_clip": 0.01154769,
      "auxiliary_loss_mlp": 0.01027268,
      "balance_loss_clip": 1.047261,
      "balance_loss_mlp": 1.01954353,
      "epoch": 0.8843864606505141,
      "flos": 25007867942400.0,
      "grad_norm": 1.5931185010293551,
      "language_loss": 0.84298933,
      "learning_rate": 1.3846941538672606e-07,
      "loss": 0.86480975,
      "num_input_tokens_seen": 159120025,
      "step": 7355,
      "time_per_iteration": 2.6801939010620117
    },
    {
      "auxiliary_loss_clip": 0.01128664,
      "auxiliary_loss_mlp": 0.00762036,
      "balance_loss_clip": 1.04457939,
      "balance_loss_mlp": 1.0005846,
      "epoch": 0.8845067035411531,
      "flos": 28183915388160.0,
      "grad_norm": 2.9966521821868755,
      "language_loss": 0.81112784,
      "learning_rate": 1.3818475112181193e-07,
      "loss": 0.83003491,
      "num_input_tokens_seen": 159138820,
      "step": 7356,
      "time_per_iteration": 2.730773448944092
    },
    {
      "auxiliary_loss_clip": 0.01138704,
      "auxiliary_loss_mlp": 0.01022449,
      "balance_loss_clip": 1.04289031,
      "balance_loss_mlp": 1.01549864,
      "epoch": 0.8846269464317922,
      "flos": 12853219311360.0,
      "grad_norm": 3.0778513005228705,
      "language_loss": 0.79824108,
      "learning_rate": 1.3790036929279091e-07,
      "loss": 0.81985259,
      "num_input_tokens_seen": 159155975,
      "step": 7357,
      "time_per_iteration": 2.6965980529785156
    },
    {
      "auxiliary_loss_clip": 0.01126573,
      "auxiliary_loss_mlp": 0.01026848,
      "balance_loss_clip": 1.04462838,
      "balance_loss_mlp": 1.0191884,
      "epoch": 0.8847471893224313,
      "flos": 18624351628800.0,
      "grad_norm": 2.3889834177259153,
      "language_loss": 0.58573949,
      "learning_rate": 1.3761626994280363e-07,
      "loss": 0.6072737,
      "num_input_tokens_seen": 159173445,
      "step": 7358,
      "time_per_iteration": 2.7358791828155518
    },
    {
      "auxiliary_loss_clip": 0.01156236,
      "auxiliary_loss_mlp": 0.00761959,
      "balance_loss_clip": 1.04354203,
      "balance_loss_mlp": 1.00051367,
      "epoch": 0.8848674322130704,
      "flos": 35769433449600.0,
      "grad_norm": 2.414202052515257,
      "language_loss": 0.73682201,
      "learning_rate": 1.3733245311494735e-07,
      "loss": 0.75600398,
      "num_input_tokens_seen": 159196100,
      "step": 7359,
      "time_per_iteration": 2.7225568294525146
    },
    {
      "auxiliary_loss_clip": 0.01094836,
      "auxiliary_loss_mlp": 0.00761684,
      "balance_loss_clip": 1.04098165,
      "balance_loss_mlp": 1.00041556,
      "epoch": 0.8849876751037095,
      "flos": 24243760897920.0,
      "grad_norm": 2.1344393690560484,
      "language_loss": 0.70944446,
      "learning_rate": 1.3704891885227676e-07,
      "loss": 0.72800964,
      "num_input_tokens_seen": 159216145,
      "step": 7360,
      "time_per_iteration": 2.8255345821380615
    },
    {
      "auxiliary_loss_clip": 0.01152971,
      "auxiliary_loss_mlp": 0.0102814,
      "balance_loss_clip": 1.04565454,
      "balance_loss_mlp": 1.01994443,
      "epoch": 0.8851079179943486,
      "flos": 21500580251520.0,
      "grad_norm": 2.8097869024520015,
      "language_loss": 0.77983946,
      "learning_rate": 1.367656671978037e-07,
      "loss": 0.80165058,
      "num_input_tokens_seen": 159233610,
      "step": 7361,
      "time_per_iteration": 3.5523979663848877
    },
    {
      "auxiliary_loss_clip": 0.0112531,
      "auxiliary_loss_mlp": 0.01021279,
      "balance_loss_clip": 1.04120719,
      "balance_loss_mlp": 1.01447845,
      "epoch": 0.8852281608849877,
      "flos": 15300711198720.0,
      "grad_norm": 1.9991834382274751,
      "language_loss": 0.74082601,
      "learning_rate": 1.36482698194498e-07,
      "loss": 0.76229191,
      "num_input_tokens_seen": 159250155,
      "step": 7362,
      "time_per_iteration": 2.6982431411743164
    },
    {
      "auxiliary_loss_clip": 0.01128008,
      "auxiliary_loss_mlp": 0.01023401,
      "balance_loss_clip": 1.04306269,
      "balance_loss_mlp": 1.01545,
      "epoch": 0.8853484037756267,
      "flos": 23295719283840.0,
      "grad_norm": 1.9905704457059255,
      "language_loss": 0.72130203,
      "learning_rate": 1.3620001188528506e-07,
      "loss": 0.74281609,
      "num_input_tokens_seen": 159270875,
      "step": 7363,
      "time_per_iteration": 2.7572484016418457
    },
    {
      "auxiliary_loss_clip": 0.01117808,
      "auxiliary_loss_mlp": 0.01024053,
      "balance_loss_clip": 1.03834665,
      "balance_loss_mlp": 1.01602471,
      "epoch": 0.8854686466662659,
      "flos": 25114773795840.0,
      "grad_norm": 2.856578187481944,
      "language_loss": 0.7342006,
      "learning_rate": 1.3591760831304865e-07,
      "loss": 0.75561923,
      "num_input_tokens_seen": 159288565,
      "step": 7364,
      "time_per_iteration": 2.710934638977051
    },
    {
      "auxiliary_loss_clip": 0.01119325,
      "auxiliary_loss_mlp": 0.01024745,
      "balance_loss_clip": 1.04215217,
      "balance_loss_mlp": 1.01696682,
      "epoch": 0.885588889556905,
      "flos": 21390873137280.0,
      "grad_norm": 1.791510951736716,
      "language_loss": 0.79460955,
      "learning_rate": 1.356354875206287e-07,
      "loss": 0.81605029,
      "num_input_tokens_seen": 159306400,
      "step": 7365,
      "time_per_iteration": 2.705418825149536
    },
    {
      "auxiliary_loss_clip": 0.01092059,
      "auxiliary_loss_mlp": 0.01020155,
      "balance_loss_clip": 1.04074359,
      "balance_loss_mlp": 1.01217949,
      "epoch": 0.885709132447544,
      "flos": 26906752431360.0,
      "grad_norm": 9.298145671469193,
      "language_loss": 0.70002657,
      "learning_rate": 1.3535364955082296e-07,
      "loss": 0.72114873,
      "num_input_tokens_seen": 159326250,
      "step": 7366,
      "time_per_iteration": 2.8399317264556885
    },
    {
      "auxiliary_loss_clip": 0.01098833,
      "auxiliary_loss_mlp": 0.01019956,
      "balance_loss_clip": 1.03784335,
      "balance_loss_mlp": 1.01287794,
      "epoch": 0.8858293753381832,
      "flos": 26103394800000.0,
      "grad_norm": 2.0520404128123957,
      "language_loss": 0.64564133,
      "learning_rate": 1.3507209444638613e-07,
      "loss": 0.66682917,
      "num_input_tokens_seen": 159348250,
      "step": 7367,
      "time_per_iteration": 2.801469326019287
    },
    {
      "auxiliary_loss_clip": 0.01075187,
      "auxiliary_loss_mlp": 0.00762102,
      "balance_loss_clip": 1.03351164,
      "balance_loss_mlp": 1.00047708,
      "epoch": 0.8859496182288222,
      "flos": 23292810282240.0,
      "grad_norm": 1.8133045068671645,
      "language_loss": 0.74524021,
      "learning_rate": 1.347908222500298e-07,
      "loss": 0.7636131,
      "num_input_tokens_seen": 159368325,
      "step": 7368,
      "time_per_iteration": 3.00106143951416
    },
    {
      "auxiliary_loss_clip": 0.01148377,
      "auxiliary_loss_mlp": 0.01021319,
      "balance_loss_clip": 1.04495907,
      "balance_loss_mlp": 1.0139544,
      "epoch": 0.8860698611194613,
      "flos": 16872916469760.0,
      "grad_norm": 2.1655300258929575,
      "language_loss": 0.70384806,
      "learning_rate": 1.3450983300442276e-07,
      "loss": 0.72554499,
      "num_input_tokens_seen": 159387555,
      "step": 7369,
      "time_per_iteration": 3.097148895263672
    },
    {
      "auxiliary_loss_clip": 0.01112544,
      "auxiliary_loss_mlp": 0.01028841,
      "balance_loss_clip": 1.03941512,
      "balance_loss_mlp": 1.02186716,
      "epoch": 0.8861901040101005,
      "flos": 24681404206080.0,
      "grad_norm": 1.9175865402732872,
      "language_loss": 0.73688042,
      "learning_rate": 1.3422912675219068e-07,
      "loss": 0.75829428,
      "num_input_tokens_seen": 159407310,
      "step": 7370,
      "time_per_iteration": 2.773935317993164
    },
    {
      "auxiliary_loss_clip": 0.01105385,
      "auxiliary_loss_mlp": 0.01021918,
      "balance_loss_clip": 1.04221606,
      "balance_loss_mlp": 1.0147357,
      "epoch": 0.8863103469007395,
      "flos": 24423026699520.0,
      "grad_norm": 1.7284346687318983,
      "language_loss": 0.79016256,
      "learning_rate": 1.339487035359166e-07,
      "loss": 0.81143558,
      "num_input_tokens_seen": 159427680,
      "step": 7371,
      "time_per_iteration": 2.7878804206848145
    },
    {
      "auxiliary_loss_clip": 0.01118359,
      "auxiliary_loss_mlp": 0.01023017,
      "balance_loss_clip": 1.04102826,
      "balance_loss_mlp": 1.0159781,
      "epoch": 0.8864305897913786,
      "flos": 22053964158720.0,
      "grad_norm": 1.6686643273035686,
      "language_loss": 0.84743536,
      "learning_rate": 1.336685633981409e-07,
      "loss": 0.86884916,
      "num_input_tokens_seen": 159448765,
      "step": 7372,
      "time_per_iteration": 2.726043939590454
    },
    {
      "auxiliary_loss_clip": 0.01089498,
      "auxiliary_loss_mlp": 0.01024914,
      "balance_loss_clip": 1.0378933,
      "balance_loss_mlp": 1.01667058,
      "epoch": 0.8865508326820177,
      "flos": 19099449843840.0,
      "grad_norm": 2.8395953589146896,
      "language_loss": 0.75140083,
      "learning_rate": 1.333887063813597e-07,
      "loss": 0.77254492,
      "num_input_tokens_seen": 159466870,
      "step": 7373,
      "time_per_iteration": 2.87349271774292
    },
    {
      "auxiliary_loss_clip": 0.01141832,
      "auxiliary_loss_mlp": 0.01021644,
      "balance_loss_clip": 1.04308128,
      "balance_loss_mlp": 1.01465857,
      "epoch": 0.8866710755726568,
      "flos": 15414189240960.0,
      "grad_norm": 1.7282431754589436,
      "language_loss": 0.66524518,
      "learning_rate": 1.331091325280278e-07,
      "loss": 0.68687993,
      "num_input_tokens_seen": 159485840,
      "step": 7374,
      "time_per_iteration": 2.7604119777679443
    },
    {
      "auxiliary_loss_clip": 0.01141818,
      "auxiliary_loss_mlp": 0.0103106,
      "balance_loss_clip": 1.04482889,
      "balance_loss_mlp": 1.02345479,
      "epoch": 0.8867913184632958,
      "flos": 20083689388800.0,
      "grad_norm": 1.9012053442494623,
      "language_loss": 0.78712624,
      "learning_rate": 1.3282984188055625e-07,
      "loss": 0.80885506,
      "num_input_tokens_seen": 159505630,
      "step": 7375,
      "time_per_iteration": 4.2217512130737305
    },
    {
      "auxiliary_loss_clip": 0.01127645,
      "auxiliary_loss_mlp": 0.01023194,
      "balance_loss_clip": 1.04367232,
      "balance_loss_mlp": 1.01507568,
      "epoch": 0.8869115613539349,
      "flos": 23365852588800.0,
      "grad_norm": 1.848488831664243,
      "language_loss": 0.79658979,
      "learning_rate": 1.3255083448131288e-07,
      "loss": 0.81809819,
      "num_input_tokens_seen": 159524675,
      "step": 7376,
      "time_per_iteration": 3.6119141578674316
    },
    {
      "auxiliary_loss_clip": 0.01157809,
      "auxiliary_loss_mlp": 0.01027445,
      "balance_loss_clip": 1.0442512,
      "balance_loss_mlp": 1.01992881,
      "epoch": 0.8870318042445741,
      "flos": 21286840371840.0,
      "grad_norm": 2.2143413504459,
      "language_loss": 0.78929007,
      "learning_rate": 1.3227211037262365e-07,
      "loss": 0.81114256,
      "num_input_tokens_seen": 159541915,
      "step": 7377,
      "time_per_iteration": 3.4957804679870605
    },
    {
      "auxiliary_loss_clip": 0.01115871,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.03909993,
      "balance_loss_mlp": 1.01907539,
      "epoch": 0.8871520471352131,
      "flos": 20010862563840.0,
      "grad_norm": 2.3388471767806434,
      "language_loss": 0.85402751,
      "learning_rate": 1.319936695967696e-07,
      "loss": 0.8754558,
      "num_input_tokens_seen": 159559740,
      "step": 7378,
      "time_per_iteration": 2.712956190109253
    },
    {
      "auxiliary_loss_clip": 0.01145619,
      "auxiliary_loss_mlp": 0.01028211,
      "balance_loss_clip": 1.04834354,
      "balance_loss_mlp": 1.01946115,
      "epoch": 0.8872722900258522,
      "flos": 22601422321920.0,
      "grad_norm": 2.396835513128908,
      "language_loss": 0.82681829,
      "learning_rate": 1.3171551219599097e-07,
      "loss": 0.84855658,
      "num_input_tokens_seen": 159578265,
      "step": 7379,
      "time_per_iteration": 2.6988401412963867
    },
    {
      "auxiliary_loss_clip": 0.01064712,
      "auxiliary_loss_mlp": 0.00762053,
      "balance_loss_clip": 1.03800738,
      "balance_loss_mlp": 1.0004518,
      "epoch": 0.8873925329164913,
      "flos": 22163276223360.0,
      "grad_norm": 2.182864340692606,
      "language_loss": 0.78338426,
      "learning_rate": 1.3143763821248377e-07,
      "loss": 0.80165195,
      "num_input_tokens_seen": 159595350,
      "step": 7380,
      "time_per_iteration": 3.023757219314575
    },
    {
      "auxiliary_loss_clip": 0.01105478,
      "auxiliary_loss_mlp": 0.00761496,
      "balance_loss_clip": 1.03931165,
      "balance_loss_mlp": 1.00043631,
      "epoch": 0.8875127758071304,
      "flos": 19208223204480.0,
      "grad_norm": 1.9842711151912098,
      "language_loss": 0.72334003,
      "learning_rate": 1.3116004768840118e-07,
      "loss": 0.74200982,
      "num_input_tokens_seen": 159613725,
      "step": 7381,
      "time_per_iteration": 2.878434896469116
    },
    {
      "auxiliary_loss_clip": 0.01151991,
      "auxiliary_loss_mlp": 0.01028024,
      "balance_loss_clip": 1.04318571,
      "balance_loss_mlp": 1.02027583,
      "epoch": 0.8876330186977694,
      "flos": 18110900666880.0,
      "grad_norm": 2.4178711518830727,
      "language_loss": 0.74399984,
      "learning_rate": 1.3088274066585348e-07,
      "loss": 0.7658,
      "num_input_tokens_seen": 159631335,
      "step": 7382,
      "time_per_iteration": 2.675098419189453
    },
    {
      "auxiliary_loss_clip": 0.01121431,
      "auxiliary_loss_mlp": 0.01024169,
      "balance_loss_clip": 1.03889418,
      "balance_loss_mlp": 1.01653957,
      "epoch": 0.8877532615884086,
      "flos": 22009434272640.0,
      "grad_norm": 2.209046121511453,
      "language_loss": 0.90671217,
      "learning_rate": 1.3060571718690749e-07,
      "loss": 0.92816818,
      "num_input_tokens_seen": 159648830,
      "step": 7383,
      "time_per_iteration": 2.7069644927978516
    },
    {
      "auxiliary_loss_clip": 0.01031155,
      "auxiliary_loss_mlp": 0.0100015,
      "balance_loss_clip": 1.01079321,
      "balance_loss_mlp": 0.99905312,
      "epoch": 0.8878735044790477,
      "flos": 72136924346880.0,
      "grad_norm": 0.7431488919970393,
      "language_loss": 0.56837487,
      "learning_rate": 1.3032897729358805e-07,
      "loss": 0.5886879,
      "num_input_tokens_seen": 159709785,
      "step": 7384,
      "time_per_iteration": 3.358717441558838
    },
    {
      "auxiliary_loss_clip": 0.01078527,
      "auxiliary_loss_mlp": 0.01028553,
      "balance_loss_clip": 1.03690493,
      "balance_loss_mlp": 1.02044725,
      "epoch": 0.8879937473696867,
      "flos": 27526355061120.0,
      "grad_norm": 3.0084888150063955,
      "language_loss": 0.80083859,
      "learning_rate": 1.3005252102787645e-07,
      "loss": 0.82190931,
      "num_input_tokens_seen": 159728725,
      "step": 7385,
      "time_per_iteration": 2.8637988567352295
    },
    {
      "auxiliary_loss_clip": 0.01142011,
      "auxiliary_loss_mlp": 0.01028309,
      "balance_loss_clip": 1.04628944,
      "balance_loss_mlp": 1.02054894,
      "epoch": 0.8881139902603259,
      "flos": 22234091886720.0,
      "grad_norm": 1.615076148914967,
      "language_loss": 0.73652023,
      "learning_rate": 1.297763484317105e-07,
      "loss": 0.75822341,
      "num_input_tokens_seen": 159747020,
      "step": 7386,
      "time_per_iteration": 2.748227834701538
    },
    {
      "auxiliary_loss_clip": 0.01112098,
      "auxiliary_loss_mlp": 0.01024169,
      "balance_loss_clip": 1.04162049,
      "balance_loss_mlp": 1.01611638,
      "epoch": 0.888234233150965,
      "flos": 20299548170880.0,
      "grad_norm": 2.7036200618150397,
      "language_loss": 0.70057476,
      "learning_rate": 1.2950045954698551e-07,
      "loss": 0.72193742,
      "num_input_tokens_seen": 159764855,
      "step": 7387,
      "time_per_iteration": 3.665152072906494
    },
    {
      "auxiliary_loss_clip": 0.01117873,
      "auxiliary_loss_mlp": 0.0102247,
      "balance_loss_clip": 1.04048753,
      "balance_loss_mlp": 1.01573753,
      "epoch": 0.888354476041604,
      "flos": 18147996437760.0,
      "grad_norm": 1.62036577794519,
      "language_loss": 0.75700533,
      "learning_rate": 1.2922485441555343e-07,
      "loss": 0.77840877,
      "num_input_tokens_seen": 159783935,
      "step": 7388,
      "time_per_iteration": 2.7833707332611084
    },
    {
      "auxiliary_loss_clip": 0.01105154,
      "auxiliary_loss_mlp": 0.01020996,
      "balance_loss_clip": 1.03686261,
      "balance_loss_mlp": 1.01429009,
      "epoch": 0.8884747189322432,
      "flos": 22014282608640.0,
      "grad_norm": 2.2657152852728126,
      "language_loss": 0.819754,
      "learning_rate": 1.2894953307922363e-07,
      "loss": 0.84101546,
      "num_input_tokens_seen": 159802895,
      "step": 7389,
      "time_per_iteration": 2.8211045265197754
    },
    {
      "auxiliary_loss_clip": 0.01113554,
      "auxiliary_loss_mlp": 0.01024882,
      "balance_loss_clip": 1.03842711,
      "balance_loss_mlp": 1.01760459,
      "epoch": 0.8885949618228822,
      "flos": 19786779567360.0,
      "grad_norm": 2.2880907311957737,
      "language_loss": 0.83868611,
      "learning_rate": 1.2867449557976208e-07,
      "loss": 0.86007059,
      "num_input_tokens_seen": 159820995,
      "step": 7390,
      "time_per_iteration": 2.683467388153076
    },
    {
      "auxiliary_loss_clip": 0.01122482,
      "auxiliary_loss_mlp": 0.01023104,
      "balance_loss_clip": 1.04323292,
      "balance_loss_mlp": 1.01540327,
      "epoch": 0.8887152047135213,
      "flos": 20047599198720.0,
      "grad_norm": 1.9902684163795354,
      "language_loss": 0.75565892,
      "learning_rate": 1.283997419588916e-07,
      "loss": 0.77711475,
      "num_input_tokens_seen": 159840465,
      "step": 7391,
      "time_per_iteration": 2.715665102005005
    },
    {
      "auxiliary_loss_clip": 0.01090275,
      "auxiliary_loss_mlp": 0.01026529,
      "balance_loss_clip": 1.03814292,
      "balance_loss_mlp": 1.01927173,
      "epoch": 0.8888354476041604,
      "flos": 18588117784320.0,
      "grad_norm": 2.001810419818935,
      "language_loss": 0.62031001,
      "learning_rate": 1.2812527225829216e-07,
      "loss": 0.64147806,
      "num_input_tokens_seen": 159858690,
      "step": 7392,
      "time_per_iteration": 2.809896469116211
    },
    {
      "auxiliary_loss_clip": 0.01172727,
      "auxiliary_loss_mlp": 0.0102919,
      "balance_loss_clip": 1.04882455,
      "balance_loss_mlp": 1.02092242,
      "epoch": 0.8889556904947995,
      "flos": 21689794120320.0,
      "grad_norm": 4.222363463594059,
      "language_loss": 0.76478159,
      "learning_rate": 1.2785108651960052e-07,
      "loss": 0.78680074,
      "num_input_tokens_seen": 159880325,
      "step": 7393,
      "time_per_iteration": 2.5862557888031006
    },
    {
      "auxiliary_loss_clip": 0.01113069,
      "auxiliary_loss_mlp": 0.01027528,
      "balance_loss_clip": 1.04434085,
      "balance_loss_mlp": 1.01955247,
      "epoch": 0.8890759333854386,
      "flos": 27381204201600.0,
      "grad_norm": 1.9285555951219109,
      "language_loss": 0.80469799,
      "learning_rate": 1.2757718478441094e-07,
      "loss": 0.82610404,
      "num_input_tokens_seen": 159901070,
      "step": 7394,
      "time_per_iteration": 2.8064587116241455
    },
    {
      "auxiliary_loss_clip": 0.01149563,
      "auxiliary_loss_mlp": 0.01023218,
      "balance_loss_clip": 1.04426527,
      "balance_loss_mlp": 1.01582122,
      "epoch": 0.8891961762760777,
      "flos": 24498834353280.0,
      "grad_norm": 1.9475182133578708,
      "language_loss": 0.77403545,
      "learning_rate": 1.2730356709427302e-07,
      "loss": 0.79576325,
      "num_input_tokens_seen": 159919750,
      "step": 7395,
      "time_per_iteration": 2.693650484085083
    },
    {
      "auxiliary_loss_clip": 0.01123586,
      "auxiliary_loss_mlp": 0.01021291,
      "balance_loss_clip": 1.0421617,
      "balance_loss_mlp": 1.01360464,
      "epoch": 0.8893164191667168,
      "flos": 41499770895360.0,
      "grad_norm": 1.9584060964638343,
      "language_loss": 0.59927797,
      "learning_rate": 1.2703023349069542e-07,
      "loss": 0.6207267,
      "num_input_tokens_seen": 159944600,
      "step": 7396,
      "time_per_iteration": 2.835523843765259
    },
    {
      "auxiliary_loss_clip": 0.01116127,
      "auxiliary_loss_mlp": 0.00761671,
      "balance_loss_clip": 1.03963327,
      "balance_loss_mlp": 1.00053918,
      "epoch": 0.8894366620573558,
      "flos": 33583623120000.0,
      "grad_norm": 1.8921361080605101,
      "language_loss": 0.61915803,
      "learning_rate": 1.2675718401514223e-07,
      "loss": 0.637936,
      "num_input_tokens_seen": 159968780,
      "step": 7397,
      "time_per_iteration": 2.784877061843872
    },
    {
      "auxiliary_loss_clip": 0.01138345,
      "auxiliary_loss_mlp": 0.0076196,
      "balance_loss_clip": 1.04657662,
      "balance_loss_mlp": 1.0005219,
      "epoch": 0.889556904947995,
      "flos": 16909832672640.0,
      "grad_norm": 2.126965480430305,
      "language_loss": 0.74449337,
      "learning_rate": 1.264844187090346e-07,
      "loss": 0.7634964,
      "num_input_tokens_seen": 159985905,
      "step": 7398,
      "time_per_iteration": 2.6824188232421875
    },
    {
      "auxiliary_loss_clip": 0.01121856,
      "auxiliary_loss_mlp": 0.01025717,
      "balance_loss_clip": 1.04032445,
      "balance_loss_mlp": 1.01815963,
      "epoch": 0.889677147838634,
      "flos": 26030855283840.0,
      "grad_norm": 2.1104191514486663,
      "language_loss": 0.75234926,
      "learning_rate": 1.262119376137516e-07,
      "loss": 0.77382499,
      "num_input_tokens_seen": 160006965,
      "step": 7399,
      "time_per_iteration": 2.7279341220855713
    },
    {
      "auxiliary_loss_clip": 0.01150892,
      "auxiliary_loss_mlp": 0.01024723,
      "balance_loss_clip": 1.04465985,
      "balance_loss_mlp": 1.01754951,
      "epoch": 0.8897973907292731,
      "flos": 26468283110400.0,
      "grad_norm": 2.0551613543776264,
      "language_loss": 0.85255706,
      "learning_rate": 1.2593974077062707e-07,
      "loss": 0.87431324,
      "num_input_tokens_seen": 160028585,
      "step": 7400,
      "time_per_iteration": 2.6481077671051025
    },
    {
      "auxiliary_loss_clip": 0.01092249,
      "auxiliary_loss_mlp": 0.01022423,
      "balance_loss_clip": 1.03892803,
      "balance_loss_mlp": 1.01490116,
      "epoch": 0.8899176336199123,
      "flos": 26249694894720.0,
      "grad_norm": 1.7850710954152367,
      "language_loss": 0.63550138,
      "learning_rate": 1.2566782822095423e-07,
      "loss": 0.6566481,
      "num_input_tokens_seen": 160048840,
      "step": 7401,
      "time_per_iteration": 4.071284055709839
    },
    {
      "auxiliary_loss_clip": 0.01123153,
      "auxiliary_loss_mlp": 0.01026397,
      "balance_loss_clip": 1.04504156,
      "balance_loss_mlp": 1.01847005,
      "epoch": 0.8900378765105513,
      "flos": 20811742156800.0,
      "grad_norm": 2.512881072981537,
      "language_loss": 0.71150863,
      "learning_rate": 1.2539620000598162e-07,
      "loss": 0.73300421,
      "num_input_tokens_seen": 160068175,
      "step": 7402,
      "time_per_iteration": 3.6131694316864014
    },
    {
      "auxiliary_loss_clip": 0.01100298,
      "auxiliary_loss_mlp": 0.0102216,
      "balance_loss_clip": 1.04054976,
      "balance_loss_mlp": 1.01515102,
      "epoch": 0.8901581194011904,
      "flos": 16472333018880.0,
      "grad_norm": 10.79487438599447,
      "language_loss": 0.80031252,
      "learning_rate": 1.2512485616691492e-07,
      "loss": 0.82153714,
      "num_input_tokens_seen": 160085230,
      "step": 7403,
      "time_per_iteration": 2.854351043701172
    },
    {
      "auxiliary_loss_clip": 0.01131533,
      "auxiliary_loss_mlp": 0.00762515,
      "balance_loss_clip": 1.04622078,
      "balance_loss_mlp": 1.00046229,
      "epoch": 0.8902783622918296,
      "flos": 35155253773440.0,
      "grad_norm": 1.5587863072122694,
      "language_loss": 0.80986458,
      "learning_rate": 1.2485379674491681e-07,
      "loss": 0.82880509,
      "num_input_tokens_seen": 160111425,
      "step": 7404,
      "time_per_iteration": 2.9008102416992188
    },
    {
      "auxiliary_loss_clip": 0.01095324,
      "auxiliary_loss_mlp": 0.01024628,
      "balance_loss_clip": 1.03886878,
      "balance_loss_mlp": 1.016397,
      "epoch": 0.8903986051824686,
      "flos": 17201068145280.0,
      "grad_norm": 2.2118252294356764,
      "language_loss": 0.79283893,
      "learning_rate": 1.2458302178110657e-07,
      "loss": 0.81403846,
      "num_input_tokens_seen": 160129790,
      "step": 7405,
      "time_per_iteration": 2.8846213817596436
    },
    {
      "auxiliary_loss_clip": 0.01153042,
      "auxiliary_loss_mlp": 0.01022247,
      "balance_loss_clip": 1.04338694,
      "balance_loss_mlp": 1.01533842,
      "epoch": 0.8905188480731077,
      "flos": 25483863997440.0,
      "grad_norm": 2.2850065731880864,
      "language_loss": 0.82488334,
      "learning_rate": 1.2431253131656118e-07,
      "loss": 0.84663624,
      "num_input_tokens_seen": 160149265,
      "step": 7406,
      "time_per_iteration": 2.6880548000335693
    },
    {
      "auxiliary_loss_clip": 0.0115404,
      "auxiliary_loss_mlp": 0.01022919,
      "balance_loss_clip": 1.04234755,
      "balance_loss_mlp": 1.01492548,
      "epoch": 0.8906390909637467,
      "flos": 23365888502400.0,
      "grad_norm": 1.867552794481422,
      "language_loss": 0.76857567,
      "learning_rate": 1.240423253923133e-07,
      "loss": 0.79034531,
      "num_input_tokens_seen": 160168870,
      "step": 7407,
      "time_per_iteration": 2.6318318843841553
    },
    {
      "auxiliary_loss_clip": 0.01115943,
      "auxiliary_loss_mlp": 0.01029945,
      "balance_loss_clip": 1.04462552,
      "balance_loss_mlp": 1.02191317,
      "epoch": 0.8907593338543859,
      "flos": 21068790860160.0,
      "grad_norm": 2.3036164341894856,
      "language_loss": 0.69766605,
      "learning_rate": 1.237724040493533e-07,
      "loss": 0.71912491,
      "num_input_tokens_seen": 160187495,
      "step": 7408,
      "time_per_iteration": 2.7592673301696777
    },
    {
      "auxiliary_loss_clip": 0.01164768,
      "auxiliary_loss_mlp": 0.00762477,
      "balance_loss_clip": 1.04917777,
      "balance_loss_mlp": 1.00051415,
      "epoch": 0.8908795767450249,
      "flos": 21869562712320.0,
      "grad_norm": 6.006608063506746,
      "language_loss": 0.72816861,
      "learning_rate": 1.2350276732862773e-07,
      "loss": 0.74744111,
      "num_input_tokens_seen": 160208520,
      "step": 7409,
      "time_per_iteration": 2.660576343536377
    },
    {
      "auxiliary_loss_clip": 0.01027611,
      "auxiliary_loss_mlp": 0.01002377,
      "balance_loss_clip": 1.01293778,
      "balance_loss_mlp": 1.0013814,
      "epoch": 0.890999819635664,
      "flos": 66307869348480.0,
      "grad_norm": 0.8407827650637906,
      "language_loss": 0.56677926,
      "learning_rate": 1.2323341527103993e-07,
      "loss": 0.58707917,
      "num_input_tokens_seen": 160263720,
      "step": 7410,
      "time_per_iteration": 3.1910197734832764
    },
    {
      "auxiliary_loss_clip": 0.01072314,
      "auxiliary_loss_mlp": 0.01026026,
      "balance_loss_clip": 1.03629935,
      "balance_loss_mlp": 1.01877201,
      "epoch": 0.8911200625263032,
      "flos": 26869908055680.0,
      "grad_norm": 3.2172747075535963,
      "language_loss": 0.85308778,
      "learning_rate": 1.2296434791745135e-07,
      "loss": 0.87407124,
      "num_input_tokens_seen": 160282170,
      "step": 7411,
      "time_per_iteration": 3.008477210998535
    },
    {
      "auxiliary_loss_clip": 0.01114275,
      "auxiliary_loss_mlp": 0.00762422,
      "balance_loss_clip": 1.04066634,
      "balance_loss_mlp": 1.00045514,
      "epoch": 0.8912403054169422,
      "flos": 20885825957760.0,
      "grad_norm": 1.6350149676795664,
      "language_loss": 0.76847267,
      "learning_rate": 1.2269556530867875e-07,
      "loss": 0.78723961,
      "num_input_tokens_seen": 160300725,
      "step": 7412,
      "time_per_iteration": 3.709923267364502
    },
    {
      "auxiliary_loss_clip": 0.01147271,
      "auxiliary_loss_mlp": 0.01025648,
      "balance_loss_clip": 1.04720902,
      "balance_loss_mlp": 1.01758909,
      "epoch": 0.8913605483075813,
      "flos": 27016567286400.0,
      "grad_norm": 2.2479784582698574,
      "language_loss": 0.81947505,
      "learning_rate": 1.2242706748549614e-07,
      "loss": 0.84120429,
      "num_input_tokens_seen": 160318720,
      "step": 7413,
      "time_per_iteration": 2.7266733646392822
    },
    {
      "auxiliary_loss_clip": 0.0108392,
      "auxiliary_loss_mlp": 0.01024123,
      "balance_loss_clip": 1.03869271,
      "balance_loss_mlp": 1.0167383,
      "epoch": 0.8914807911982204,
      "flos": 23621500661760.0,
      "grad_norm": 1.9751823299872746,
      "language_loss": 0.82612509,
      "learning_rate": 1.2215885448863473e-07,
      "loss": 0.84720552,
      "num_input_tokens_seen": 160339595,
      "step": 7414,
      "time_per_iteration": 2.8279473781585693
    },
    {
      "auxiliary_loss_clip": 0.01142295,
      "auxiliary_loss_mlp": 0.01025748,
      "balance_loss_clip": 1.04882407,
      "balance_loss_mlp": 1.01817226,
      "epoch": 0.8916010340888595,
      "flos": 24462277286400.0,
      "grad_norm": 1.9258936180219302,
      "language_loss": 0.80783081,
      "learning_rate": 1.2189092635878152e-07,
      "loss": 0.82951117,
      "num_input_tokens_seen": 160361045,
      "step": 7415,
      "time_per_iteration": 2.8218162059783936
    },
    {
      "auxiliary_loss_clip": 0.0114262,
      "auxiliary_loss_mlp": 0.01023007,
      "balance_loss_clip": 1.04525864,
      "balance_loss_mlp": 1.01492465,
      "epoch": 0.8917212769794985,
      "flos": 21215773313280.0,
      "grad_norm": 1.9741627685323935,
      "language_loss": 0.77342194,
      "learning_rate": 1.216232831365822e-07,
      "loss": 0.79507822,
      "num_input_tokens_seen": 160379990,
      "step": 7416,
      "time_per_iteration": 2.7633538246154785
    },
    {
      "auxiliary_loss_clip": 0.01075703,
      "auxiliary_loss_mlp": 0.01026253,
      "balance_loss_clip": 1.03727031,
      "balance_loss_mlp": 1.01872492,
      "epoch": 0.8918415198701377,
      "flos": 25513992529920.0,
      "grad_norm": 1.8054313918669962,
      "language_loss": 0.80990207,
      "learning_rate": 1.2135592486263678e-07,
      "loss": 0.83092165,
      "num_input_tokens_seen": 160399240,
      "step": 7417,
      "time_per_iteration": 2.858520030975342
    },
    {
      "auxiliary_loss_clip": 0.01136482,
      "auxiliary_loss_mlp": 0.01023602,
      "balance_loss_clip": 1.04530096,
      "balance_loss_mlp": 1.01566267,
      "epoch": 0.8919617627607768,
      "flos": 37853006693760.0,
      "grad_norm": 1.8162477448931704,
      "language_loss": 0.61223751,
      "learning_rate": 1.2108885157750415e-07,
      "loss": 0.63383842,
      "num_input_tokens_seen": 160421600,
      "step": 7418,
      "time_per_iteration": 2.8800477981567383
    },
    {
      "auxiliary_loss_clip": 0.01054896,
      "auxiliary_loss_mlp": 0.01029121,
      "balance_loss_clip": 1.03633273,
      "balance_loss_mlp": 1.02122974,
      "epoch": 0.8920820056514158,
      "flos": 26213676531840.0,
      "grad_norm": 1.7482139757709803,
      "language_loss": 0.80311304,
      "learning_rate": 1.2082206332169897e-07,
      "loss": 0.82395315,
      "num_input_tokens_seen": 160441695,
      "step": 7419,
      "time_per_iteration": 2.956247091293335
    },
    {
      "auxiliary_loss_clip": 0.01143489,
      "auxiliary_loss_mlp": 0.01023355,
      "balance_loss_clip": 1.04233372,
      "balance_loss_mlp": 1.01588356,
      "epoch": 0.892202248542055,
      "flos": 17383135207680.0,
      "grad_norm": 2.41356010650322,
      "language_loss": 0.73761594,
      "learning_rate": 1.2055556013569225e-07,
      "loss": 0.75928438,
      "num_input_tokens_seen": 160457205,
      "step": 7420,
      "time_per_iteration": 3.2877113819122314
    },
    {
      "auxiliary_loss_clip": 0.01118469,
      "auxiliary_loss_mlp": 0.0102389,
      "balance_loss_clip": 1.04065096,
      "balance_loss_mlp": 1.01656437,
      "epoch": 0.892322491432694,
      "flos": 21324223451520.0,
      "grad_norm": 1.7876786750263347,
      "language_loss": 0.82171863,
      "learning_rate": 1.2028934205991315e-07,
      "loss": 0.84314221,
      "num_input_tokens_seen": 160476525,
      "step": 7421,
      "time_per_iteration": 2.7632524967193604
    },
    {
      "auxiliary_loss_clip": 0.01090347,
      "auxiliary_loss_mlp": 0.0102251,
      "balance_loss_clip": 1.03672135,
      "balance_loss_mlp": 1.01493156,
      "epoch": 0.8924427343233331,
      "flos": 24029374573440.0,
      "grad_norm": 1.5839134841015703,
      "language_loss": 0.76772314,
      "learning_rate": 1.2002340913474607e-07,
      "loss": 0.78885168,
      "num_input_tokens_seen": 160500160,
      "step": 7422,
      "time_per_iteration": 2.999013662338257
    },
    {
      "auxiliary_loss_clip": 0.01138848,
      "auxiliary_loss_mlp": 0.00762478,
      "balance_loss_clip": 1.04200578,
      "balance_loss_mlp": 1.00045443,
      "epoch": 0.8925629772139723,
      "flos": 30008069631360.0,
      "grad_norm": 1.930911015275123,
      "language_loss": 0.74351394,
      "learning_rate": 1.1975776140053317e-07,
      "loss": 0.76252717,
      "num_input_tokens_seen": 160520130,
      "step": 7423,
      "time_per_iteration": 2.7983100414276123
    },
    {
      "auxiliary_loss_clip": 0.0110898,
      "auxiliary_loss_mlp": 0.01028148,
      "balance_loss_clip": 1.04249024,
      "balance_loss_mlp": 1.01998234,
      "epoch": 0.8926832201046113,
      "flos": 22601709630720.0,
      "grad_norm": 2.1366628016098894,
      "language_loss": 0.73708439,
      "learning_rate": 1.194923988975729e-07,
      "loss": 0.75845563,
      "num_input_tokens_seen": 160539730,
      "step": 7424,
      "time_per_iteration": 2.7749788761138916
    },
    {
      "auxiliary_loss_clip": 0.01119691,
      "auxiliary_loss_mlp": 0.00761944,
      "balance_loss_clip": 1.04217863,
      "balance_loss_mlp": 1.00049496,
      "epoch": 0.8928034629952504,
      "flos": 13297722117120.0,
      "grad_norm": 2.2005454537332882,
      "language_loss": 0.73756546,
      "learning_rate": 1.192273216661206e-07,
      "loss": 0.75638187,
      "num_input_tokens_seen": 160557820,
      "step": 7425,
      "time_per_iteration": 2.732532501220703
    },
    {
      "auxiliary_loss_clip": 0.01036904,
      "auxiliary_loss_mlp": 0.01001801,
      "balance_loss_clip": 1.0109998,
      "balance_loss_mlp": 1.00075841,
      "epoch": 0.8929237058858895,
      "flos": 54854556744960.0,
      "grad_norm": 0.766457829964775,
      "language_loss": 0.5744428,
      "learning_rate": 1.189625297463881e-07,
      "loss": 0.59482986,
      "num_input_tokens_seen": 160619510,
      "step": 7426,
      "time_per_iteration": 4.264575719833374
    },
    {
      "auxiliary_loss_clip": 0.01121269,
      "auxiliary_loss_mlp": 0.01024354,
      "balance_loss_clip": 1.04010582,
      "balance_loss_mlp": 1.016361,
      "epoch": 0.8930439487765286,
      "flos": 28883850785280.0,
      "grad_norm": 1.7046453995593065,
      "language_loss": 0.79909283,
      "learning_rate": 1.1869802317854394e-07,
      "loss": 0.82054901,
      "num_input_tokens_seen": 160643295,
      "step": 7427,
      "time_per_iteration": 4.055570125579834
    },
    {
      "auxiliary_loss_clip": 0.01094114,
      "auxiliary_loss_mlp": 0.01025104,
      "balance_loss_clip": 1.03791952,
      "balance_loss_mlp": 1.01719701,
      "epoch": 0.8931641916671677,
      "flos": 22419283432320.0,
      "grad_norm": 2.8714958214572377,
      "language_loss": 0.72010362,
      "learning_rate": 1.1843380200271425e-07,
      "loss": 0.74129581,
      "num_input_tokens_seen": 160662495,
      "step": 7428,
      "time_per_iteration": 3.757835865020752
    },
    {
      "auxiliary_loss_clip": 0.01093073,
      "auxiliary_loss_mlp": 0.01026461,
      "balance_loss_clip": 1.03876007,
      "balance_loss_mlp": 1.01843786,
      "epoch": 0.8932844345578068,
      "flos": 25843149786240.0,
      "grad_norm": 2.3904653569272987,
      "language_loss": 0.80657905,
      "learning_rate": 1.181698662589805e-07,
      "loss": 0.82777441,
      "num_input_tokens_seen": 160682080,
      "step": 7429,
      "time_per_iteration": 2.868600606918335
    },
    {
      "auxiliary_loss_clip": 0.01130313,
      "auxiliary_loss_mlp": 0.01022824,
      "balance_loss_clip": 1.04101992,
      "balance_loss_mlp": 1.01521218,
      "epoch": 0.8934046774484459,
      "flos": 22925803069440.0,
      "grad_norm": 2.5855697722964908,
      "language_loss": 0.76389396,
      "learning_rate": 1.1790621598738249e-07,
      "loss": 0.78542536,
      "num_input_tokens_seen": 160700395,
      "step": 7430,
      "time_per_iteration": 2.766213893890381
    },
    {
      "auxiliary_loss_clip": 0.01087047,
      "auxiliary_loss_mlp": 0.01022159,
      "balance_loss_clip": 1.0385747,
      "balance_loss_mlp": 1.01515269,
      "epoch": 0.8935249203390849,
      "flos": 24462097718400.0,
      "grad_norm": 2.0859251756863886,
      "language_loss": 0.74923766,
      "learning_rate": 1.1764285122791461e-07,
      "loss": 0.77032971,
      "num_input_tokens_seen": 160721115,
      "step": 7431,
      "time_per_iteration": 2.8380963802337646
    },
    {
      "auxiliary_loss_clip": 0.01154828,
      "auxiliary_loss_mlp": 0.01024226,
      "balance_loss_clip": 1.04626036,
      "balance_loss_mlp": 1.01697505,
      "epoch": 0.8936451632297241,
      "flos": 15742735966080.0,
      "grad_norm": 2.05144552676613,
      "language_loss": 0.77524853,
      "learning_rate": 1.173797720205294e-07,
      "loss": 0.79703903,
      "num_input_tokens_seen": 160739150,
      "step": 7432,
      "time_per_iteration": 2.6368343830108643
    },
    {
      "auxiliary_loss_clip": 0.01143076,
      "auxiliary_loss_mlp": 0.01031019,
      "balance_loss_clip": 1.04668939,
      "balance_loss_mlp": 1.02258515,
      "epoch": 0.8937654061203631,
      "flos": 35115500396160.0,
      "grad_norm": 3.7915902260017877,
      "language_loss": 0.71630132,
      "learning_rate": 1.1711697840513602e-07,
      "loss": 0.73804224,
      "num_input_tokens_seen": 160758585,
      "step": 7433,
      "time_per_iteration": 2.7690343856811523
    },
    {
      "auxiliary_loss_clip": 0.01122718,
      "auxiliary_loss_mlp": 0.00761889,
      "balance_loss_clip": 1.03839731,
      "balance_loss_mlp": 1.00044274,
      "epoch": 0.8938856490110022,
      "flos": 16107444708480.0,
      "grad_norm": 6.1292700605550925,
      "language_loss": 0.71230292,
      "learning_rate": 1.1685447042160012e-07,
      "loss": 0.73114896,
      "num_input_tokens_seen": 160776620,
      "step": 7434,
      "time_per_iteration": 2.6644229888916016
    },
    {
      "auxiliary_loss_clip": 0.01077301,
      "auxiliary_loss_mlp": 0.01029065,
      "balance_loss_clip": 1.03619838,
      "balance_loss_mlp": 1.02079463,
      "epoch": 0.8940058919016414,
      "flos": 20704189858560.0,
      "grad_norm": 1.5831168291977107,
      "language_loss": 0.71596622,
      "learning_rate": 1.1659224810974367e-07,
      "loss": 0.73702991,
      "num_input_tokens_seen": 160796580,
      "step": 7435,
      "time_per_iteration": 2.9393930435180664
    },
    {
      "auxiliary_loss_clip": 0.01110278,
      "auxiliary_loss_mlp": 0.01027255,
      "balance_loss_clip": 1.0429492,
      "balance_loss_mlp": 1.01971543,
      "epoch": 0.8941261347922804,
      "flos": 25229041937280.0,
      "grad_norm": 1.7420913353484546,
      "language_loss": 0.68599898,
      "learning_rate": 1.1633031150934591e-07,
      "loss": 0.70737433,
      "num_input_tokens_seen": 160819610,
      "step": 7436,
      "time_per_iteration": 3.016413927078247
    },
    {
      "auxiliary_loss_clip": 0.01140142,
      "auxiliary_loss_mlp": 0.01032075,
      "balance_loss_clip": 1.04387498,
      "balance_loss_mlp": 1.02373624,
      "epoch": 0.8942463776829195,
      "flos": 19537236806400.0,
      "grad_norm": 1.9731744792107377,
      "language_loss": 0.79707432,
      "learning_rate": 1.1606866066014176e-07,
      "loss": 0.81879652,
      "num_input_tokens_seen": 160838660,
      "step": 7437,
      "time_per_iteration": 2.7582147121429443
    },
    {
      "auxiliary_loss_clip": 0.01119757,
      "auxiliary_loss_mlp": 0.01022375,
      "balance_loss_clip": 1.04137945,
      "balance_loss_mlp": 1.01439357,
      "epoch": 0.8943666205735585,
      "flos": 22301567585280.0,
      "grad_norm": 5.949009850650719,
      "language_loss": 0.75409645,
      "learning_rate": 1.1580729560182434e-07,
      "loss": 0.77551776,
      "num_input_tokens_seen": 160854515,
      "step": 7438,
      "time_per_iteration": 3.659611940383911
    },
    {
      "auxiliary_loss_clip": 0.01152341,
      "auxiliary_loss_mlp": 0.01022455,
      "balance_loss_clip": 1.04484797,
      "balance_loss_mlp": 1.01525486,
      "epoch": 0.8944868634641977,
      "flos": 18912893581440.0,
      "grad_norm": 2.6817907053730634,
      "language_loss": 0.70888269,
      "learning_rate": 1.1554621637404171e-07,
      "loss": 0.73063064,
      "num_input_tokens_seen": 160872605,
      "step": 7439,
      "time_per_iteration": 2.627345085144043
    },
    {
      "auxiliary_loss_clip": 0.01111491,
      "auxiliary_loss_mlp": 0.01025243,
      "balance_loss_clip": 1.04018044,
      "balance_loss_mlp": 1.01724672,
      "epoch": 0.8946071063548368,
      "flos": 14460904241280.0,
      "grad_norm": 2.4106994276171165,
      "language_loss": 0.61165363,
      "learning_rate": 1.1528542301639999e-07,
      "loss": 0.633021,
      "num_input_tokens_seen": 160889395,
      "step": 7440,
      "time_per_iteration": 2.662688732147217
    },
    {
      "auxiliary_loss_clip": 0.01130164,
      "auxiliary_loss_mlp": 0.00762279,
      "balance_loss_clip": 1.0442965,
      "balance_loss_mlp": 1.00043714,
      "epoch": 0.8947273492454758,
      "flos": 20084084438400.0,
      "grad_norm": 2.4503324223828957,
      "language_loss": 0.82647508,
      "learning_rate": 1.1502491556846105e-07,
      "loss": 0.8453995,
      "num_input_tokens_seen": 160907890,
      "step": 7441,
      "time_per_iteration": 2.7733848094940186
    },
    {
      "auxiliary_loss_clip": 0.01115449,
      "auxiliary_loss_mlp": 0.01021385,
      "balance_loss_clip": 1.03761458,
      "balance_loss_mlp": 1.01365161,
      "epoch": 0.894847592136115,
      "flos": 18550555136640.0,
      "grad_norm": 2.4389837388529814,
      "language_loss": 0.8130033,
      "learning_rate": 1.1476469406974331e-07,
      "loss": 0.83437157,
      "num_input_tokens_seen": 160923490,
      "step": 7442,
      "time_per_iteration": 2.7224082946777344
    },
    {
      "auxiliary_loss_clip": 0.01068863,
      "auxiliary_loss_mlp": 0.0102355,
      "balance_loss_clip": 1.03797364,
      "balance_loss_mlp": 1.01635563,
      "epoch": 0.894967835026754,
      "flos": 23478468704640.0,
      "grad_norm": 1.7410950345876044,
      "language_loss": 0.77166367,
      "learning_rate": 1.1450475855972341e-07,
      "loss": 0.79258776,
      "num_input_tokens_seen": 160944280,
      "step": 7443,
      "time_per_iteration": 2.950169801712036
    },
    {
      "auxiliary_loss_clip": 0.01153373,
      "auxiliary_loss_mlp": 0.01024044,
      "balance_loss_clip": 1.04378176,
      "balance_loss_mlp": 1.01645637,
      "epoch": 0.8950880779173931,
      "flos": 15188310564480.0,
      "grad_norm": 2.295278663557358,
      "language_loss": 0.70869923,
      "learning_rate": 1.1424510907783158e-07,
      "loss": 0.7304734,
      "num_input_tokens_seen": 160961560,
      "step": 7444,
      "time_per_iteration": 3.5844168663024902
    },
    {
      "auxiliary_loss_clip": 0.01121364,
      "auxiliary_loss_mlp": 0.01021462,
      "balance_loss_clip": 1.0424459,
      "balance_loss_mlp": 1.01434875,
      "epoch": 0.8952083208080323,
      "flos": 22091957769600.0,
      "grad_norm": 2.054286317332953,
      "language_loss": 0.82770956,
      "learning_rate": 1.1398574566345787e-07,
      "loss": 0.84913778,
      "num_input_tokens_seen": 160982195,
      "step": 7445,
      "time_per_iteration": 2.9328925609588623
    },
    {
      "auxiliary_loss_clip": 0.0114453,
      "auxiliary_loss_mlp": 0.01029847,
      "balance_loss_clip": 1.04118252,
      "balance_loss_mlp": 1.02160406,
      "epoch": 0.8953285636986713,
      "flos": 23254026572160.0,
      "grad_norm": 3.032703635325532,
      "language_loss": 0.82846081,
      "learning_rate": 1.1372666835594702e-07,
      "loss": 0.85020459,
      "num_input_tokens_seen": 161000520,
      "step": 7446,
      "time_per_iteration": 2.725437641143799
    },
    {
      "auxiliary_loss_clip": 0.01141771,
      "auxiliary_loss_mlp": 0.00762025,
      "balance_loss_clip": 1.04187548,
      "balance_loss_mlp": 1.00047874,
      "epoch": 0.8954488065893104,
      "flos": 16362661818240.0,
      "grad_norm": 2.363309480276,
      "language_loss": 0.71865165,
      "learning_rate": 1.1346787719460071e-07,
      "loss": 0.73768961,
      "num_input_tokens_seen": 161019405,
      "step": 7447,
      "time_per_iteration": 2.7555668354034424
    },
    {
      "auxiliary_loss_clip": 0.01090152,
      "auxiliary_loss_mlp": 0.01024811,
      "balance_loss_clip": 1.03701806,
      "balance_loss_mlp": 1.01715147,
      "epoch": 0.8955690494799495,
      "flos": 18257883120000.0,
      "grad_norm": 2.1076472675454485,
      "language_loss": 0.72323835,
      "learning_rate": 1.1320937221867732e-07,
      "loss": 0.74438798,
      "num_input_tokens_seen": 161036985,
      "step": 7448,
      "time_per_iteration": 2.8823440074920654
    },
    {
      "auxiliary_loss_clip": 0.01098141,
      "auxiliary_loss_mlp": 0.00761433,
      "balance_loss_clip": 1.04284358,
      "balance_loss_mlp": 1.00053024,
      "epoch": 0.8956892923705886,
      "flos": 25447486498560.0,
      "grad_norm": 1.7651519246896326,
      "language_loss": 0.7976687,
      "learning_rate": 1.1295115346739192e-07,
      "loss": 0.81626439,
      "num_input_tokens_seen": 161056985,
      "step": 7449,
      "time_per_iteration": 2.832667827606201
    },
    {
      "auxiliary_loss_clip": 0.0113723,
      "auxiliary_loss_mlp": 0.01024757,
      "balance_loss_clip": 1.04259586,
      "balance_loss_mlp": 1.01648974,
      "epoch": 0.8958095352612276,
      "flos": 52661883939840.0,
      "grad_norm": 3.038203806594275,
      "language_loss": 0.73328519,
      "learning_rate": 1.1269322097991629e-07,
      "loss": 0.75490505,
      "num_input_tokens_seen": 161080270,
      "step": 7450,
      "time_per_iteration": 2.9762635231018066
    },
    {
      "auxiliary_loss_clip": 0.01127444,
      "auxiliary_loss_mlp": 0.00762783,
      "balance_loss_clip": 1.04397154,
      "balance_loss_mlp": 1.00042474,
      "epoch": 0.8959297781518668,
      "flos": 23186335392000.0,
      "grad_norm": 1.8390443048729097,
      "language_loss": 0.68472254,
      "learning_rate": 1.1243557479537846e-07,
      "loss": 0.70362484,
      "num_input_tokens_seen": 161100160,
      "step": 7451,
      "time_per_iteration": 2.730618953704834
    },
    {
      "auxiliary_loss_clip": 0.01151155,
      "auxiliary_loss_mlp": 0.01027693,
      "balance_loss_clip": 1.044204,
      "balance_loss_mlp": 1.0198555,
      "epoch": 0.8960500210425059,
      "flos": 20334309557760.0,
      "grad_norm": 2.170031933026438,
      "language_loss": 0.68952149,
      "learning_rate": 1.121782149528634e-07,
      "loss": 0.71131003,
      "num_input_tokens_seen": 161117260,
      "step": 7452,
      "time_per_iteration": 4.419187545776367
    },
    {
      "auxiliary_loss_clip": 0.0112664,
      "auxiliary_loss_mlp": 0.0102405,
      "balance_loss_clip": 1.04056716,
      "balance_loss_mlp": 1.01697493,
      "epoch": 0.8961702639331449,
      "flos": 19901694153600.0,
      "grad_norm": 2.3838641260886098,
      "language_loss": 0.79226494,
      "learning_rate": 1.1192114149141208e-07,
      "loss": 0.81377184,
      "num_input_tokens_seen": 161136895,
      "step": 7453,
      "time_per_iteration": 2.7817113399505615
    },
    {
      "auxiliary_loss_clip": 0.01097661,
      "auxiliary_loss_mlp": 0.01025691,
      "balance_loss_clip": 1.03999352,
      "balance_loss_mlp": 1.01746571,
      "epoch": 0.8962905068237841,
      "flos": 12896348567040.0,
      "grad_norm": 2.2287509286504337,
      "language_loss": 0.65368497,
      "learning_rate": 1.1166435445002197e-07,
      "loss": 0.67491847,
      "num_input_tokens_seen": 161154565,
      "step": 7454,
      "time_per_iteration": 3.6974687576293945
    },
    {
      "auxiliary_loss_clip": 0.01139402,
      "auxiliary_loss_mlp": 0.01027323,
      "balance_loss_clip": 1.04456663,
      "balance_loss_mlp": 1.01930046,
      "epoch": 0.8964107497144231,
      "flos": 23440331439360.0,
      "grad_norm": 2.0536521147084383,
      "language_loss": 0.68496901,
      "learning_rate": 1.1140785386764818e-07,
      "loss": 0.70663631,
      "num_input_tokens_seen": 161173265,
      "step": 7455,
      "time_per_iteration": 2.6997358798980713
    },
    {
      "auxiliary_loss_clip": 0.01115269,
      "auxiliary_loss_mlp": 0.00761862,
      "balance_loss_clip": 1.04434943,
      "balance_loss_mlp": 1.0005703,
      "epoch": 0.8965309926050622,
      "flos": 19500176949120.0,
      "grad_norm": 2.420696557124073,
      "language_loss": 0.69587845,
      "learning_rate": 1.1115163978320153e-07,
      "loss": 0.71464968,
      "num_input_tokens_seen": 161191995,
      "step": 7456,
      "time_per_iteration": 2.812937021255493
    },
    {
      "auxiliary_loss_clip": 0.01135047,
      "auxiliary_loss_mlp": 0.01023972,
      "balance_loss_clip": 1.04626536,
      "balance_loss_mlp": 1.01566887,
      "epoch": 0.8966512354957014,
      "flos": 28658008022400.0,
      "grad_norm": 2.1197268927192456,
      "language_loss": 0.82360595,
      "learning_rate": 1.1089571223554917e-07,
      "loss": 0.84519613,
      "num_input_tokens_seen": 161212880,
      "step": 7457,
      "time_per_iteration": 2.7383673191070557
    },
    {
      "auxiliary_loss_clip": 0.01098044,
      "auxiliary_loss_mlp": 0.01026753,
      "balance_loss_clip": 1.03823733,
      "balance_loss_mlp": 1.01918316,
      "epoch": 0.8967714783863404,
      "flos": 23370916406400.0,
      "grad_norm": 2.129710659479486,
      "language_loss": 0.85743427,
      "learning_rate": 1.1064007126351537e-07,
      "loss": 0.87868226,
      "num_input_tokens_seen": 161233595,
      "step": 7458,
      "time_per_iteration": 2.813314199447632
    },
    {
      "auxiliary_loss_clip": 0.01072188,
      "auxiliary_loss_mlp": 0.0102269,
      "balance_loss_clip": 1.0348947,
      "balance_loss_mlp": 1.0147897,
      "epoch": 0.8968917212769795,
      "flos": 24535175938560.0,
      "grad_norm": 2.2610393581121584,
      "language_loss": 0.76590836,
      "learning_rate": 1.1038471690588003e-07,
      "loss": 0.78685713,
      "num_input_tokens_seen": 161252740,
      "step": 7459,
      "time_per_iteration": 2.884650230407715
    },
    {
      "auxiliary_loss_clip": 0.01140826,
      "auxiliary_loss_mlp": 0.01019891,
      "balance_loss_clip": 1.04232037,
      "balance_loss_mlp": 1.01258302,
      "epoch": 0.8970119641676186,
      "flos": 23475416048640.0,
      "grad_norm": 1.9664903458636949,
      "language_loss": 0.80370855,
      "learning_rate": 1.1012964920138145e-07,
      "loss": 0.82531571,
      "num_input_tokens_seen": 161272325,
      "step": 7460,
      "time_per_iteration": 2.7692863941192627
    },
    {
      "auxiliary_loss_clip": 0.01102077,
      "auxiliary_loss_mlp": 0.01022493,
      "balance_loss_clip": 1.03696072,
      "balance_loss_mlp": 1.01525092,
      "epoch": 0.8971322070582577,
      "flos": 24538192680960.0,
      "grad_norm": 2.0995519930525535,
      "language_loss": 0.75730252,
      "learning_rate": 1.0987486818871205e-07,
      "loss": 0.77854824,
      "num_input_tokens_seen": 161295915,
      "step": 7461,
      "time_per_iteration": 2.8619132041931152
    },
    {
      "auxiliary_loss_clip": 0.01086228,
      "auxiliary_loss_mlp": 0.01021827,
      "balance_loss_clip": 1.04032695,
      "balance_loss_mlp": 1.01397157,
      "epoch": 0.8972524499488967,
      "flos": 21797454159360.0,
      "grad_norm": 2.1220061771496317,
      "language_loss": 0.72796881,
      "learning_rate": 1.0962037390652245e-07,
      "loss": 0.74904937,
      "num_input_tokens_seen": 161314935,
      "step": 7462,
      "time_per_iteration": 2.8364877700805664
    },
    {
      "auxiliary_loss_clip": 0.01140781,
      "auxiliary_loss_mlp": 0.01027065,
      "balance_loss_clip": 1.04472566,
      "balance_loss_mlp": 1.01908422,
      "epoch": 0.8973726928395359,
      "flos": 21726243446400.0,
      "grad_norm": 1.8650976494160263,
      "language_loss": 0.71955037,
      "learning_rate": 1.0936616639341911e-07,
      "loss": 0.74122888,
      "num_input_tokens_seen": 161335225,
      "step": 7463,
      "time_per_iteration": 3.2993826866149902
    },
    {
      "auxiliary_loss_clip": 0.01046116,
      "auxiliary_loss_mlp": 0.01001025,
      "balance_loss_clip": 1.01085329,
      "balance_loss_mlp": 1.0000174,
      "epoch": 0.897492935730175,
      "flos": 53837100097920.0,
      "grad_norm": 0.7371356746297469,
      "language_loss": 0.54700524,
      "learning_rate": 1.0911224568796473e-07,
      "loss": 0.56747663,
      "num_input_tokens_seen": 161393420,
      "step": 7464,
      "time_per_iteration": 4.9333086013793945
    },
    {
      "auxiliary_loss_clip": 0.01125594,
      "auxiliary_loss_mlp": 0.00761562,
      "balance_loss_clip": 1.04574752,
      "balance_loss_mlp": 1.00056672,
      "epoch": 0.897613178620814,
      "flos": 18290346036480.0,
      "grad_norm": 3.130765473366468,
      "language_loss": 0.71219528,
      "learning_rate": 1.0885861182867984e-07,
      "loss": 0.73106682,
      "num_input_tokens_seen": 161411525,
      "step": 7465,
      "time_per_iteration": 2.8957409858703613
    },
    {
      "auxiliary_loss_clip": 0.01125974,
      "auxiliary_loss_mlp": 0.01025467,
      "balance_loss_clip": 1.04455447,
      "balance_loss_mlp": 1.0182786,
      "epoch": 0.8977334215114532,
      "flos": 32993718059520.0,
      "grad_norm": 1.8928573660268038,
      "language_loss": 0.70668674,
      "learning_rate": 1.0860526485403942e-07,
      "loss": 0.72820115,
      "num_input_tokens_seen": 161432800,
      "step": 7466,
      "time_per_iteration": 2.811947822570801
    },
    {
      "auxiliary_loss_clip": 0.01157256,
      "auxiliary_loss_mlp": 0.00761629,
      "balance_loss_clip": 1.04557037,
      "balance_loss_mlp": 1.00046229,
      "epoch": 0.8978536644020922,
      "flos": 15195636938880.0,
      "grad_norm": 1.6766207822696773,
      "language_loss": 0.77044284,
      "learning_rate": 1.0835220480247675e-07,
      "loss": 0.78963166,
      "num_input_tokens_seen": 161451295,
      "step": 7467,
      "time_per_iteration": 2.6906816959381104
    },
    {
      "auxiliary_loss_clip": 0.01088095,
      "auxiliary_loss_mlp": 0.00762549,
      "balance_loss_clip": 1.03665996,
      "balance_loss_mlp": 1.00048327,
      "epoch": 0.8979739072927313,
      "flos": 18004389863040.0,
      "grad_norm": 2.4489894057473496,
      "language_loss": 0.83948231,
      "learning_rate": 1.0809943171238067e-07,
      "loss": 0.85798872,
      "num_input_tokens_seen": 161469220,
      "step": 7468,
      "time_per_iteration": 2.792147636413574
    },
    {
      "auxiliary_loss_clip": 0.01128863,
      "auxiliary_loss_mlp": 0.01032014,
      "balance_loss_clip": 1.04545891,
      "balance_loss_mlp": 1.02353859,
      "epoch": 0.8980941501833704,
      "flos": 22271546793600.0,
      "grad_norm": 9.673666199340436,
      "language_loss": 0.62793475,
      "learning_rate": 1.078469456220965e-07,
      "loss": 0.64954352,
      "num_input_tokens_seen": 161489375,
      "step": 7469,
      "time_per_iteration": 2.723923921585083
    },
    {
      "auxiliary_loss_clip": 0.01096411,
      "auxiliary_loss_mlp": 0.0076204,
      "balance_loss_clip": 1.0396924,
      "balance_loss_mlp": 1.00039148,
      "epoch": 0.8982143930740095,
      "flos": 37560729726720.0,
      "grad_norm": 1.8092185041746733,
      "language_loss": 0.70109069,
      "learning_rate": 1.0759474656992606e-07,
      "loss": 0.7196753,
      "num_input_tokens_seen": 161512145,
      "step": 7470,
      "time_per_iteration": 2.9947986602783203
    },
    {
      "auxiliary_loss_clip": 0.01100177,
      "auxiliary_loss_mlp": 0.0102196,
      "balance_loss_clip": 1.03759325,
      "balance_loss_mlp": 1.01404405,
      "epoch": 0.8983346359646486,
      "flos": 18076893465600.0,
      "grad_norm": 2.859992163911746,
      "language_loss": 0.77839649,
      "learning_rate": 1.0734283459412785e-07,
      "loss": 0.79961783,
      "num_input_tokens_seen": 161528995,
      "step": 7471,
      "time_per_iteration": 2.8289966583251953
    },
    {
      "auxiliary_loss_clip": 0.01170658,
      "auxiliary_loss_mlp": 0.01024251,
      "balance_loss_clip": 1.04802537,
      "balance_loss_mlp": 1.01621318,
      "epoch": 0.8984548788552876,
      "flos": 20558895344640.0,
      "grad_norm": 1.9050857154507765,
      "language_loss": 0.80534971,
      "learning_rate": 1.0709120973291707e-07,
      "loss": 0.82729876,
      "num_input_tokens_seen": 161548775,
      "step": 7472,
      "time_per_iteration": 2.6391332149505615
    },
    {
      "auxiliary_loss_clip": 0.01118065,
      "auxiliary_loss_mlp": 0.01029138,
      "balance_loss_clip": 1.04107976,
      "balance_loss_mlp": 1.02075791,
      "epoch": 0.8985751217459268,
      "flos": 17785442511360.0,
      "grad_norm": 2.280893722388335,
      "language_loss": 0.7773124,
      "learning_rate": 1.0683987202446475e-07,
      "loss": 0.79878449,
      "num_input_tokens_seen": 161566960,
      "step": 7473,
      "time_per_iteration": 2.7598917484283447
    },
    {
      "auxiliary_loss_clip": 0.01099922,
      "auxiliary_loss_mlp": 0.01026209,
      "balance_loss_clip": 1.04176652,
      "balance_loss_mlp": 1.01825213,
      "epoch": 0.8986953646365659,
      "flos": 21617003208960.0,
      "grad_norm": 1.770829081037783,
      "language_loss": 0.69953114,
      "learning_rate": 1.0658882150689862e-07,
      "loss": 0.72079247,
      "num_input_tokens_seen": 161585820,
      "step": 7474,
      "time_per_iteration": 2.796884298324585
    },
    {
      "auxiliary_loss_clip": 0.01117537,
      "auxiliary_loss_mlp": 0.01023878,
      "balance_loss_clip": 1.0421288,
      "balance_loss_mlp": 1.01569402,
      "epoch": 0.8988156075272049,
      "flos": 14027355083520.0,
      "grad_norm": 2.600970783858761,
      "language_loss": 0.78937435,
      "learning_rate": 1.0633805821830288e-07,
      "loss": 0.81078851,
      "num_input_tokens_seen": 161602505,
      "step": 7475,
      "time_per_iteration": 2.783588171005249
    },
    {
      "auxiliary_loss_clip": 0.01142037,
      "auxiliary_loss_mlp": 0.01023875,
      "balance_loss_clip": 1.0424304,
      "balance_loss_mlp": 1.01586676,
      "epoch": 0.8989358504178441,
      "flos": 29059202004480.0,
      "grad_norm": 2.1911513236551152,
      "language_loss": 0.82894576,
      "learning_rate": 1.0608758219671753e-07,
      "loss": 0.85060489,
      "num_input_tokens_seen": 161621545,
      "step": 7476,
      "time_per_iteration": 2.7408924102783203
    },
    {
      "auxiliary_loss_clip": 0.01160092,
      "auxiliary_loss_mlp": 0.01023612,
      "balance_loss_clip": 1.04692721,
      "balance_loss_mlp": 1.01583362,
      "epoch": 0.8990560933084831,
      "flos": 20230420446720.0,
      "grad_norm": 1.6870059786166145,
      "language_loss": 0.70705134,
      "learning_rate": 1.0583739348014065e-07,
      "loss": 0.72888839,
      "num_input_tokens_seen": 161642630,
      "step": 7477,
      "time_per_iteration": 2.6275150775909424
    },
    {
      "auxiliary_loss_clip": 0.01144706,
      "auxiliary_loss_mlp": 0.01022809,
      "balance_loss_clip": 1.04463434,
      "balance_loss_mlp": 1.01516175,
      "epoch": 0.8991763361991222,
      "flos": 25520672459520.0,
      "grad_norm": 7.346942658509201,
      "language_loss": 0.84768355,
      "learning_rate": 1.0558749210652518e-07,
      "loss": 0.86935866,
      "num_input_tokens_seen": 161662560,
      "step": 7478,
      "time_per_iteration": 4.71118426322937
    },
    {
      "auxiliary_loss_clip": 0.01131895,
      "auxiliary_loss_mlp": 0.0102304,
      "balance_loss_clip": 1.04406762,
      "balance_loss_mlp": 1.01552975,
      "epoch": 0.8992965790897613,
      "flos": 25119191168640.0,
      "grad_norm": 3.2603954814179574,
      "language_loss": 0.85709375,
      "learning_rate": 1.053378781137808e-07,
      "loss": 0.8786431,
      "num_input_tokens_seen": 161683480,
      "step": 7479,
      "time_per_iteration": 2.7688019275665283
    },
    {
      "auxiliary_loss_clip": 0.01131439,
      "auxiliary_loss_mlp": 0.01025889,
      "balance_loss_clip": 1.03858829,
      "balance_loss_mlp": 1.01780963,
      "epoch": 0.8994168219804004,
      "flos": 16070815814400.0,
      "grad_norm": 2.083424002449527,
      "language_loss": 0.77731317,
      "learning_rate": 1.0508855153977392e-07,
      "loss": 0.79888642,
      "num_input_tokens_seen": 161699945,
      "step": 7480,
      "time_per_iteration": 3.628556489944458
    },
    {
      "auxiliary_loss_clip": 0.01112732,
      "auxiliary_loss_mlp": 0.01021022,
      "balance_loss_clip": 1.04048872,
      "balance_loss_mlp": 1.01355326,
      "epoch": 0.8995370648710395,
      "flos": 24825764966400.0,
      "grad_norm": 2.4509073650244564,
      "language_loss": 0.67282712,
      "learning_rate": 1.0483951242232669e-07,
      "loss": 0.69416469,
      "num_input_tokens_seen": 161720420,
      "step": 7481,
      "time_per_iteration": 2.776324510574341
    },
    {
      "auxiliary_loss_clip": 0.01017753,
      "auxiliary_loss_mlp": 0.01002131,
      "balance_loss_clip": 1.01272941,
      "balance_loss_mlp": 1.00115907,
      "epoch": 0.8996573077616786,
      "flos": 63116238378240.0,
      "grad_norm": 0.9866138761003209,
      "language_loss": 0.57694179,
      "learning_rate": 1.0459076079921936e-07,
      "loss": 0.59714067,
      "num_input_tokens_seen": 161773080,
      "step": 7482,
      "time_per_iteration": 3.513566493988037
    },
    {
      "auxiliary_loss_clip": 0.01129546,
      "auxiliary_loss_mlp": 0.01023921,
      "balance_loss_clip": 1.04563212,
      "balance_loss_mlp": 1.01662517,
      "epoch": 0.8997775506523177,
      "flos": 18219674027520.0,
      "grad_norm": 2.6445998016160677,
      "language_loss": 0.85342628,
      "learning_rate": 1.0434229670818618e-07,
      "loss": 0.8749609,
      "num_input_tokens_seen": 161789755,
      "step": 7483,
      "time_per_iteration": 2.6791434288024902
    },
    {
      "auxiliary_loss_clip": 0.01139114,
      "auxiliary_loss_mlp": 0.01024003,
      "balance_loss_clip": 1.04395413,
      "balance_loss_mlp": 1.01668024,
      "epoch": 0.8998977935429567,
      "flos": 24166768095360.0,
      "grad_norm": 1.6539155701345858,
      "language_loss": 0.80135965,
      "learning_rate": 1.0409412018691944e-07,
      "loss": 0.82299078,
      "num_input_tokens_seen": 161810220,
      "step": 7484,
      "time_per_iteration": 2.7249608039855957
    },
    {
      "auxiliary_loss_clip": 0.01141611,
      "auxiliary_loss_mlp": 0.01022093,
      "balance_loss_clip": 1.04740858,
      "balance_loss_mlp": 1.01356387,
      "epoch": 0.9000180364335959,
      "flos": 20773030273920.0,
      "grad_norm": 1.9829631833255568,
      "language_loss": 0.7467972,
      "learning_rate": 1.0384623127306724e-07,
      "loss": 0.76843429,
      "num_input_tokens_seen": 161827565,
      "step": 7485,
      "time_per_iteration": 2.679743766784668
    },
    {
      "auxiliary_loss_clip": 0.01127283,
      "auxiliary_loss_mlp": 0.01024476,
      "balance_loss_clip": 1.04614377,
      "balance_loss_mlp": 1.01738906,
      "epoch": 0.900138279324235,
      "flos": 19205745166080.0,
      "grad_norm": 1.934202693427422,
      "language_loss": 0.7959373,
      "learning_rate": 1.0359863000423397e-07,
      "loss": 0.81745493,
      "num_input_tokens_seen": 161845700,
      "step": 7486,
      "time_per_iteration": 2.7321155071258545
    },
    {
      "auxiliary_loss_clip": 0.01127115,
      "auxiliary_loss_mlp": 0.01024868,
      "balance_loss_clip": 1.04145849,
      "balance_loss_mlp": 1.01741481,
      "epoch": 0.900258522214874,
      "flos": 28731158069760.0,
      "grad_norm": 1.6878053333038407,
      "language_loss": 0.7185905,
      "learning_rate": 1.0335131641798112e-07,
      "loss": 0.74011034,
      "num_input_tokens_seen": 161867660,
      "step": 7487,
      "time_per_iteration": 2.783708333969116
    },
    {
      "auxiliary_loss_clip": 0.01045026,
      "auxiliary_loss_mlp": 0.01005424,
      "balance_loss_clip": 1.02354693,
      "balance_loss_mlp": 1.00441074,
      "epoch": 0.9003787651055132,
      "flos": 58280685655680.0,
      "grad_norm": 0.8121410259187307,
      "language_loss": 0.55538386,
      "learning_rate": 1.0310429055182512e-07,
      "loss": 0.57588834,
      "num_input_tokens_seen": 161921980,
      "step": 7488,
      "time_per_iteration": 3.128176212310791
    },
    {
      "auxiliary_loss_clip": 0.01149468,
      "auxiliary_loss_mlp": 0.01022589,
      "balance_loss_clip": 1.04600239,
      "balance_loss_mlp": 1.01442337,
      "epoch": 0.9004990079961522,
      "flos": 25556475340800.0,
      "grad_norm": 2.2432062875992025,
      "language_loss": 0.74333346,
      "learning_rate": 1.0285755244324024e-07,
      "loss": 0.76505405,
      "num_input_tokens_seen": 161942725,
      "step": 7489,
      "time_per_iteration": 2.8008952140808105
    },
    {
      "auxiliary_loss_clip": 0.01124144,
      "auxiliary_loss_mlp": 0.01021065,
      "balance_loss_clip": 1.04150486,
      "balance_loss_mlp": 1.01382613,
      "epoch": 0.9006192508867913,
      "flos": 23335185352320.0,
      "grad_norm": 1.8196381420175591,
      "language_loss": 0.68736744,
      "learning_rate": 1.0261110212965629e-07,
      "loss": 0.70881951,
      "num_input_tokens_seen": 161964520,
      "step": 7490,
      "time_per_iteration": 3.6951022148132324
    },
    {
      "auxiliary_loss_clip": 0.0109654,
      "auxiliary_loss_mlp": 0.00762098,
      "balance_loss_clip": 1.04032791,
      "balance_loss_mlp": 1.00050104,
      "epoch": 0.9007394937774305,
      "flos": 18040300485120.0,
      "grad_norm": 2.0026202178253056,
      "language_loss": 0.79566306,
      "learning_rate": 1.023649396484596e-07,
      "loss": 0.8142494,
      "num_input_tokens_seen": 161983575,
      "step": 7491,
      "time_per_iteration": 2.824953556060791
    },
    {
      "auxiliary_loss_clip": 0.01138169,
      "auxiliary_loss_mlp": 0.0102551,
      "balance_loss_clip": 1.04553699,
      "balance_loss_mlp": 1.01722503,
      "epoch": 0.9008597366680695,
      "flos": 43068456633600.0,
      "grad_norm": 2.113274381726441,
      "language_loss": 0.67786884,
      "learning_rate": 1.0211906503699275e-07,
      "loss": 0.69950563,
      "num_input_tokens_seen": 162006550,
      "step": 7492,
      "time_per_iteration": 2.936836004257202
    },
    {
      "auxiliary_loss_clip": 0.01158463,
      "auxiliary_loss_mlp": 0.0102833,
      "balance_loss_clip": 1.04972625,
      "balance_loss_mlp": 1.02046251,
      "epoch": 0.9009799795587086,
      "flos": 14939055112320.0,
      "grad_norm": 2.776369957727004,
      "language_loss": 0.82362843,
      "learning_rate": 1.0187347833255455e-07,
      "loss": 0.84549636,
      "num_input_tokens_seen": 162022455,
      "step": 7493,
      "time_per_iteration": 2.6903417110443115
    },
    {
      "auxiliary_loss_clip": 0.011536,
      "auxiliary_loss_mlp": 0.01024368,
      "balance_loss_clip": 1.04520297,
      "balance_loss_mlp": 1.01700091,
      "epoch": 0.9011002224493477,
      "flos": 21579584215680.0,
      "grad_norm": 2.1876648197841986,
      "language_loss": 0.79252785,
      "learning_rate": 1.0162817957240056e-07,
      "loss": 0.81430751,
      "num_input_tokens_seen": 162042350,
      "step": 7494,
      "time_per_iteration": 2.7245917320251465
    },
    {
      "auxiliary_loss_clip": 0.01020091,
      "auxiliary_loss_mlp": 0.00754185,
      "balance_loss_clip": 1.01432204,
      "balance_loss_mlp": 1.00086117,
      "epoch": 0.9012204653399868,
      "flos": 71166367883520.0,
      "grad_norm": 0.8825161038182414,
      "language_loss": 0.63003075,
      "learning_rate": 1.0138316879374253e-07,
      "loss": 0.6477735,
      "num_input_tokens_seen": 162111640,
      "step": 7495,
      "time_per_iteration": 3.439812660217285
    },
    {
      "auxiliary_loss_clip": 0.01095583,
      "auxiliary_loss_mlp": 0.01027797,
      "balance_loss_clip": 1.04104638,
      "balance_loss_mlp": 1.01991761,
      "epoch": 0.9013407082306258,
      "flos": 15594963413760.0,
      "grad_norm": 2.442314465234192,
      "language_loss": 0.74339354,
      "learning_rate": 1.0113844603374833e-07,
      "loss": 0.76462734,
      "num_input_tokens_seen": 162128165,
      "step": 7496,
      "time_per_iteration": 2.784722089767456
    },
    {
      "auxiliary_loss_clip": 0.01138809,
      "auxiliary_loss_mlp": 0.01031561,
      "balance_loss_clip": 1.04410684,
      "balance_loss_mlp": 1.0234313,
      "epoch": 0.901460951121265,
      "flos": 15049157276160.0,
      "grad_norm": 2.669297569516453,
      "language_loss": 0.72026587,
      "learning_rate": 1.0089401132954178e-07,
      "loss": 0.74196959,
      "num_input_tokens_seen": 162146145,
      "step": 7497,
      "time_per_iteration": 2.6665756702423096
    },
    {
      "auxiliary_loss_clip": 0.01119376,
      "auxiliary_loss_mlp": 0.01026199,
      "balance_loss_clip": 1.04121959,
      "balance_loss_mlp": 1.01846838,
      "epoch": 0.9015811940119041,
      "flos": 22236857233920.0,
      "grad_norm": 1.7420183697194178,
      "language_loss": 0.7224766,
      "learning_rate": 1.006498647182037e-07,
      "loss": 0.74393231,
      "num_input_tokens_seen": 162164800,
      "step": 7498,
      "time_per_iteration": 2.8161561489105225
    },
    {
      "auxiliary_loss_clip": 0.01139937,
      "auxiliary_loss_mlp": 0.01026281,
      "balance_loss_clip": 1.04455125,
      "balance_loss_mlp": 1.01844311,
      "epoch": 0.9017014369025431,
      "flos": 24973824827520.0,
      "grad_norm": 3.866393642898566,
      "language_loss": 0.71394962,
      "learning_rate": 1.004060062367713e-07,
      "loss": 0.7356118,
      "num_input_tokens_seen": 162185895,
      "step": 7499,
      "time_per_iteration": 2.710590124130249
    },
    {
      "auxiliary_loss_clip": 0.01120404,
      "auxiliary_loss_mlp": 0.01025497,
      "balance_loss_clip": 1.03902614,
      "balance_loss_mlp": 1.01811171,
      "epoch": 0.9018216797931822,
      "flos": 18114168804480.0,
      "grad_norm": 1.7158069967384784,
      "language_loss": 0.69425946,
      "learning_rate": 1.0016243592223728e-07,
      "loss": 0.71571851,
      "num_input_tokens_seen": 162206295,
      "step": 7500,
      "time_per_iteration": 2.751634120941162
    },
    {
      "auxiliary_loss_clip": 0.01115981,
      "auxiliary_loss_mlp": 0.01026623,
      "balance_loss_clip": 1.04012799,
      "balance_loss_mlp": 1.01899409,
      "epoch": 0.9019419226838213,
      "flos": 37268452759680.0,
      "grad_norm": 2.027070297357289,
      "language_loss": 0.65897799,
      "learning_rate": 9.991915381155114e-08,
      "loss": 0.68040401,
      "num_input_tokens_seen": 162229275,
      "step": 7501,
      "time_per_iteration": 2.940398693084717
    },
    {
      "auxiliary_loss_clip": 0.01117958,
      "auxiliary_loss_mlp": 0.01027769,
      "balance_loss_clip": 1.04148316,
      "balance_loss_mlp": 1.01995516,
      "epoch": 0.9020621655744604,
      "flos": 23441121538560.0,
      "grad_norm": 7.695630644132355,
      "language_loss": 0.7525202,
      "learning_rate": 9.967615994161871e-08,
      "loss": 0.77397746,
      "num_input_tokens_seen": 162248935,
      "step": 7502,
      "time_per_iteration": 2.6977715492248535
    },
    {
      "auxiliary_loss_clip": 0.01135299,
      "auxiliary_loss_mlp": 0.01024304,
      "balance_loss_clip": 1.04405808,
      "balance_loss_mlp": 1.01645446,
      "epoch": 0.9021824084650995,
      "flos": 22857465444480.0,
      "grad_norm": 1.8172004231656038,
      "language_loss": 0.78635919,
      "learning_rate": 9.943345434930161e-08,
      "loss": 0.80795521,
      "num_input_tokens_seen": 162269185,
      "step": 7503,
      "time_per_iteration": 3.6809804439544678
    },
    {
      "auxiliary_loss_clip": 0.01154579,
      "auxiliary_loss_mlp": 0.01026737,
      "balance_loss_clip": 1.04722714,
      "balance_loss_mlp": 1.01945937,
      "epoch": 0.9023026513557386,
      "flos": 22127581082880.0,
      "grad_norm": 1.9860026151340142,
      "language_loss": 0.68740541,
      "learning_rate": 9.919103707141885e-08,
      "loss": 0.70921856,
      "num_input_tokens_seen": 162288065,
      "step": 7504,
      "time_per_iteration": 3.559194326400757
    },
    {
      "auxiliary_loss_clip": 0.01135549,
      "auxiliary_loss_mlp": 0.01026955,
      "balance_loss_clip": 1.0447129,
      "balance_loss_mlp": 1.01839626,
      "epoch": 0.9024228942463777,
      "flos": 24199087357440.0,
      "grad_norm": 3.8557548965077877,
      "language_loss": 0.76883829,
      "learning_rate": 9.89489081447441e-08,
      "loss": 0.79046333,
      "num_input_tokens_seen": 162305265,
      "step": 7505,
      "time_per_iteration": 2.8091256618499756
    },
    {
      "auxiliary_loss_clip": 0.01145835,
      "auxiliary_loss_mlp": 0.00762571,
      "balance_loss_clip": 1.04529929,
      "balance_loss_mlp": 1.00049162,
      "epoch": 0.9025431371370167,
      "flos": 25008262992000.0,
      "grad_norm": 1.8282324755269443,
      "language_loss": 0.82824916,
      "learning_rate": 9.870706760600844e-08,
      "loss": 0.84733325,
      "num_input_tokens_seen": 162325215,
      "step": 7506,
      "time_per_iteration": 3.713393449783325
    },
    {
      "auxiliary_loss_clip": 0.01163658,
      "auxiliary_loss_mlp": 0.01030439,
      "balance_loss_clip": 1.04959941,
      "balance_loss_mlp": 1.02234805,
      "epoch": 0.9026633800276559,
      "flos": 18952862440320.0,
      "grad_norm": 1.801378292433967,
      "language_loss": 0.72914994,
      "learning_rate": 9.846551549189918e-08,
      "loss": 0.75109088,
      "num_input_tokens_seen": 162344820,
      "step": 7507,
      "time_per_iteration": 2.683391809463501
    },
    {
      "auxiliary_loss_clip": 0.01097247,
      "auxiliary_loss_mlp": 0.01025343,
      "balance_loss_clip": 1.04057276,
      "balance_loss_mlp": 1.01775527,
      "epoch": 0.902783622918295,
      "flos": 32416059536640.0,
      "grad_norm": 12.97634121009251,
      "language_loss": 0.68861961,
      "learning_rate": 9.822425183905902e-08,
      "loss": 0.70984548,
      "num_input_tokens_seen": 162365345,
      "step": 7508,
      "time_per_iteration": 2.8907907009124756
    },
    {
      "auxiliary_loss_clip": 0.0104322,
      "auxiliary_loss_mlp": 0.01001159,
      "balance_loss_clip": 1.00915432,
      "balance_loss_mlp": 1.00017536,
      "epoch": 0.902903865808934,
      "flos": 63717453244800.0,
      "grad_norm": 0.9253215498525688,
      "language_loss": 0.75205171,
      "learning_rate": 9.798327668408823e-08,
      "loss": 0.77249551,
      "num_input_tokens_seen": 162426980,
      "step": 7509,
      "time_per_iteration": 3.3193717002868652
    },
    {
      "auxiliary_loss_clip": 0.01127423,
      "auxiliary_loss_mlp": 0.01030826,
      "balance_loss_clip": 1.04717875,
      "balance_loss_mlp": 1.02224278,
      "epoch": 0.9030241086995732,
      "flos": 23804034600960.0,
      "grad_norm": 2.7924671800282335,
      "language_loss": 0.68939817,
      "learning_rate": 9.774259006354158e-08,
      "loss": 0.71098065,
      "num_input_tokens_seen": 162447050,
      "step": 7510,
      "time_per_iteration": 2.831115245819092
    },
    {
      "auxiliary_loss_clip": 0.01140294,
      "auxiliary_loss_mlp": 0.01024016,
      "balance_loss_clip": 1.04585254,
      "balance_loss_mlp": 1.01572764,
      "epoch": 0.9031443515902122,
      "flos": 26395887248640.0,
      "grad_norm": 2.04519840114275,
      "language_loss": 0.76169276,
      "learning_rate": 9.750219201393184e-08,
      "loss": 0.7833358,
      "num_input_tokens_seen": 162467015,
      "step": 7511,
      "time_per_iteration": 2.9169421195983887
    },
    {
      "auxiliary_loss_clip": 0.01119472,
      "auxiliary_loss_mlp": 0.01023909,
      "balance_loss_clip": 1.04133248,
      "balance_loss_mlp": 1.0159936,
      "epoch": 0.9032645944808513,
      "flos": 24939350749440.0,
      "grad_norm": 3.5367828140967155,
      "language_loss": 0.77993953,
      "learning_rate": 9.726208257172697e-08,
      "loss": 0.80137336,
      "num_input_tokens_seen": 162488710,
      "step": 7512,
      "time_per_iteration": 2.7401249408721924
    },
    {
      "auxiliary_loss_clip": 0.01104898,
      "auxiliary_loss_mlp": 0.01021189,
      "balance_loss_clip": 1.04411542,
      "balance_loss_mlp": 1.01354814,
      "epoch": 0.9033848373714904,
      "flos": 21178821196800.0,
      "grad_norm": 2.378782705608996,
      "language_loss": 0.74706411,
      "learning_rate": 9.702226177335115e-08,
      "loss": 0.76832497,
      "num_input_tokens_seen": 162507205,
      "step": 7513,
      "time_per_iteration": 2.828240156173706
    },
    {
      "auxiliary_loss_clip": 0.01149855,
      "auxiliary_loss_mlp": 0.01030431,
      "balance_loss_clip": 1.04542708,
      "balance_loss_mlp": 1.02281976,
      "epoch": 0.9035050802621295,
      "flos": 26286359702400.0,
      "grad_norm": 1.609667671065519,
      "language_loss": 0.72585678,
      "learning_rate": 9.67827296551853e-08,
      "loss": 0.74765968,
      "num_input_tokens_seen": 162528490,
      "step": 7514,
      "time_per_iteration": 2.6734778881073
    },
    {
      "auxiliary_loss_clip": 0.01143829,
      "auxiliary_loss_mlp": 0.01024439,
      "balance_loss_clip": 1.04661798,
      "balance_loss_mlp": 1.01684201,
      "epoch": 0.9036253231527686,
      "flos": 24204546224640.0,
      "grad_norm": 3.344764508062149,
      "language_loss": 0.68540341,
      "learning_rate": 9.65434862535659e-08,
      "loss": 0.70708609,
      "num_input_tokens_seen": 162547860,
      "step": 7515,
      "time_per_iteration": 2.6596577167510986
    },
    {
      "auxiliary_loss_clip": 0.01134424,
      "auxiliary_loss_mlp": 0.01022445,
      "balance_loss_clip": 1.0412246,
      "balance_loss_mlp": 1.01379061,
      "epoch": 0.9037455660434077,
      "flos": 18072655660800.0,
      "grad_norm": 2.958972496624188,
      "language_loss": 0.65162963,
      "learning_rate": 9.630453160478635e-08,
      "loss": 0.67319834,
      "num_input_tokens_seen": 162563215,
      "step": 7516,
      "time_per_iteration": 3.5394465923309326
    },
    {
      "auxiliary_loss_clip": 0.01137056,
      "auxiliary_loss_mlp": 0.00761617,
      "balance_loss_clip": 1.0445435,
      "balance_loss_mlp": 1.00044012,
      "epoch": 0.9038658089340468,
      "flos": 24060795995520.0,
      "grad_norm": 1.7118704582884667,
      "language_loss": 0.82341564,
      "learning_rate": 9.60658657450959e-08,
      "loss": 0.8424024,
      "num_input_tokens_seen": 162583515,
      "step": 7517,
      "time_per_iteration": 2.754115581512451
    },
    {
      "auxiliary_loss_clip": 0.01115132,
      "auxiliary_loss_mlp": 0.01025359,
      "balance_loss_clip": 1.03718936,
      "balance_loss_mlp": 1.01762247,
      "epoch": 0.9039860518246858,
      "flos": 21834298535040.0,
      "grad_norm": 1.879885352890459,
      "language_loss": 0.79456949,
      "learning_rate": 9.582748871069979e-08,
      "loss": 0.81597435,
      "num_input_tokens_seen": 162602955,
      "step": 7518,
      "time_per_iteration": 2.7351417541503906
    },
    {
      "auxiliary_loss_clip": 0.01101991,
      "auxiliary_loss_mlp": 0.01025194,
      "balance_loss_clip": 1.0388999,
      "balance_loss_mlp": 1.01744604,
      "epoch": 0.904106294715325,
      "flos": 26614870513920.0,
      "grad_norm": 2.38869850490198,
      "language_loss": 0.83474976,
      "learning_rate": 9.558940053775954e-08,
      "loss": 0.85602164,
      "num_input_tokens_seen": 162621595,
      "step": 7519,
      "time_per_iteration": 2.7584476470947266
    },
    {
      "auxiliary_loss_clip": 0.01136934,
      "auxiliary_loss_mlp": 0.01031401,
      "balance_loss_clip": 1.04642713,
      "balance_loss_mlp": 1.02346814,
      "epoch": 0.904226537605964,
      "flos": 17785693906560.0,
      "grad_norm": 2.567479090083006,
      "language_loss": 0.68258202,
      "learning_rate": 9.535160126239294e-08,
      "loss": 0.70426536,
      "num_input_tokens_seen": 162638220,
      "step": 7520,
      "time_per_iteration": 2.6546380519866943
    },
    {
      "auxiliary_loss_clip": 0.01109878,
      "auxiliary_loss_mlp": 0.01025292,
      "balance_loss_clip": 1.04003787,
      "balance_loss_mlp": 1.01750183,
      "epoch": 0.9043467804966031,
      "flos": 24790428961920.0,
      "grad_norm": 1.618643476342625,
      "language_loss": 0.70755702,
      "learning_rate": 9.511409092067424e-08,
      "loss": 0.72890878,
      "num_input_tokens_seen": 162658575,
      "step": 7521,
      "time_per_iteration": 2.800341844558716
    },
    {
      "auxiliary_loss_clip": 0.01060952,
      "auxiliary_loss_mlp": 0.01026322,
      "balance_loss_clip": 1.03255033,
      "balance_loss_mlp": 1.01890171,
      "epoch": 0.9044670233872423,
      "flos": 22632125472000.0,
      "grad_norm": 1.9156202391324137,
      "language_loss": 0.67419547,
      "learning_rate": 9.487686954863327e-08,
      "loss": 0.69506818,
      "num_input_tokens_seen": 162678295,
      "step": 7522,
      "time_per_iteration": 3.1847023963928223
    },
    {
      "auxiliary_loss_clip": 0.01124702,
      "auxiliary_loss_mlp": 0.01027369,
      "balance_loss_clip": 1.0437398,
      "balance_loss_mlp": 1.01985872,
      "epoch": 0.9045872662778813,
      "flos": 23771320289280.0,
      "grad_norm": 2.207349767420366,
      "language_loss": 0.77108634,
      "learning_rate": 9.46399371822566e-08,
      "loss": 0.79260707,
      "num_input_tokens_seen": 162698070,
      "step": 7523,
      "time_per_iteration": 2.959568977355957
    },
    {
      "auxiliary_loss_clip": 0.01120182,
      "auxiliary_loss_mlp": 0.01022487,
      "balance_loss_clip": 1.04157579,
      "balance_loss_mlp": 1.0144459,
      "epoch": 0.9047075091685204,
      "flos": 15191039998080.0,
      "grad_norm": 3.3202024179581096,
      "language_loss": 0.72424746,
      "learning_rate": 9.440329385748657e-08,
      "loss": 0.74567413,
      "num_input_tokens_seen": 162715140,
      "step": 7524,
      "time_per_iteration": 2.7476396560668945
    },
    {
      "auxiliary_loss_clip": 0.01117025,
      "auxiliary_loss_mlp": 0.01023042,
      "balance_loss_clip": 1.04194474,
      "balance_loss_mlp": 1.0157702,
      "epoch": 0.9048277520591596,
      "flos": 18003707504640.0,
      "grad_norm": 1.9454642377904607,
      "language_loss": 0.70888704,
      "learning_rate": 9.416693961022137e-08,
      "loss": 0.73028767,
      "num_input_tokens_seen": 162733390,
      "step": 7525,
      "time_per_iteration": 2.7680411338806152
    },
    {
      "auxiliary_loss_clip": 0.01105895,
      "auxiliary_loss_mlp": 0.01020857,
      "balance_loss_clip": 1.03799951,
      "balance_loss_mlp": 1.01294112,
      "epoch": 0.9049479949497986,
      "flos": 21872471713920.0,
      "grad_norm": 1.984991316718632,
      "language_loss": 0.76990426,
      "learning_rate": 9.393087447631654e-08,
      "loss": 0.79117179,
      "num_input_tokens_seen": 162751670,
      "step": 7526,
      "time_per_iteration": 2.7666807174682617
    },
    {
      "auxiliary_loss_clip": 0.01074049,
      "auxiliary_loss_mlp": 0.01021121,
      "balance_loss_clip": 1.03391266,
      "balance_loss_mlp": 1.01383162,
      "epoch": 0.9050682378404377,
      "flos": 20773928113920.0,
      "grad_norm": 1.8299103834637487,
      "language_loss": 0.72410899,
      "learning_rate": 9.36950984915823e-08,
      "loss": 0.74506068,
      "num_input_tokens_seen": 162770025,
      "step": 7527,
      "time_per_iteration": 2.8600287437438965
    },
    {
      "auxiliary_loss_clip": 0.01136734,
      "auxiliary_loss_mlp": 0.01031546,
      "balance_loss_clip": 1.04673862,
      "balance_loss_mlp": 1.02322555,
      "epoch": 0.9051884807310768,
      "flos": 21580015178880.0,
      "grad_norm": 1.9085333707726586,
      "language_loss": 0.69215298,
      "learning_rate": 9.345961169178607e-08,
      "loss": 0.71383584,
      "num_input_tokens_seen": 162789710,
      "step": 7528,
      "time_per_iteration": 2.7239511013031006
    },
    {
      "auxiliary_loss_clip": 0.01142477,
      "auxiliary_loss_mlp": 0.01027718,
      "balance_loss_clip": 1.04386556,
      "balance_loss_mlp": 1.02029979,
      "epoch": 0.9053087236217159,
      "flos": 21908059113600.0,
      "grad_norm": 1.4639759099272218,
      "language_loss": 0.72734189,
      "learning_rate": 9.322441411265081e-08,
      "loss": 0.74904382,
      "num_input_tokens_seen": 162810695,
      "step": 7529,
      "time_per_iteration": 3.774233341217041
    },
    {
      "auxiliary_loss_clip": 0.01115296,
      "auxiliary_loss_mlp": 0.01023876,
      "balance_loss_clip": 1.03933132,
      "balance_loss_mlp": 1.01635718,
      "epoch": 0.9054289665123549,
      "flos": 17055809544960.0,
      "grad_norm": 1.9880938152940713,
      "language_loss": 0.73522896,
      "learning_rate": 9.298950578985554e-08,
      "loss": 0.75662065,
      "num_input_tokens_seen": 162827770,
      "step": 7530,
      "time_per_iteration": 3.6769492626190186
    },
    {
      "auxiliary_loss_clip": 0.01142316,
      "auxiliary_loss_mlp": 0.01027843,
      "balance_loss_clip": 1.04456091,
      "balance_loss_mlp": 1.02015352,
      "epoch": 0.905549209402994,
      "flos": 20777268078720.0,
      "grad_norm": 2.4384093797713637,
      "language_loss": 0.70855796,
      "learning_rate": 9.275488675903665e-08,
      "loss": 0.73025954,
      "num_input_tokens_seen": 162846715,
      "step": 7531,
      "time_per_iteration": 2.7275359630584717
    },
    {
      "auxiliary_loss_clip": 0.01140305,
      "auxiliary_loss_mlp": 0.0102505,
      "balance_loss_clip": 1.0475421,
      "balance_loss_mlp": 1.01681304,
      "epoch": 0.9056694522936332,
      "flos": 21686813291520.0,
      "grad_norm": 2.054725613958685,
      "language_loss": 0.73956752,
      "learning_rate": 9.252055705578454e-08,
      "loss": 0.76122105,
      "num_input_tokens_seen": 162866215,
      "step": 7532,
      "time_per_iteration": 3.6325745582580566
    },
    {
      "auxiliary_loss_clip": 0.01150415,
      "auxiliary_loss_mlp": 0.01024227,
      "balance_loss_clip": 1.04443216,
      "balance_loss_mlp": 1.0175004,
      "epoch": 0.9057896951842722,
      "flos": 29569133433600.0,
      "grad_norm": 2.152193773181579,
      "language_loss": 0.72137678,
      "learning_rate": 9.228651671564747e-08,
      "loss": 0.74312317,
      "num_input_tokens_seen": 162888245,
      "step": 7533,
      "time_per_iteration": 2.70854115486145
    },
    {
      "auxiliary_loss_clip": 0.01079876,
      "auxiliary_loss_mlp": 0.01030266,
      "balance_loss_clip": 1.03566158,
      "balance_loss_mlp": 1.02273512,
      "epoch": 0.9059099380749113,
      "flos": 27892248952320.0,
      "grad_norm": 1.5618292189491734,
      "language_loss": 0.77753025,
      "learning_rate": 9.205276577412901e-08,
      "loss": 0.79863167,
      "num_input_tokens_seen": 162911025,
      "step": 7534,
      "time_per_iteration": 2.900083303451538
    },
    {
      "auxiliary_loss_clip": 0.01073063,
      "auxiliary_loss_mlp": 0.00763232,
      "balance_loss_clip": 1.03830934,
      "balance_loss_mlp": 1.00042236,
      "epoch": 0.9060301809655504,
      "flos": 17748993185280.0,
      "grad_norm": 2.663207801198533,
      "language_loss": 0.77310109,
      "learning_rate": 9.181930426668905e-08,
      "loss": 0.79146409,
      "num_input_tokens_seen": 162927820,
      "step": 7535,
      "time_per_iteration": 2.8961925506591797
    },
    {
      "auxiliary_loss_clip": 0.01153554,
      "auxiliary_loss_mlp": 0.01025336,
      "balance_loss_clip": 1.04637218,
      "balance_loss_mlp": 1.0175519,
      "epoch": 0.9061504238561895,
      "flos": 31759432963200.0,
      "grad_norm": 1.746833266621036,
      "language_loss": 0.67459315,
      "learning_rate": 9.158613222874346e-08,
      "loss": 0.69638211,
      "num_input_tokens_seen": 162949445,
      "step": 7536,
      "time_per_iteration": 2.764326333999634
    },
    {
      "auxiliary_loss_clip": 0.01138919,
      "auxiliary_loss_mlp": 0.01021166,
      "balance_loss_clip": 1.04443383,
      "balance_loss_mlp": 1.01386476,
      "epoch": 0.9062706667468285,
      "flos": 20048066075520.0,
      "grad_norm": 1.6650422863432284,
      "language_loss": 0.82207668,
      "learning_rate": 9.135324969566394e-08,
      "loss": 0.84367752,
      "num_input_tokens_seen": 162968945,
      "step": 7537,
      "time_per_iteration": 2.7093489170074463
    },
    {
      "auxiliary_loss_clip": 0.01137788,
      "auxiliary_loss_mlp": 0.01027927,
      "balance_loss_clip": 1.04529822,
      "balance_loss_mlp": 1.01936817,
      "epoch": 0.9063909096374677,
      "flos": 18437292576000.0,
      "grad_norm": 2.8435712569128584,
      "language_loss": 0.75574911,
      "learning_rate": 9.112065670277913e-08,
      "loss": 0.77740622,
      "num_input_tokens_seen": 162985310,
      "step": 7538,
      "time_per_iteration": 2.631361484527588
    },
    {
      "auxiliary_loss_clip": 0.01122563,
      "auxiliary_loss_mlp": 0.01022334,
      "balance_loss_clip": 1.03964424,
      "balance_loss_mlp": 1.01517606,
      "epoch": 0.9065111525281068,
      "flos": 33547353361920.0,
      "grad_norm": 1.8417051420171264,
      "language_loss": 0.72975492,
      "learning_rate": 9.088835328537303e-08,
      "loss": 0.75120389,
      "num_input_tokens_seen": 163006900,
      "step": 7539,
      "time_per_iteration": 2.8192079067230225
    },
    {
      "auxiliary_loss_clip": 0.01139157,
      "auxiliary_loss_mlp": 0.0102674,
      "balance_loss_clip": 1.04362154,
      "balance_loss_mlp": 1.01904464,
      "epoch": 0.9066313954187458,
      "flos": 23367863750400.0,
      "grad_norm": 2.848847837511454,
      "language_loss": 0.71676469,
      "learning_rate": 9.065633947868568e-08,
      "loss": 0.73842365,
      "num_input_tokens_seen": 163026505,
      "step": 7540,
      "time_per_iteration": 2.695504665374756
    },
    {
      "auxiliary_loss_clip": 0.0111213,
      "auxiliary_loss_mlp": 0.01022763,
      "balance_loss_clip": 1.04023993,
      "balance_loss_mlp": 1.01554537,
      "epoch": 0.906751638309385,
      "flos": 26249623067520.0,
      "grad_norm": 3.4960317865338033,
      "language_loss": 0.80206102,
      "learning_rate": 9.042461531791379e-08,
      "loss": 0.82340991,
      "num_input_tokens_seen": 163044925,
      "step": 7541,
      "time_per_iteration": 4.300300598144531
    },
    {
      "auxiliary_loss_clip": 0.01103785,
      "auxiliary_loss_mlp": 0.01020781,
      "balance_loss_clip": 1.0395031,
      "balance_loss_mlp": 1.01389647,
      "epoch": 0.906871881200024,
      "flos": 16544477485440.0,
      "grad_norm": 1.8551090019615604,
      "language_loss": 0.77997398,
      "learning_rate": 9.019318083820903e-08,
      "loss": 0.80121964,
      "num_input_tokens_seen": 163063505,
      "step": 7542,
      "time_per_iteration": 2.95216965675354
    },
    {
      "auxiliary_loss_clip": 0.01154233,
      "auxiliary_loss_mlp": 0.01025859,
      "balance_loss_clip": 1.04551792,
      "balance_loss_mlp": 1.01783049,
      "epoch": 0.9069921240906631,
      "flos": 24605129675520.0,
      "grad_norm": 1.8917762904770927,
      "language_loss": 0.85336381,
      "learning_rate": 8.996203607468045e-08,
      "loss": 0.87516475,
      "num_input_tokens_seen": 163082505,
      "step": 7543,
      "time_per_iteration": 2.670011281967163
    },
    {
      "auxiliary_loss_clip": 0.01137868,
      "auxiliary_loss_mlp": 0.01023735,
      "balance_loss_clip": 1.04366231,
      "balance_loss_mlp": 1.01655245,
      "epoch": 0.9071123669813023,
      "flos": 25374731500800.0,
      "grad_norm": 1.8659791231235612,
      "language_loss": 0.75522876,
      "learning_rate": 8.973118106239241e-08,
      "loss": 0.7768448,
      "num_input_tokens_seen": 163105110,
      "step": 7544,
      "time_per_iteration": 2.703784704208374
    },
    {
      "auxiliary_loss_clip": 0.01128365,
      "auxiliary_loss_mlp": 0.01026553,
      "balance_loss_clip": 1.04080713,
      "balance_loss_mlp": 1.01914072,
      "epoch": 0.9072326098719413,
      "flos": 26725798690560.0,
      "grad_norm": 3.021045377629564,
      "language_loss": 0.95115578,
      "learning_rate": 8.95006158363656e-08,
      "loss": 0.97270501,
      "num_input_tokens_seen": 163125295,
      "step": 7545,
      "time_per_iteration": 2.7814207077026367
    },
    {
      "auxiliary_loss_clip": 0.01140379,
      "auxiliary_loss_mlp": 0.01028991,
      "balance_loss_clip": 1.04763424,
      "balance_loss_mlp": 1.02055717,
      "epoch": 0.9073528527625804,
      "flos": 23878800760320.0,
      "grad_norm": 1.8942098071687188,
      "language_loss": 0.77207249,
      "learning_rate": 8.9270340431576e-08,
      "loss": 0.7937662,
      "num_input_tokens_seen": 163144385,
      "step": 7546,
      "time_per_iteration": 2.703707695007324
    },
    {
      "auxiliary_loss_clip": 0.01095236,
      "auxiliary_loss_mlp": 0.01023649,
      "balance_loss_clip": 1.03669024,
      "balance_loss_mlp": 1.01633596,
      "epoch": 0.9074730956532195,
      "flos": 37852144767360.0,
      "grad_norm": 4.792816973936877,
      "language_loss": 0.73567158,
      "learning_rate": 8.904035488295658e-08,
      "loss": 0.75686038,
      "num_input_tokens_seen": 163163885,
      "step": 7547,
      "time_per_iteration": 2.981877326965332
    },
    {
      "auxiliary_loss_clip": 0.01035713,
      "auxiliary_loss_mlp": 0.01003097,
      "balance_loss_clip": 1.0127306,
      "balance_loss_mlp": 1.00216138,
      "epoch": 0.9075933385438586,
      "flos": 65173307385600.0,
      "grad_norm": 0.6583648661184007,
      "language_loss": 0.53181207,
      "learning_rate": 8.881065922539632e-08,
      "loss": 0.5522002,
      "num_input_tokens_seen": 163224325,
      "step": 7548,
      "time_per_iteration": 3.2490172386169434
    },
    {
      "auxiliary_loss_clip": 0.01146985,
      "auxiliary_loss_mlp": 0.01020701,
      "balance_loss_clip": 1.0433687,
      "balance_loss_mlp": 1.01325309,
      "epoch": 0.9077135814344977,
      "flos": 19931571290880.0,
      "grad_norm": 4.297259274907283,
      "language_loss": 0.73326713,
      "learning_rate": 8.85812534937389e-08,
      "loss": 0.75494397,
      "num_input_tokens_seen": 163242425,
      "step": 7549,
      "time_per_iteration": 2.6334433555603027
    },
    {
      "auxiliary_loss_clip": 0.0113364,
      "auxiliary_loss_mlp": 0.0102409,
      "balance_loss_clip": 1.0460391,
      "balance_loss_mlp": 1.01607072,
      "epoch": 0.9078338243251368,
      "flos": 17529650784000.0,
      "grad_norm": 3.045165959887559,
      "language_loss": 0.67587459,
      "learning_rate": 8.835213772278583e-08,
      "loss": 0.69745189,
      "num_input_tokens_seen": 163259280,
      "step": 7550,
      "time_per_iteration": 2.7154598236083984
    },
    {
      "auxiliary_loss_clip": 0.01075671,
      "auxiliary_loss_mlp": 0.01023331,
      "balance_loss_clip": 1.03621602,
      "balance_loss_mlp": 1.01635993,
      "epoch": 0.9079540672157759,
      "flos": 28803410277120.0,
      "grad_norm": 2.149615849892267,
      "language_loss": 0.79007828,
      "learning_rate": 8.812331194729373e-08,
      "loss": 0.8110683,
      "num_input_tokens_seen": 163278925,
      "step": 7551,
      "time_per_iteration": 2.8838558197021484
    },
    {
      "auxiliary_loss_clip": 0.01136791,
      "auxiliary_loss_mlp": 0.01032674,
      "balance_loss_clip": 1.04665291,
      "balance_loss_mlp": 1.02395439,
      "epoch": 0.9080743101064149,
      "flos": 23513840622720.0,
      "grad_norm": 2.271238674751256,
      "language_loss": 0.72337437,
      "learning_rate": 8.789477620197461e-08,
      "loss": 0.74506903,
      "num_input_tokens_seen": 163298450,
      "step": 7552,
      "time_per_iteration": 2.764956474304199
    },
    {
      "auxiliary_loss_clip": 0.01079895,
      "auxiliary_loss_mlp": 0.00762281,
      "balance_loss_clip": 1.03913486,
      "balance_loss_mlp": 1.00048208,
      "epoch": 0.9081945529970541,
      "flos": 22778102344320.0,
      "grad_norm": 3.181695285333396,
      "language_loss": 0.78976804,
      "learning_rate": 8.766653052149831e-08,
      "loss": 0.80818975,
      "num_input_tokens_seen": 163313635,
      "step": 7553,
      "time_per_iteration": 2.833526849746704
    },
    {
      "auxiliary_loss_clip": 0.0113566,
      "auxiliary_loss_mlp": 0.01028455,
      "balance_loss_clip": 1.04200625,
      "balance_loss_mlp": 1.02026534,
      "epoch": 0.9083147958876931,
      "flos": 18873714821760.0,
      "grad_norm": 2.3034596704679577,
      "language_loss": 0.75021946,
      "learning_rate": 8.743857494048823e-08,
      "loss": 0.7718606,
      "num_input_tokens_seen": 163330450,
      "step": 7554,
      "time_per_iteration": 2.7038674354553223
    },
    {
      "auxiliary_loss_clip": 0.01098169,
      "auxiliary_loss_mlp": 0.01019485,
      "balance_loss_clip": 1.04084861,
      "balance_loss_mlp": 1.0120585,
      "epoch": 0.9084350387783322,
      "flos": 18909374048640.0,
      "grad_norm": 1.8979040402794618,
      "language_loss": 0.63210177,
      "learning_rate": 8.721090949352605e-08,
      "loss": 0.65327835,
      "num_input_tokens_seen": 163346690,
      "step": 7555,
      "time_per_iteration": 4.5723114013671875
    },
    {
      "auxiliary_loss_clip": 0.01159806,
      "auxiliary_loss_mlp": 0.01024176,
      "balance_loss_clip": 1.04970026,
      "balance_loss_mlp": 1.01614785,
      "epoch": 0.9085552816689714,
      "flos": 20595488325120.0,
      "grad_norm": 6.386662276702787,
      "language_loss": 0.73135531,
      "learning_rate": 8.698353421514793e-08,
      "loss": 0.75319517,
      "num_input_tokens_seen": 163365065,
      "step": 7556,
      "time_per_iteration": 2.8084661960601807
    },
    {
      "auxiliary_loss_clip": 0.01103162,
      "auxiliary_loss_mlp": 0.01028749,
      "balance_loss_clip": 1.04336274,
      "balance_loss_mlp": 1.02180815,
      "epoch": 0.9086755245596104,
      "flos": 18113163223680.0,
      "grad_norm": 2.500716050487871,
      "language_loss": 0.80758262,
      "learning_rate": 8.67564491398467e-08,
      "loss": 0.82890177,
      "num_input_tokens_seen": 163382070,
      "step": 7557,
      "time_per_iteration": 2.846824884414673
    },
    {
      "auxiliary_loss_clip": 0.01150151,
      "auxiliary_loss_mlp": 0.01027156,
      "balance_loss_clip": 1.04484689,
      "balance_loss_mlp": 1.01970863,
      "epoch": 0.9087957674502495,
      "flos": 19129793857920.0,
      "grad_norm": 2.2332524997138226,
      "language_loss": 0.74041837,
      "learning_rate": 8.652965430207104e-08,
      "loss": 0.76219141,
      "num_input_tokens_seen": 163399975,
      "step": 7558,
      "time_per_iteration": 3.7512149810791016
    },
    {
      "auxiliary_loss_clip": 0.01109536,
      "auxiliary_loss_mlp": 0.01020328,
      "balance_loss_clip": 1.03983903,
      "balance_loss_mlp": 1.01294899,
      "epoch": 0.9089160103408886,
      "flos": 18109930999680.0,
      "grad_norm": 2.100574932982088,
      "language_loss": 0.65833509,
      "learning_rate": 8.630314973622521e-08,
      "loss": 0.67963374,
      "num_input_tokens_seen": 163417520,
      "step": 7559,
      "time_per_iteration": 2.779015302658081
    },
    {
      "auxiliary_loss_clip": 0.01135561,
      "auxiliary_loss_mlp": 0.01029601,
      "balance_loss_clip": 1.04479039,
      "balance_loss_mlp": 1.02235341,
      "epoch": 0.9090362532315277,
      "flos": 33364855336320.0,
      "grad_norm": 1.9251056269194442,
      "language_loss": 0.71178734,
      "learning_rate": 8.607693547666995e-08,
      "loss": 0.73343903,
      "num_input_tokens_seen": 163440060,
      "step": 7560,
      "time_per_iteration": 2.8063433170318604
    },
    {
      "auxiliary_loss_clip": 0.01047696,
      "auxiliary_loss_mlp": 0.01001316,
      "balance_loss_clip": 1.01059556,
      "balance_loss_mlp": 1.00031495,
      "epoch": 0.9091564961221668,
      "flos": 71480585082240.0,
      "grad_norm": 0.882087028317968,
      "language_loss": 0.57969731,
      "learning_rate": 8.585101155772201e-08,
      "loss": 0.60018742,
      "num_input_tokens_seen": 163502180,
      "step": 7561,
      "time_per_iteration": 3.371852159500122
    },
    {
      "auxiliary_loss_clip": 0.01157383,
      "auxiliary_loss_mlp": 0.01021356,
      "balance_loss_clip": 1.04375648,
      "balance_loss_mlp": 1.01375055,
      "epoch": 0.9092767390128058,
      "flos": 24712574232960.0,
      "grad_norm": 2.035306471428879,
      "language_loss": 0.68876255,
      "learning_rate": 8.562537801365377e-08,
      "loss": 0.71054995,
      "num_input_tokens_seen": 163521915,
      "step": 7562,
      "time_per_iteration": 2.7403106689453125
    },
    {
      "auxiliary_loss_clip": 0.01130866,
      "auxiliary_loss_mlp": 0.01022892,
      "balance_loss_clip": 1.04324973,
      "balance_loss_mlp": 1.01503575,
      "epoch": 0.909396981903445,
      "flos": 23586487879680.0,
      "grad_norm": 3.27592178182606,
      "language_loss": 0.70140409,
      "learning_rate": 8.540003487869362e-08,
      "loss": 0.72294164,
      "num_input_tokens_seen": 163543585,
      "step": 7563,
      "time_per_iteration": 2.73433518409729
    },
    {
      "auxiliary_loss_clip": 0.01094493,
      "auxiliary_loss_mlp": 0.01018664,
      "balance_loss_clip": 1.04027128,
      "balance_loss_mlp": 1.01116538,
      "epoch": 0.909517224794084,
      "flos": 23404169422080.0,
      "grad_norm": 1.941342330291705,
      "language_loss": 0.79759574,
      "learning_rate": 8.517498218702557e-08,
      "loss": 0.81872731,
      "num_input_tokens_seen": 163561515,
      "step": 7564,
      "time_per_iteration": 2.783871650695801
    },
    {
      "auxiliary_loss_clip": 0.01151008,
      "auxiliary_loss_mlp": 0.01027071,
      "balance_loss_clip": 1.0422895,
      "balance_loss_mlp": 1.019835,
      "epoch": 0.9096374676847231,
      "flos": 19208618254080.0,
      "grad_norm": 1.83807378764725,
      "language_loss": 0.69543982,
      "learning_rate": 8.49502199727905e-08,
      "loss": 0.7172206,
      "num_input_tokens_seen": 163579540,
      "step": 7565,
      "time_per_iteration": 2.687417984008789
    },
    {
      "auxiliary_loss_clip": 0.01134431,
      "auxiliary_loss_mlp": 0.01024296,
      "balance_loss_clip": 1.04086185,
      "balance_loss_mlp": 1.01654196,
      "epoch": 0.9097577105753623,
      "flos": 33292495388160.0,
      "grad_norm": 2.1025375974713367,
      "language_loss": 0.66292167,
      "learning_rate": 8.472574827008428e-08,
      "loss": 0.68450892,
      "num_input_tokens_seen": 163600425,
      "step": 7566,
      "time_per_iteration": 2.7777938842773438
    },
    {
      "auxiliary_loss_clip": 0.01142996,
      "auxiliary_loss_mlp": 0.00762134,
      "balance_loss_clip": 1.04039478,
      "balance_loss_mlp": 1.00056076,
      "epoch": 0.9098779534660013,
      "flos": 21906443001600.0,
      "grad_norm": 1.9941945592060561,
      "language_loss": 0.83875525,
      "learning_rate": 8.450156711295942e-08,
      "loss": 0.85780656,
      "num_input_tokens_seen": 163620595,
      "step": 7567,
      "time_per_iteration": 3.7756717205047607
    },
    {
      "auxiliary_loss_clip": 0.0111863,
      "auxiliary_loss_mlp": 0.01025254,
      "balance_loss_clip": 1.04059482,
      "balance_loss_mlp": 1.0181433,
      "epoch": 0.9099981963566404,
      "flos": 25730354102400.0,
      "grad_norm": 3.6924473162325415,
      "language_loss": 0.86337495,
      "learning_rate": 8.427767653542383e-08,
      "loss": 0.88481385,
      "num_input_tokens_seen": 163635765,
      "step": 7568,
      "time_per_iteration": 2.8017983436584473
    },
    {
      "auxiliary_loss_clip": 0.01151559,
      "auxiliary_loss_mlp": 0.01026273,
      "balance_loss_clip": 1.04646397,
      "balance_loss_mlp": 1.01863766,
      "epoch": 0.9101184392472795,
      "flos": 21069437304960.0,
      "grad_norm": 3.9305101115756016,
      "language_loss": 0.70611846,
      "learning_rate": 8.405407657144125e-08,
      "loss": 0.72789669,
      "num_input_tokens_seen": 163654925,
      "step": 7569,
      "time_per_iteration": 2.6703720092773438
    },
    {
      "auxiliary_loss_clip": 0.01139919,
      "auxiliary_loss_mlp": 0.01025628,
      "balance_loss_clip": 1.04530931,
      "balance_loss_mlp": 1.01837993,
      "epoch": 0.9102386821379186,
      "flos": 24752614919040.0,
      "grad_norm": 1.8801033734559829,
      "language_loss": 0.72619241,
      "learning_rate": 8.383076725493232e-08,
      "loss": 0.74784791,
      "num_input_tokens_seen": 163672245,
      "step": 7570,
      "time_per_iteration": 2.6886186599731445
    },
    {
      "auxiliary_loss_clip": 0.01120607,
      "auxiliary_loss_mlp": 0.01027111,
      "balance_loss_clip": 1.0399127,
      "balance_loss_mlp": 1.01919842,
      "epoch": 0.9103589250285576,
      "flos": 22562818179840.0,
      "grad_norm": 1.9373395844456476,
      "language_loss": 0.68131506,
      "learning_rate": 8.360774861977216e-08,
      "loss": 0.70279229,
      "num_input_tokens_seen": 163691365,
      "step": 7571,
      "time_per_iteration": 2.694821357727051
    },
    {
      "auxiliary_loss_clip": 0.01105438,
      "auxiliary_loss_mlp": 0.01022449,
      "balance_loss_clip": 1.04188526,
      "balance_loss_mlp": 1.01540971,
      "epoch": 0.9104791679191968,
      "flos": 25373474524800.0,
      "grad_norm": 2.1266693677773483,
      "language_loss": 0.74617743,
      "learning_rate": 8.338502069979281e-08,
      "loss": 0.76745629,
      "num_input_tokens_seen": 163711675,
      "step": 7572,
      "time_per_iteration": 2.770413398742676
    },
    {
      "auxiliary_loss_clip": 0.01115238,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.04388535,
      "balance_loss_mlp": 1.01852512,
      "epoch": 0.9105994108098359,
      "flos": 14426681558400.0,
      "grad_norm": 3.26160914934773,
      "language_loss": 0.80005777,
      "learning_rate": 8.316258352878214e-08,
      "loss": 0.82147712,
      "num_input_tokens_seen": 163728095,
      "step": 7573,
      "time_per_iteration": 2.7430191040039062
    },
    {
      "auxiliary_loss_clip": 0.01152402,
      "auxiliary_loss_mlp": 0.01022193,
      "balance_loss_clip": 1.04495406,
      "balance_loss_mlp": 1.01423895,
      "epoch": 0.9107196537004749,
      "flos": 26718292748160.0,
      "grad_norm": 2.008444798425987,
      "language_loss": 0.71170497,
      "learning_rate": 8.294043714048338e-08,
      "loss": 0.73345101,
      "num_input_tokens_seen": 163747175,
      "step": 7574,
      "time_per_iteration": 2.684579610824585
    },
    {
      "auxiliary_loss_clip": 0.01043758,
      "auxiliary_loss_mlp": 0.01001485,
      "balance_loss_clip": 1.02914321,
      "balance_loss_mlp": 1.00051332,
      "epoch": 0.9108398965911141,
      "flos": 66532634703360.0,
      "grad_norm": 0.7565564570862447,
      "language_loss": 0.60424209,
      "learning_rate": 8.271858156859624e-08,
      "loss": 0.62469447,
      "num_input_tokens_seen": 163812545,
      "step": 7575,
      "time_per_iteration": 3.3379878997802734
    },
    {
      "auxiliary_loss_clip": 0.01103436,
      "auxiliary_loss_mlp": 0.01021605,
      "balance_loss_clip": 1.03861165,
      "balance_loss_mlp": 1.0139215,
      "epoch": 0.9109601394817531,
      "flos": 25411073086080.0,
      "grad_norm": 1.707090933999928,
      "language_loss": 0.73922729,
      "learning_rate": 8.249701684677557e-08,
      "loss": 0.76047766,
      "num_input_tokens_seen": 163833870,
      "step": 7576,
      "time_per_iteration": 2.8070971965789795
    },
    {
      "auxiliary_loss_clip": 0.01130958,
      "auxiliary_loss_mlp": 0.01022662,
      "balance_loss_clip": 1.04406118,
      "balance_loss_mlp": 1.01502657,
      "epoch": 0.9110803823723922,
      "flos": 22747794243840.0,
      "grad_norm": 2.211762930824561,
      "language_loss": 0.81659013,
      "learning_rate": 8.227574300863294e-08,
      "loss": 0.8381263,
      "num_input_tokens_seen": 163854040,
      "step": 7577,
      "time_per_iteration": 2.722337007522583
    },
    {
      "auxiliary_loss_clip": 0.01121204,
      "auxiliary_loss_mlp": 0.0102864,
      "balance_loss_clip": 1.04257417,
      "balance_loss_mlp": 1.02006876,
      "epoch": 0.9112006252630314,
      "flos": 48469924131840.0,
      "grad_norm": 3.8315171701414776,
      "language_loss": 0.69446254,
      "learning_rate": 8.205476008773548e-08,
      "loss": 0.71596098,
      "num_input_tokens_seen": 163878040,
      "step": 7578,
      "time_per_iteration": 2.9681332111358643
    },
    {
      "auxiliary_loss_clip": 0.01127883,
      "auxiliary_loss_mlp": 0.01023641,
      "balance_loss_clip": 1.04194927,
      "balance_loss_mlp": 1.01589251,
      "epoch": 0.9113208681536704,
      "flos": 30009649829760.0,
      "grad_norm": 2.2211702820562422,
      "language_loss": 0.82825488,
      "learning_rate": 8.183406811760596e-08,
      "loss": 0.84977007,
      "num_input_tokens_seen": 163897770,
      "step": 7579,
      "time_per_iteration": 2.80534291267395
    },
    {
      "auxiliary_loss_clip": 0.01100173,
      "auxiliary_loss_mlp": 0.00762273,
      "balance_loss_clip": 1.03543782,
      "balance_loss_mlp": 1.00045776,
      "epoch": 0.9114411110443095,
      "flos": 25594971742080.0,
      "grad_norm": 3.637109428143423,
      "language_loss": 0.7416147,
      "learning_rate": 8.161366713172313e-08,
      "loss": 0.76023912,
      "num_input_tokens_seen": 163920160,
      "step": 7580,
      "time_per_iteration": 2.7670507431030273
    },
    {
      "auxiliary_loss_clip": 0.01139507,
      "auxiliary_loss_mlp": 0.01026006,
      "balance_loss_clip": 1.04291642,
      "balance_loss_mlp": 1.01805496,
      "epoch": 0.9115613539349486,
      "flos": 18399729928320.0,
      "grad_norm": 2.7235316207345375,
      "language_loss": 0.84535223,
      "learning_rate": 8.139355716352137e-08,
      "loss": 0.86700737,
      "num_input_tokens_seen": 163935000,
      "step": 7581,
      "time_per_iteration": 3.577345371246338
    },
    {
      "auxiliary_loss_clip": 0.01136581,
      "auxiliary_loss_mlp": 0.01022948,
      "balance_loss_clip": 1.04346621,
      "balance_loss_mlp": 1.01489544,
      "epoch": 0.9116815968255877,
      "flos": 21726171619200.0,
      "grad_norm": 1.644706534797065,
      "language_loss": 0.70067316,
      "learning_rate": 8.117373824639196e-08,
      "loss": 0.72226846,
      "num_input_tokens_seen": 163955265,
      "step": 7582,
      "time_per_iteration": 2.7021048069000244
    },
    {
      "auxiliary_loss_clip": 0.01049197,
      "auxiliary_loss_mlp": 0.01000793,
      "balance_loss_clip": 1.01085544,
      "balance_loss_mlp": 0.99975628,
      "epoch": 0.9118018397162267,
      "flos": 65363526835200.0,
      "grad_norm": 0.7241159255340995,
      "language_loss": 0.59237504,
      "learning_rate": 8.095421041368067e-08,
      "loss": 0.61287493,
      "num_input_tokens_seen": 164014680,
      "step": 7583,
      "time_per_iteration": 3.20635986328125
    },
    {
      "auxiliary_loss_clip": 0.01105694,
      "auxiliary_loss_mlp": 0.0102375,
      "balance_loss_clip": 1.03995597,
      "balance_loss_mlp": 1.01621604,
      "epoch": 0.9119220826068659,
      "flos": 20922885815040.0,
      "grad_norm": 3.269478489653178,
      "language_loss": 0.70574892,
      "learning_rate": 8.073497369868999e-08,
      "loss": 0.72704339,
      "num_input_tokens_seen": 164033140,
      "step": 7584,
      "time_per_iteration": 3.733626365661621
    },
    {
      "auxiliary_loss_clip": 0.01111384,
      "auxiliary_loss_mlp": 0.01022126,
      "balance_loss_clip": 1.04075861,
      "balance_loss_mlp": 1.01432371,
      "epoch": 0.912042325497505,
      "flos": 28366449327360.0,
      "grad_norm": 1.605064706248508,
      "language_loss": 0.75850594,
      "learning_rate": 8.051602813467772e-08,
      "loss": 0.77984101,
      "num_input_tokens_seen": 164054995,
      "step": 7585,
      "time_per_iteration": 2.800105571746826
    },
    {
      "auxiliary_loss_clip": 0.01137413,
      "auxiliary_loss_mlp": 0.01022139,
      "balance_loss_clip": 1.042274,
      "balance_loss_mlp": 1.01490295,
      "epoch": 0.912162568388144,
      "flos": 17566782468480.0,
      "grad_norm": 1.617817339553619,
      "language_loss": 0.71398187,
      "learning_rate": 8.029737375485756e-08,
      "loss": 0.7355774,
      "num_input_tokens_seen": 164074225,
      "step": 7586,
      "time_per_iteration": 2.6762728691101074
    },
    {
      "auxiliary_loss_clip": 0.01135355,
      "auxiliary_loss_mlp": 0.01024781,
      "balance_loss_clip": 1.04169631,
      "balance_loss_mlp": 1.01674938,
      "epoch": 0.9122828112787832,
      "flos": 19827897661440.0,
      "grad_norm": 2.0047585077586714,
      "language_loss": 0.72766024,
      "learning_rate": 8.007901059239986e-08,
      "loss": 0.74926162,
      "num_input_tokens_seen": 164093505,
      "step": 7587,
      "time_per_iteration": 2.664206027984619
    },
    {
      "auxiliary_loss_clip": 0.01095739,
      "auxiliary_loss_mlp": 0.01021225,
      "balance_loss_clip": 1.04001343,
      "balance_loss_mlp": 1.01389337,
      "epoch": 0.9124030541694222,
      "flos": 20813789232000.0,
      "grad_norm": 1.98706318429892,
      "language_loss": 0.80309021,
      "learning_rate": 7.986093868042964e-08,
      "loss": 0.82425982,
      "num_input_tokens_seen": 164113750,
      "step": 7588,
      "time_per_iteration": 2.8219919204711914
    },
    {
      "auxiliary_loss_clip": 0.01072234,
      "auxiliary_loss_mlp": 0.01026443,
      "balance_loss_clip": 1.03692019,
      "balance_loss_mlp": 1.01907599,
      "epoch": 0.9125232970600613,
      "flos": 25192305302400.0,
      "grad_norm": 2.0563500418002416,
      "language_loss": 0.67459249,
      "learning_rate": 7.964315805202826e-08,
      "loss": 0.69557935,
      "num_input_tokens_seen": 164134330,
      "step": 7589,
      "time_per_iteration": 2.8464841842651367
    },
    {
      "auxiliary_loss_clip": 0.01113436,
      "auxiliary_loss_mlp": 0.01024831,
      "balance_loss_clip": 1.04496264,
      "balance_loss_mlp": 1.01703501,
      "epoch": 0.9126435399507005,
      "flos": 19719591177600.0,
      "grad_norm": 2.144667544287393,
      "language_loss": 0.73347086,
      "learning_rate": 7.942566874023304e-08,
      "loss": 0.75485355,
      "num_input_tokens_seen": 164153515,
      "step": 7590,
      "time_per_iteration": 2.851496934890747
    },
    {
      "auxiliary_loss_clip": 0.01141522,
      "auxiliary_loss_mlp": 0.010232,
      "balance_loss_clip": 1.04320455,
      "balance_loss_mlp": 1.01488853,
      "epoch": 0.9127637828413395,
      "flos": 19573614305280.0,
      "grad_norm": 2.0608513045183074,
      "language_loss": 0.69344747,
      "learning_rate": 7.920847077803649e-08,
      "loss": 0.71509469,
      "num_input_tokens_seen": 164171305,
      "step": 7591,
      "time_per_iteration": 2.6999406814575195
    },
    {
      "auxiliary_loss_clip": 0.01120365,
      "auxiliary_loss_mlp": 0.01022477,
      "balance_loss_clip": 1.03999662,
      "balance_loss_mlp": 1.01509202,
      "epoch": 0.9128840257319786,
      "flos": 20230635928320.0,
      "grad_norm": 1.9141816408979968,
      "language_loss": 0.82352197,
      "learning_rate": 7.899156419838826e-08,
      "loss": 0.84495044,
      "num_input_tokens_seen": 164190275,
      "step": 7592,
      "time_per_iteration": 3.6529436111450195
    },
    {
      "auxiliary_loss_clip": 0.01106361,
      "auxiliary_loss_mlp": 0.01021899,
      "balance_loss_clip": 1.0412724,
      "balance_loss_mlp": 1.01391768,
      "epoch": 0.9130042686226177,
      "flos": 24858658846080.0,
      "grad_norm": 1.8341161785009923,
      "language_loss": 0.65905702,
      "learning_rate": 7.87749490341918e-08,
      "loss": 0.68033957,
      "num_input_tokens_seen": 164210550,
      "step": 7593,
      "time_per_iteration": 2.852989435195923
    },
    {
      "auxiliary_loss_clip": 0.01126594,
      "auxiliary_loss_mlp": 0.01026249,
      "balance_loss_clip": 1.04257417,
      "balance_loss_mlp": 1.01713598,
      "epoch": 0.9131245115132568,
      "flos": 23581747284480.0,
      "grad_norm": 2.0067432000616217,
      "language_loss": 0.83371031,
      "learning_rate": 7.855862531830836e-08,
      "loss": 0.85523868,
      "num_input_tokens_seen": 164226660,
      "step": 7594,
      "time_per_iteration": 2.6944491863250732
    },
    {
      "auxiliary_loss_clip": 0.01113836,
      "auxiliary_loss_mlp": 0.01021698,
      "balance_loss_clip": 1.04005599,
      "balance_loss_mlp": 1.01425934,
      "epoch": 0.9132447544038959,
      "flos": 19931607204480.0,
      "grad_norm": 2.1244118830672583,
      "language_loss": 0.72635996,
      "learning_rate": 7.834259308355373e-08,
      "loss": 0.74771535,
      "num_input_tokens_seen": 164245425,
      "step": 7595,
      "time_per_iteration": 2.795289993286133
    },
    {
      "auxiliary_loss_clip": 0.01106863,
      "auxiliary_loss_mlp": 0.01027072,
      "balance_loss_clip": 1.04004061,
      "balance_loss_mlp": 1.019526,
      "epoch": 0.9133649972945349,
      "flos": 21981747864960.0,
      "grad_norm": 2.180148136566148,
      "language_loss": 0.74838591,
      "learning_rate": 7.812685236269989e-08,
      "loss": 0.76972526,
      "num_input_tokens_seen": 164264085,
      "step": 7596,
      "time_per_iteration": 2.7793917655944824
    },
    {
      "auxiliary_loss_clip": 0.01009003,
      "auxiliary_loss_mlp": 0.01000497,
      "balance_loss_clip": 1.00756168,
      "balance_loss_mlp": 0.99941236,
      "epoch": 0.9134852401851741,
      "flos": 71240523511680.0,
      "grad_norm": 0.7876369097770157,
      "language_loss": 0.58614266,
      "learning_rate": 7.791140318847445e-08,
      "loss": 0.60623765,
      "num_input_tokens_seen": 164322220,
      "step": 7597,
      "time_per_iteration": 3.39770770072937
    },
    {
      "auxiliary_loss_clip": 0.01154017,
      "auxiliary_loss_mlp": 0.0102525,
      "balance_loss_clip": 1.04709721,
      "balance_loss_mlp": 1.01828825,
      "epoch": 0.9136054830758131,
      "flos": 23626923615360.0,
      "grad_norm": 1.7794250524500248,
      "language_loss": 0.80319184,
      "learning_rate": 7.769624559356081e-08,
      "loss": 0.82498455,
      "num_input_tokens_seen": 164345615,
      "step": 7598,
      "time_per_iteration": 3.2068464756011963
    },
    {
      "auxiliary_loss_clip": 0.01113131,
      "auxiliary_loss_mlp": 0.01028372,
      "balance_loss_clip": 1.04457676,
      "balance_loss_mlp": 1.02020621,
      "epoch": 0.9137257259664522,
      "flos": 23438858981760.0,
      "grad_norm": 6.14835143966804,
      "language_loss": 0.75458086,
      "learning_rate": 7.748137961059842e-08,
      "loss": 0.77599585,
      "num_input_tokens_seen": 164359595,
      "step": 7599,
      "time_per_iteration": 2.8097481727600098
    },
    {
      "auxiliary_loss_clip": 0.01153137,
      "auxiliary_loss_mlp": 0.01024599,
      "balance_loss_clip": 1.04446399,
      "balance_loss_mlp": 1.01783943,
      "epoch": 0.9138459688570914,
      "flos": 19127854523520.0,
      "grad_norm": 3.051451250661027,
      "language_loss": 0.65538019,
      "learning_rate": 7.726680527218211e-08,
      "loss": 0.67715752,
      "num_input_tokens_seen": 164376635,
      "step": 7600,
      "time_per_iteration": 2.6685445308685303
    },
    {
      "auxiliary_loss_clip": 0.01123814,
      "auxiliary_loss_mlp": 0.01022471,
      "balance_loss_clip": 1.03953111,
      "balance_loss_mlp": 1.01495516,
      "epoch": 0.9139662117477304,
      "flos": 46281240714240.0,
      "grad_norm": 1.7926552173695924,
      "language_loss": 0.7570551,
      "learning_rate": 7.70525226108627e-08,
      "loss": 0.77851796,
      "num_input_tokens_seen": 164400305,
      "step": 7601,
      "time_per_iteration": 2.9015426635742188
    },
    {
      "auxiliary_loss_clip": 0.01093299,
      "auxiliary_loss_mlp": 0.0102835,
      "balance_loss_clip": 1.03800535,
      "balance_loss_mlp": 1.02106071,
      "epoch": 0.9140864546383695,
      "flos": 22273198819200.0,
      "grad_norm": 2.0863384200775648,
      "language_loss": 0.80033207,
      "learning_rate": 7.683853165914666e-08,
      "loss": 0.82154858,
      "num_input_tokens_seen": 164418075,
      "step": 7602,
      "time_per_iteration": 2.7810184955596924
    },
    {
      "auxiliary_loss_clip": 0.01144998,
      "auxiliary_loss_mlp": 0.01026678,
      "balance_loss_clip": 1.0437808,
      "balance_loss_mlp": 1.01932263,
      "epoch": 0.9142066975290086,
      "flos": 17530009920000.0,
      "grad_norm": 1.9713691462481928,
      "language_loss": 0.76997471,
      "learning_rate": 7.662483244949602e-08,
      "loss": 0.79169154,
      "num_input_tokens_seen": 164435335,
      "step": 7603,
      "time_per_iteration": 2.6374361515045166
    },
    {
      "auxiliary_loss_clip": 0.01070014,
      "auxiliary_loss_mlp": 0.01025134,
      "balance_loss_clip": 1.03886735,
      "balance_loss_mlp": 1.01754606,
      "epoch": 0.9143269404196477,
      "flos": 17712148809600.0,
      "grad_norm": 4.0708654378102125,
      "language_loss": 0.80704868,
      "learning_rate": 7.641142501432951e-08,
      "loss": 0.82800019,
      "num_input_tokens_seen": 164451530,
      "step": 7604,
      "time_per_iteration": 2.858670949935913
    },
    {
      "auxiliary_loss_clip": 0.01146299,
      "auxiliary_loss_mlp": 0.01024564,
      "balance_loss_clip": 1.04489899,
      "balance_loss_mlp": 1.01807642,
      "epoch": 0.9144471833102867,
      "flos": 33323414019840.0,
      "grad_norm": 1.7368732296105527,
      "language_loss": 0.74078524,
      "learning_rate": 7.619830938602013e-08,
      "loss": 0.76249385,
      "num_input_tokens_seen": 164472755,
      "step": 7605,
      "time_per_iteration": 3.2439377307891846
    },
    {
      "auxiliary_loss_clip": 0.01157437,
      "auxiliary_loss_mlp": 0.01022226,
      "balance_loss_clip": 1.04889131,
      "balance_loss_mlp": 1.01459074,
      "epoch": 0.9145674262009259,
      "flos": 21068970428160.0,
      "grad_norm": 2.252077932269188,
      "language_loss": 0.82639313,
      "learning_rate": 7.598548559689777e-08,
      "loss": 0.84818977,
      "num_input_tokens_seen": 164491155,
      "step": 7606,
      "time_per_iteration": 3.550154209136963
    },
    {
      "auxiliary_loss_clip": 0.01168396,
      "auxiliary_loss_mlp": 0.01024206,
      "balance_loss_clip": 1.04583955,
      "balance_loss_mlp": 1.01642752,
      "epoch": 0.914687669091565,
      "flos": 16800269212800.0,
      "grad_norm": 2.6947845344393357,
      "language_loss": 0.81184483,
      "learning_rate": 7.577295367924751e-08,
      "loss": 0.83377087,
      "num_input_tokens_seen": 164507555,
      "step": 7607,
      "time_per_iteration": 3.580331325531006
    },
    {
      "auxiliary_loss_clip": 0.0108734,
      "auxiliary_loss_mlp": 0.00762536,
      "balance_loss_clip": 1.03968227,
      "balance_loss_mlp": 1.00040972,
      "epoch": 0.914807911982204,
      "flos": 25773627012480.0,
      "grad_norm": 6.099880660938896,
      "language_loss": 0.82515603,
      "learning_rate": 7.556071366531002e-08,
      "loss": 0.84365475,
      "num_input_tokens_seen": 164528525,
      "step": 7608,
      "time_per_iteration": 2.848438024520874
    },
    {
      "auxiliary_loss_clip": 0.01154284,
      "auxiliary_loss_mlp": 0.0102835,
      "balance_loss_clip": 1.04331577,
      "balance_loss_mlp": 1.02040446,
      "epoch": 0.9149281548728432,
      "flos": 19208043636480.0,
      "grad_norm": 3.0395912912790037,
      "language_loss": 0.79083645,
      "learning_rate": 7.53487655872822e-08,
      "loss": 0.81266278,
      "num_input_tokens_seen": 164547695,
      "step": 7609,
      "time_per_iteration": 2.69132924079895
    },
    {
      "auxiliary_loss_clip": 0.01129225,
      "auxiliary_loss_mlp": 0.00761871,
      "balance_loss_clip": 1.04043257,
      "balance_loss_mlp": 1.00043631,
      "epoch": 0.9150483977634822,
      "flos": 26870554500480.0,
      "grad_norm": 2.0645220093555428,
      "language_loss": 0.74086982,
      "learning_rate": 7.513710947731656e-08,
      "loss": 0.75978076,
      "num_input_tokens_seen": 164568905,
      "step": 7610,
      "time_per_iteration": 3.65913724899292
    },
    {
      "auxiliary_loss_clip": 0.01130565,
      "auxiliary_loss_mlp": 0.01030654,
      "balance_loss_clip": 1.04450011,
      "balance_loss_mlp": 1.02329326,
      "epoch": 0.9151686406541213,
      "flos": 21908956953600.0,
      "grad_norm": 1.8241925335794469,
      "language_loss": 0.85385036,
      "learning_rate": 7.492574536752095e-08,
      "loss": 0.87546253,
      "num_input_tokens_seen": 164588895,
      "step": 7611,
      "time_per_iteration": 2.6947665214538574
    },
    {
      "auxiliary_loss_clip": 0.01137423,
      "auxiliary_loss_mlp": 0.01028041,
      "balance_loss_clip": 1.04266906,
      "balance_loss_mlp": 1.02064419,
      "epoch": 0.9152888835447605,
      "flos": 27308556944640.0,
      "grad_norm": 1.7477203831062658,
      "language_loss": 0.78135753,
      "learning_rate": 7.471467328995907e-08,
      "loss": 0.80301225,
      "num_input_tokens_seen": 164607705,
      "step": 7612,
      "time_per_iteration": 2.6952850818634033
    },
    {
      "auxiliary_loss_clip": 0.01049551,
      "auxiliary_loss_mlp": 0.010233,
      "balance_loss_clip": 1.03140497,
      "balance_loss_mlp": 1.01581955,
      "epoch": 0.9154091264353995,
      "flos": 13370728510080.0,
      "grad_norm": 2.3224323258883754,
      "language_loss": 0.60850942,
      "learning_rate": 7.450389327665018e-08,
      "loss": 0.62923789,
      "num_input_tokens_seen": 164625540,
      "step": 7613,
      "time_per_iteration": 3.0870532989501953
    },
    {
      "auxiliary_loss_clip": 0.01111787,
      "auxiliary_loss_mlp": 0.01024614,
      "balance_loss_clip": 1.04360092,
      "balance_loss_mlp": 1.01668036,
      "epoch": 0.9155293693260386,
      "flos": 20193037367040.0,
      "grad_norm": 2.610306295396331,
      "language_loss": 0.67525458,
      "learning_rate": 7.429340535957029e-08,
      "loss": 0.69661868,
      "num_input_tokens_seen": 164640735,
      "step": 7614,
      "time_per_iteration": 3.0558207035064697
    },
    {
      "auxiliary_loss_clip": 0.01121552,
      "auxiliary_loss_mlp": 0.01022974,
      "balance_loss_clip": 1.04317153,
      "balance_loss_mlp": 1.01519895,
      "epoch": 0.9156496122166777,
      "flos": 19354990176000.0,
      "grad_norm": 2.196327098442628,
      "language_loss": 0.70308673,
      "learning_rate": 7.40832095706494e-08,
      "loss": 0.72453201,
      "num_input_tokens_seen": 164657430,
      "step": 7615,
      "time_per_iteration": 2.717777967453003
    },
    {
      "auxiliary_loss_clip": 0.01124519,
      "auxiliary_loss_mlp": 0.01030054,
      "balance_loss_clip": 1.04139972,
      "balance_loss_mlp": 1.0227735,
      "epoch": 0.9157698551073168,
      "flos": 21107287261440.0,
      "grad_norm": 2.931740325450068,
      "language_loss": 0.80234134,
      "learning_rate": 7.387330594177443e-08,
      "loss": 0.82388705,
      "num_input_tokens_seen": 164679505,
      "step": 7616,
      "time_per_iteration": 2.735410690307617
    },
    {
      "auxiliary_loss_clip": 0.01108704,
      "auxiliary_loss_mlp": 0.01025279,
      "balance_loss_clip": 1.04033661,
      "balance_loss_mlp": 1.01787019,
      "epoch": 0.9158900979979558,
      "flos": 25193167228800.0,
      "grad_norm": 4.590597581338073,
      "language_loss": 0.79133749,
      "learning_rate": 7.366369450478749e-08,
      "loss": 0.81267732,
      "num_input_tokens_seen": 164700615,
      "step": 7617,
      "time_per_iteration": 4.1775243282318115
    },
    {
      "auxiliary_loss_clip": 0.01105998,
      "auxiliary_loss_mlp": 0.01023292,
      "balance_loss_clip": 1.04028392,
      "balance_loss_mlp": 1.01582384,
      "epoch": 0.916010340888595,
      "flos": 30146648302080.0,
      "grad_norm": 2.9516965931622794,
      "language_loss": 0.66191065,
      "learning_rate": 7.345437529148646e-08,
      "loss": 0.68320352,
      "num_input_tokens_seen": 164719625,
      "step": 7618,
      "time_per_iteration": 2.948195457458496
    },
    {
      "auxiliary_loss_clip": 0.01109313,
      "auxiliary_loss_mlp": 0.01023297,
      "balance_loss_clip": 1.04013443,
      "balance_loss_mlp": 1.01608157,
      "epoch": 0.9161305837792341,
      "flos": 17091827907840.0,
      "grad_norm": 2.21063909715395,
      "language_loss": 0.73054957,
      "learning_rate": 7.324534833362483e-08,
      "loss": 0.75187564,
      "num_input_tokens_seen": 164737200,
      "step": 7619,
      "time_per_iteration": 2.757870674133301
    },
    {
      "auxiliary_loss_clip": 0.01137404,
      "auxiliary_loss_mlp": 0.01022367,
      "balance_loss_clip": 1.04617572,
      "balance_loss_mlp": 1.01505613,
      "epoch": 0.9162508266698731,
      "flos": 22893699288960.0,
      "grad_norm": 2.8759519397694224,
      "language_loss": 0.68530875,
      "learning_rate": 7.303661366291192e-08,
      "loss": 0.70690644,
      "num_input_tokens_seen": 164757870,
      "step": 7620,
      "time_per_iteration": 2.7762362957000732
    },
    {
      "auxiliary_loss_clip": 0.01077586,
      "auxiliary_loss_mlp": 0.01022787,
      "balance_loss_clip": 1.04201865,
      "balance_loss_mlp": 1.01517832,
      "epoch": 0.9163710695605123,
      "flos": 19974808287360.0,
      "grad_norm": 2.0321527204253194,
      "language_loss": 0.81621969,
      "learning_rate": 7.28281713110126e-08,
      "loss": 0.83722341,
      "num_input_tokens_seen": 164775945,
      "step": 7621,
      "time_per_iteration": 2.8789432048797607
    },
    {
      "auxiliary_loss_clip": 0.01129366,
      "auxiliary_loss_mlp": 0.01024682,
      "balance_loss_clip": 1.04534757,
      "balance_loss_mlp": 1.01757097,
      "epoch": 0.9164913124511513,
      "flos": 22783812606720.0,
      "grad_norm": 2.9301566749437127,
      "language_loss": 0.7718426,
      "learning_rate": 7.262002130954759e-08,
      "loss": 0.79338312,
      "num_input_tokens_seen": 164794400,
      "step": 7622,
      "time_per_iteration": 2.783890724182129
    },
    {
      "auxiliary_loss_clip": 0.01115531,
      "auxiliary_loss_mlp": 0.01030192,
      "balance_loss_clip": 1.04828131,
      "balance_loss_mlp": 1.02199054,
      "epoch": 0.9166115553417904,
      "flos": 24900854348160.0,
      "grad_norm": 1.8847434219778678,
      "language_loss": 0.78912711,
      "learning_rate": 7.241216369009296e-08,
      "loss": 0.81058431,
      "num_input_tokens_seen": 164814585,
      "step": 7623,
      "time_per_iteration": 2.8284497261047363
    },
    {
      "auxiliary_loss_clip": 0.01103933,
      "auxiliary_loss_mlp": 0.01021604,
      "balance_loss_clip": 1.04017162,
      "balance_loss_mlp": 1.01395726,
      "epoch": 0.9167317982324296,
      "flos": 25702919089920.0,
      "grad_norm": 2.817374263552367,
      "language_loss": 0.66562414,
      "learning_rate": 7.220459848418037e-08,
      "loss": 0.68687952,
      "num_input_tokens_seen": 164834660,
      "step": 7624,
      "time_per_iteration": 2.7618024349212646
    },
    {
      "auxiliary_loss_clip": 0.01111493,
      "auxiliary_loss_mlp": 0.01021297,
      "balance_loss_clip": 1.04061079,
      "balance_loss_mlp": 1.01366138,
      "epoch": 0.9168520411230686,
      "flos": 15632813370240.0,
      "grad_norm": 2.4365334151681055,
      "language_loss": 0.79845715,
      "learning_rate": 7.199732572329708e-08,
      "loss": 0.819785,
      "num_input_tokens_seen": 164852560,
      "step": 7625,
      "time_per_iteration": 2.766972780227661
    },
    {
      "auxiliary_loss_clip": 0.01061025,
      "auxiliary_loss_mlp": 0.01028937,
      "balance_loss_clip": 1.03849006,
      "balance_loss_mlp": 1.02117062,
      "epoch": 0.9169722840137077,
      "flos": 30258151096320.0,
      "grad_norm": 2.455877054627479,
      "language_loss": 0.7597208,
      "learning_rate": 7.179034543888684e-08,
      "loss": 0.7806204,
      "num_input_tokens_seen": 164872065,
      "step": 7626,
      "time_per_iteration": 3.344627857208252
    },
    {
      "auxiliary_loss_clip": 0.01103803,
      "auxiliary_loss_mlp": 0.01023772,
      "balance_loss_clip": 1.04007411,
      "balance_loss_mlp": 1.01627922,
      "epoch": 0.9170925269043467,
      "flos": 22491643380480.0,
      "grad_norm": 2.115107029100762,
      "language_loss": 0.77830583,
      "learning_rate": 7.158365766234808e-08,
      "loss": 0.79958153,
      "num_input_tokens_seen": 164890915,
      "step": 7627,
      "time_per_iteration": 3.3347597122192383
    },
    {
      "auxiliary_loss_clip": 0.01129226,
      "auxiliary_loss_mlp": 0.01030037,
      "balance_loss_clip": 1.04480219,
      "balance_loss_mlp": 1.02169251,
      "epoch": 0.9172127697949859,
      "flos": 22893914770560.0,
      "grad_norm": 1.9757566806316025,
      "language_loss": 0.72425818,
      "learning_rate": 7.137726242503527e-08,
      "loss": 0.7458508,
      "num_input_tokens_seen": 164909835,
      "step": 7628,
      "time_per_iteration": 2.7709860801696777
    },
    {
      "auxiliary_loss_clip": 0.01123539,
      "auxiliary_loss_mlp": 0.01023033,
      "balance_loss_clip": 1.04444289,
      "balance_loss_mlp": 1.01530766,
      "epoch": 0.917333012685625,
      "flos": 17451867882240.0,
      "grad_norm": 2.630135040779435,
      "language_loss": 0.78117681,
      "learning_rate": 7.11711597582585e-08,
      "loss": 0.80264246,
      "num_input_tokens_seen": 164927195,
      "step": 7629,
      "time_per_iteration": 2.7249882221221924
    },
    {
      "auxiliary_loss_clip": 0.01092124,
      "auxiliary_loss_mlp": 0.01023182,
      "balance_loss_clip": 1.03776598,
      "balance_loss_mlp": 1.01618099,
      "epoch": 0.917453255576264,
      "flos": 14318949692160.0,
      "grad_norm": 2.3877403113366773,
      "language_loss": 0.80436081,
      "learning_rate": 7.096534969328271e-08,
      "loss": 0.82551384,
      "num_input_tokens_seen": 164944640,
      "step": 7630,
      "time_per_iteration": 2.8275082111358643
    },
    {
      "auxiliary_loss_clip": 0.01153801,
      "auxiliary_loss_mlp": 0.01022184,
      "balance_loss_clip": 1.04342318,
      "balance_loss_mlp": 1.01523089,
      "epoch": 0.9175734984669032,
      "flos": 20741177888640.0,
      "grad_norm": 2.807165618721204,
      "language_loss": 0.83977002,
      "learning_rate": 7.075983226132987e-08,
      "loss": 0.86152983,
      "num_input_tokens_seen": 164963570,
      "step": 7631,
      "time_per_iteration": 2.6581265926361084
    },
    {
      "auxiliary_loss_clip": 0.0112501,
      "auxiliary_loss_mlp": 0.01027928,
      "balance_loss_clip": 1.04394221,
      "balance_loss_mlp": 1.01988745,
      "epoch": 0.9176937413575422,
      "flos": 14830497233280.0,
      "grad_norm": 2.917343745536187,
      "language_loss": 0.79365021,
      "learning_rate": 7.055460749357656e-08,
      "loss": 0.81517959,
      "num_input_tokens_seen": 164979850,
      "step": 7632,
      "time_per_iteration": 4.848465919494629
    },
    {
      "auxiliary_loss_clip": 0.01121547,
      "auxiliary_loss_mlp": 0.00761903,
      "balance_loss_clip": 1.04107952,
      "balance_loss_mlp": 1.00048232,
      "epoch": 0.9178139842481813,
      "flos": 18474603828480.0,
      "grad_norm": 2.3873101000493837,
      "language_loss": 0.70617139,
      "learning_rate": 7.034967542115521e-08,
      "loss": 0.72500592,
      "num_input_tokens_seen": 164998115,
      "step": 7633,
      "time_per_iteration": 2.722325086593628
    },
    {
      "auxiliary_loss_clip": 0.01165747,
      "auxiliary_loss_mlp": 0.00761996,
      "balance_loss_clip": 1.04659939,
      "balance_loss_mlp": 1.00043368,
      "epoch": 0.9179342271388204,
      "flos": 20047455544320.0,
      "grad_norm": 2.2600140168516174,
      "language_loss": 0.75396216,
      "learning_rate": 7.014503607515388e-08,
      "loss": 0.77323949,
      "num_input_tokens_seen": 165017420,
      "step": 7634,
      "time_per_iteration": 2.6357388496398926
    },
    {
      "auxiliary_loss_clip": 0.01130282,
      "auxiliary_loss_mlp": 0.01027787,
      "balance_loss_clip": 1.04547524,
      "balance_loss_mlp": 1.02067971,
      "epoch": 0.9180544700294595,
      "flos": 24676232647680.0,
      "grad_norm": 3.3728718213044115,
      "language_loss": 0.67955065,
      "learning_rate": 6.994068948661592e-08,
      "loss": 0.70113134,
      "num_input_tokens_seen": 165035575,
      "step": 7635,
      "time_per_iteration": 2.765364170074463
    },
    {
      "auxiliary_loss_clip": 0.01120451,
      "auxiliary_loss_mlp": 0.0102965,
      "balance_loss_clip": 1.03991961,
      "balance_loss_mlp": 1.02135313,
      "epoch": 0.9181747129200986,
      "flos": 16727478301440.0,
      "grad_norm": 2.1806492148737995,
      "language_loss": 0.76961803,
      "learning_rate": 6.973663568654142e-08,
      "loss": 0.79111898,
      "num_input_tokens_seen": 165053280,
      "step": 7636,
      "time_per_iteration": 3.6544742584228516
    },
    {
      "auxiliary_loss_clip": 0.01112836,
      "auxiliary_loss_mlp": 0.01033524,
      "balance_loss_clip": 1.04129648,
      "balance_loss_mlp": 1.02501869,
      "epoch": 0.9182949558107377,
      "flos": 24271626873600.0,
      "grad_norm": 2.157627612595263,
      "language_loss": 0.6541419,
      "learning_rate": 6.953287470588386e-08,
      "loss": 0.67560554,
      "num_input_tokens_seen": 165071235,
      "step": 7637,
      "time_per_iteration": 2.815131902694702
    },
    {
      "auxiliary_loss_clip": 0.01152624,
      "auxiliary_loss_mlp": 0.01024637,
      "balance_loss_clip": 1.0462749,
      "balance_loss_mlp": 1.01673067,
      "epoch": 0.9184151987013768,
      "flos": 22082117443200.0,
      "grad_norm": 2.455116610801295,
      "language_loss": 0.85609275,
      "learning_rate": 6.932940657555452e-08,
      "loss": 0.87786531,
      "num_input_tokens_seen": 165087365,
      "step": 7638,
      "time_per_iteration": 2.6244261264801025
    },
    {
      "auxiliary_loss_clip": 0.0113064,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.04181194,
      "balance_loss_mlp": 1.01539063,
      "epoch": 0.9185354415920158,
      "flos": 32166732257280.0,
      "grad_norm": 1.505964113215491,
      "language_loss": 0.76529098,
      "learning_rate": 6.912623132641938e-08,
      "loss": 0.7868275,
      "num_input_tokens_seen": 165112455,
      "step": 7639,
      "time_per_iteration": 2.8021240234375
    },
    {
      "auxiliary_loss_clip": 0.01128787,
      "auxiliary_loss_mlp": 0.01030521,
      "balance_loss_clip": 1.04425097,
      "balance_loss_mlp": 1.02257013,
      "epoch": 0.918655684482655,
      "flos": 20997831542400.0,
      "grad_norm": 2.085844462847014,
      "language_loss": 0.76771259,
      "learning_rate": 6.892334898929952e-08,
      "loss": 0.78930569,
      "num_input_tokens_seen": 165132700,
      "step": 7640,
      "time_per_iteration": 2.716186761856079
    },
    {
      "auxiliary_loss_clip": 0.01119868,
      "auxiliary_loss_mlp": 0.01025141,
      "balance_loss_clip": 1.04166663,
      "balance_loss_mlp": 1.01760066,
      "epoch": 0.918775927373294,
      "flos": 15560704817280.0,
      "grad_norm": 2.845940693655713,
      "language_loss": 0.84741819,
      "learning_rate": 6.872075959497236e-08,
      "loss": 0.86886823,
      "num_input_tokens_seen": 165151475,
      "step": 7641,
      "time_per_iteration": 2.6212551593780518
    },
    {
      "auxiliary_loss_clip": 0.01124809,
      "auxiliary_loss_mlp": 0.01024432,
      "balance_loss_clip": 1.04126072,
      "balance_loss_mlp": 1.01708841,
      "epoch": 0.9188961702639331,
      "flos": 29934057657600.0,
      "grad_norm": 1.872983343274557,
      "language_loss": 0.83094937,
      "learning_rate": 6.85184631741702e-08,
      "loss": 0.85244179,
      "num_input_tokens_seen": 165172040,
      "step": 7642,
      "time_per_iteration": 2.652710437774658
    },
    {
      "auxiliary_loss_clip": 0.01123654,
      "auxiliary_loss_mlp": 0.01022638,
      "balance_loss_clip": 1.03847909,
      "balance_loss_mlp": 1.01543152,
      "epoch": 0.9190164131545723,
      "flos": 20701244943360.0,
      "grad_norm": 3.3342170279742582,
      "language_loss": 0.7749784,
      "learning_rate": 6.831645975758161e-08,
      "loss": 0.79644132,
      "num_input_tokens_seen": 165189980,
      "step": 7643,
      "time_per_iteration": 3.397568702697754
    },
    {
      "auxiliary_loss_clip": 0.01142333,
      "auxiliary_loss_mlp": 0.01023601,
      "balance_loss_clip": 1.04470778,
      "balance_loss_mlp": 1.01580441,
      "epoch": 0.9191366560452113,
      "flos": 25629912696960.0,
      "grad_norm": 2.211329167117406,
      "language_loss": 0.67399794,
      "learning_rate": 6.811474937585026e-08,
      "loss": 0.69565725,
      "num_input_tokens_seen": 165209770,
      "step": 7644,
      "time_per_iteration": 2.6793406009674072
    },
    {
      "auxiliary_loss_clip": 0.01153422,
      "auxiliary_loss_mlp": 0.01022162,
      "balance_loss_clip": 1.04739916,
      "balance_loss_mlp": 1.01504552,
      "epoch": 0.9192568989358504,
      "flos": 21434325615360.0,
      "grad_norm": 2.454018925016314,
      "language_loss": 0.79352105,
      "learning_rate": 6.79133320595755e-08,
      "loss": 0.81527692,
      "num_input_tokens_seen": 165229690,
      "step": 7645,
      "time_per_iteration": 2.625246047973633
    },
    {
      "auxiliary_loss_clip": 0.01123555,
      "auxiliary_loss_mlp": 0.01025085,
      "balance_loss_clip": 1.0419271,
      "balance_loss_mlp": 1.0174315,
      "epoch": 0.9193771418264896,
      "flos": 23185078416000.0,
      "grad_norm": 1.884475949776951,
      "language_loss": 0.75458872,
      "learning_rate": 6.771220783931198e-08,
      "loss": 0.77607512,
      "num_input_tokens_seen": 165249850,
      "step": 7646,
      "time_per_iteration": 2.6775004863739014
    },
    {
      "auxiliary_loss_clip": 0.01014396,
      "auxiliary_loss_mlp": 0.01001179,
      "balance_loss_clip": 1.00866866,
      "balance_loss_mlp": 1.00010586,
      "epoch": 0.9194973847171286,
      "flos": 70582963184640.0,
      "grad_norm": 0.8467690787807002,
      "language_loss": 0.64513361,
      "learning_rate": 6.751137674556994e-08,
      "loss": 0.66528934,
      "num_input_tokens_seen": 165310235,
      "step": 7647,
      "time_per_iteration": 3.515638589859009
    },
    {
      "auxiliary_loss_clip": 0.01148304,
      "auxiliary_loss_mlp": 0.01019686,
      "balance_loss_clip": 1.04407549,
      "balance_loss_mlp": 1.01172853,
      "epoch": 0.9196176276077677,
      "flos": 14720682378240.0,
      "grad_norm": 2.519580582132067,
      "language_loss": 0.77411562,
      "learning_rate": 6.731083880881572e-08,
      "loss": 0.79579556,
      "num_input_tokens_seen": 165326455,
      "step": 7648,
      "time_per_iteration": 3.030540704727173
    },
    {
      "auxiliary_loss_clip": 0.01061491,
      "auxiliary_loss_mlp": 0.00761874,
      "balance_loss_clip": 1.03434074,
      "balance_loss_mlp": 1.00038266,
      "epoch": 0.9197378704984068,
      "flos": 23294893271040.0,
      "grad_norm": 2.066813429292795,
      "language_loss": 0.81233293,
      "learning_rate": 6.711059405947072e-08,
      "loss": 0.83056659,
      "num_input_tokens_seen": 165344645,
      "step": 7649,
      "time_per_iteration": 2.935617208480835
    },
    {
      "auxiliary_loss_clip": 0.01125699,
      "auxiliary_loss_mlp": 0.01022579,
      "balance_loss_clip": 1.0423975,
      "balance_loss_mlp": 1.0149225,
      "epoch": 0.9198581133890459,
      "flos": 20302564913280.0,
      "grad_norm": 4.760478162403137,
      "language_loss": 0.76938474,
      "learning_rate": 6.691064252791156e-08,
      "loss": 0.79086751,
      "num_input_tokens_seen": 165364120,
      "step": 7650,
      "time_per_iteration": 3.7339723110198975
    },
    {
      "auxiliary_loss_clip": 0.01130901,
      "auxiliary_loss_mlp": 0.010239,
      "balance_loss_clip": 1.04079044,
      "balance_loss_mlp": 1.0168488,
      "epoch": 0.9199783562796849,
      "flos": 17675663569920.0,
      "grad_norm": 1.7275439611744696,
      "language_loss": 0.7801429,
      "learning_rate": 6.67109842444713e-08,
      "loss": 0.80169094,
      "num_input_tokens_seen": 165383050,
      "step": 7651,
      "time_per_iteration": 2.7103021144866943
    },
    {
      "auxiliary_loss_clip": 0.0107789,
      "auxiliary_loss_mlp": 0.01028297,
      "balance_loss_clip": 1.03517711,
      "balance_loss_mlp": 1.02014887,
      "epoch": 0.9200985991703241,
      "flos": 17676022705920.0,
      "grad_norm": 2.7882315575225034,
      "language_loss": 0.77045548,
      "learning_rate": 6.651161923943704e-08,
      "loss": 0.79151732,
      "num_input_tokens_seen": 165400955,
      "step": 7652,
      "time_per_iteration": 2.7713212966918945
    },
    {
      "auxiliary_loss_clip": 0.01111038,
      "auxiliary_loss_mlp": 0.01025294,
      "balance_loss_clip": 1.04105735,
      "balance_loss_mlp": 1.01708031,
      "epoch": 0.9202188420609632,
      "flos": 20996574566400.0,
      "grad_norm": 2.541608917685375,
      "language_loss": 0.77120626,
      "learning_rate": 6.631254754305326e-08,
      "loss": 0.79256952,
      "num_input_tokens_seen": 165420415,
      "step": 7653,
      "time_per_iteration": 2.687771797180176
    },
    {
      "auxiliary_loss_clip": 0.01112322,
      "auxiliary_loss_mlp": 0.01020842,
      "balance_loss_clip": 1.03984404,
      "balance_loss_mlp": 1.01326013,
      "epoch": 0.9203390849516022,
      "flos": 13918222586880.0,
      "grad_norm": 1.7850619445762148,
      "language_loss": 0.78441548,
      "learning_rate": 6.611376918551848e-08,
      "loss": 0.80574715,
      "num_input_tokens_seen": 165439200,
      "step": 7654,
      "time_per_iteration": 2.6254801750183105
    },
    {
      "auxiliary_loss_clip": 0.01142802,
      "auxiliary_loss_mlp": 0.010266,
      "balance_loss_clip": 1.04382157,
      "balance_loss_mlp": 1.01910782,
      "epoch": 0.9204593278422414,
      "flos": 21175912195200.0,
      "grad_norm": 2.0133240189464705,
      "language_loss": 0.79511237,
      "learning_rate": 6.591528419698744e-08,
      "loss": 0.81680644,
      "num_input_tokens_seen": 165458985,
      "step": 7655,
      "time_per_iteration": 2.6639692783355713
    },
    {
      "auxiliary_loss_clip": 0.01112081,
      "auxiliary_loss_mlp": 0.01028049,
      "balance_loss_clip": 1.04442394,
      "balance_loss_mlp": 1.02082777,
      "epoch": 0.9205795707328804,
      "flos": 14501375890560.0,
      "grad_norm": 3.4777624807353082,
      "language_loss": 0.8329308,
      "learning_rate": 6.571709260756986e-08,
      "loss": 0.85433209,
      "num_input_tokens_seen": 165475630,
      "step": 7656,
      "time_per_iteration": 2.75909686088562
    },
    {
      "auxiliary_loss_clip": 0.01109367,
      "auxiliary_loss_mlp": 0.01027519,
      "balance_loss_clip": 1.04393864,
      "balance_loss_mlp": 1.01971364,
      "epoch": 0.9206998136235195,
      "flos": 22417559579520.0,
      "grad_norm": 3.5043442038210233,
      "language_loss": 0.76435673,
      "learning_rate": 6.551919444733122e-08,
      "loss": 0.78572559,
      "num_input_tokens_seen": 165493445,
      "step": 7657,
      "time_per_iteration": 3.73091197013855
    },
    {
      "auxiliary_loss_clip": 0.01133159,
      "auxiliary_loss_mlp": 0.0102389,
      "balance_loss_clip": 1.04315388,
      "balance_loss_mlp": 1.0162251,
      "epoch": 0.9208200565141585,
      "flos": 53358407544960.0,
      "grad_norm": 7.744720846467778,
      "language_loss": 0.65699261,
      "learning_rate": 6.53215897462931e-08,
      "loss": 0.67856306,
      "num_input_tokens_seen": 165517200,
      "step": 7658,
      "time_per_iteration": 3.8066415786743164
    },
    {
      "auxiliary_loss_clip": 0.01105158,
      "auxiliary_loss_mlp": 0.01020486,
      "balance_loss_clip": 1.03826106,
      "balance_loss_mlp": 1.01291037,
      "epoch": 0.9209402994047977,
      "flos": 30589139946240.0,
      "grad_norm": 2.0455742996328032,
      "language_loss": 0.74948162,
      "learning_rate": 6.512427853443103e-08,
      "loss": 0.77073807,
      "num_input_tokens_seen": 165539280,
      "step": 7659,
      "time_per_iteration": 2.7883193492889404
    },
    {
      "auxiliary_loss_clip": 0.01135534,
      "auxiliary_loss_mlp": 0.01032504,
      "balance_loss_clip": 1.04581356,
      "balance_loss_mlp": 1.02494621,
      "epoch": 0.9210605422954368,
      "flos": 29132711187840.0,
      "grad_norm": 1.737056725240179,
      "language_loss": 0.75615126,
      "learning_rate": 6.492726084167799e-08,
      "loss": 0.77783167,
      "num_input_tokens_seen": 165561395,
      "step": 7660,
      "time_per_iteration": 2.758995771408081
    },
    {
      "auxiliary_loss_clip": 0.0104646,
      "auxiliary_loss_mlp": 0.00754094,
      "balance_loss_clip": 1.01035821,
      "balance_loss_mlp": 1.00084066,
      "epoch": 0.9211807851860758,
      "flos": 54853838472960.0,
      "grad_norm": 0.7706637050125663,
      "language_loss": 0.57465619,
      "learning_rate": 6.473053669792072e-08,
      "loss": 0.59266174,
      "num_input_tokens_seen": 165616085,
      "step": 7661,
      "time_per_iteration": 4.0109570026397705
    },
    {
      "auxiliary_loss_clip": 0.01121028,
      "auxiliary_loss_mlp": 0.01026926,
      "balance_loss_clip": 1.04181278,
      "balance_loss_mlp": 1.01932013,
      "epoch": 0.921301028076715,
      "flos": 19201974238080.0,
      "grad_norm": 2.8516925463156313,
      "language_loss": 0.7285285,
      "learning_rate": 6.453410613300248e-08,
      "loss": 0.75000799,
      "num_input_tokens_seen": 165634015,
      "step": 7662,
      "time_per_iteration": 2.7749664783477783
    },
    {
      "auxiliary_loss_clip": 0.01141412,
      "auxiliary_loss_mlp": 0.01020476,
      "balance_loss_clip": 1.0463388,
      "balance_loss_mlp": 1.01296592,
      "epoch": 0.921421270967354,
      "flos": 27526893765120.0,
      "grad_norm": 2.0623849062269386,
      "language_loss": 0.58230382,
      "learning_rate": 6.43379691767214e-08,
      "loss": 0.60392272,
      "num_input_tokens_seen": 165653220,
      "step": 7663,
      "time_per_iteration": 2.7791342735290527
    },
    {
      "auxiliary_loss_clip": 0.01042831,
      "auxiliary_loss_mlp": 0.0100128,
      "balance_loss_clip": 1.01133394,
      "balance_loss_mlp": 1.00024307,
      "epoch": 0.9215415138579931,
      "flos": 70209311955840.0,
      "grad_norm": 0.7200681062382037,
      "language_loss": 0.55126119,
      "learning_rate": 6.414212585883105e-08,
      "loss": 0.5717023,
      "num_input_tokens_seen": 165715850,
      "step": 7664,
      "time_per_iteration": 3.3498148918151855
    },
    {
      "auxiliary_loss_clip": 0.01126029,
      "auxiliary_loss_mlp": 0.01023361,
      "balance_loss_clip": 1.0483489,
      "balance_loss_mlp": 1.01539159,
      "epoch": 0.9216617567486323,
      "flos": 35553107790720.0,
      "grad_norm": 1.545440316812092,
      "language_loss": 0.69950676,
      "learning_rate": 6.394657620904143e-08,
      "loss": 0.72100055,
      "num_input_tokens_seen": 165738960,
      "step": 7665,
      "time_per_iteration": 2.871730327606201
    },
    {
      "auxiliary_loss_clip": 0.01173198,
      "auxiliary_loss_mlp": 0.0102356,
      "balance_loss_clip": 1.04930592,
      "balance_loss_mlp": 1.01555538,
      "epoch": 0.9217819996392713,
      "flos": 29533330552320.0,
      "grad_norm": 1.894625308241934,
      "language_loss": 0.71804291,
      "learning_rate": 6.375132025701657e-08,
      "loss": 0.7400105,
      "num_input_tokens_seen": 165761260,
      "step": 7666,
      "time_per_iteration": 2.6502695083618164
    },
    {
      "auxiliary_loss_clip": 0.01130538,
      "auxiliary_loss_mlp": 0.01027771,
      "balance_loss_clip": 1.04816937,
      "balance_loss_mlp": 1.01910448,
      "epoch": 0.9219022425299104,
      "flos": 14574669592320.0,
      "grad_norm": 3.3562990301310527,
      "language_loss": 0.69723511,
      "learning_rate": 6.355635803237724e-08,
      "loss": 0.71881819,
      "num_input_tokens_seen": 165776960,
      "step": 7667,
      "time_per_iteration": 2.7044103145599365
    },
    {
      "auxiliary_loss_clip": 0.01130882,
      "auxiliary_loss_mlp": 0.01025136,
      "balance_loss_clip": 1.0448823,
      "balance_loss_mlp": 1.0178349,
      "epoch": 0.9220224854205495,
      "flos": 18077503996800.0,
      "grad_norm": 23.26642185380316,
      "language_loss": 0.79872823,
      "learning_rate": 6.336168956469867e-08,
      "loss": 0.82028836,
      "num_input_tokens_seen": 165795435,
      "step": 7668,
      "time_per_iteration": 3.7187371253967285
    },
    {
      "auxiliary_loss_clip": 0.0109511,
      "auxiliary_loss_mlp": 0.01023142,
      "balance_loss_clip": 1.03902376,
      "balance_loss_mlp": 1.01556659,
      "epoch": 0.9221427283111886,
      "flos": 24790464875520.0,
      "grad_norm": 2.2225258237035868,
      "language_loss": 0.72041869,
      "learning_rate": 6.316731488351168e-08,
      "loss": 0.74160123,
      "num_input_tokens_seen": 165816625,
      "step": 7669,
      "time_per_iteration": 2.798501491546631
    },
    {
      "auxiliary_loss_clip": 0.01044407,
      "auxiliary_loss_mlp": 0.01023344,
      "balance_loss_clip": 1.03234518,
      "balance_loss_mlp": 1.01518989,
      "epoch": 0.9222629712018277,
      "flos": 13845036625920.0,
      "grad_norm": 1.7530045046267841,
      "language_loss": 0.63221061,
      "learning_rate": 6.297323401830334e-08,
      "loss": 0.65288818,
      "num_input_tokens_seen": 165835410,
      "step": 7670,
      "time_per_iteration": 3.1806600093841553
    },
    {
      "auxiliary_loss_clip": 0.01138637,
      "auxiliary_loss_mlp": 0.01023982,
      "balance_loss_clip": 1.04199135,
      "balance_loss_mlp": 1.01576805,
      "epoch": 0.9223832140924668,
      "flos": 21616177196160.0,
      "grad_norm": 2.5863581442258163,
      "language_loss": 0.69182795,
      "learning_rate": 6.277944699851523e-08,
      "loss": 0.71345413,
      "num_input_tokens_seen": 165854930,
      "step": 7671,
      "time_per_iteration": 2.855532646179199
    },
    {
      "auxiliary_loss_clip": 0.01149622,
      "auxiliary_loss_mlp": 0.01027311,
      "balance_loss_clip": 1.04446185,
      "balance_loss_mlp": 1.01949096,
      "epoch": 0.9225034569831059,
      "flos": 21142084561920.0,
      "grad_norm": 2.0269473681776944,
      "language_loss": 0.73496276,
      "learning_rate": 6.25859538535447e-08,
      "loss": 0.75673211,
      "num_input_tokens_seen": 165875725,
      "step": 7672,
      "time_per_iteration": 2.6633243560791016
    },
    {
      "auxiliary_loss_clip": 0.01116263,
      "auxiliary_loss_mlp": 0.01024321,
      "balance_loss_clip": 1.04368639,
      "balance_loss_mlp": 1.01577997,
      "epoch": 0.9226236998737449,
      "flos": 12495046844160.0,
      "grad_norm": 2.843031388500655,
      "language_loss": 0.77662659,
      "learning_rate": 6.239275461274474e-08,
      "loss": 0.7980324,
      "num_input_tokens_seen": 165892100,
      "step": 7673,
      "time_per_iteration": 2.6852223873138428
    },
    {
      "auxiliary_loss_clip": 0.01157528,
      "auxiliary_loss_mlp": 0.01025284,
      "balance_loss_clip": 1.04714954,
      "balance_loss_mlp": 1.01810813,
      "epoch": 0.9227439427643841,
      "flos": 26214071581440.0,
      "grad_norm": 1.7393055555377,
      "language_loss": 0.86212993,
      "learning_rate": 6.219984930542299e-08,
      "loss": 0.8839581,
      "num_input_tokens_seen": 165912840,
      "step": 7674,
      "time_per_iteration": 2.7077105045318604
    },
    {
      "auxiliary_loss_clip": 0.010789,
      "auxiliary_loss_mlp": 0.00761859,
      "balance_loss_clip": 1.03769946,
      "balance_loss_mlp": 1.00039291,
      "epoch": 0.9228641856550232,
      "flos": 17967581400960.0,
      "grad_norm": 2.149461234315487,
      "language_loss": 0.75719786,
      "learning_rate": 6.200723796084383e-08,
      "loss": 0.77560544,
      "num_input_tokens_seen": 165930935,
      "step": 7675,
      "time_per_iteration": 2.845818281173706
    },
    {
      "auxiliary_loss_clip": 0.01043036,
      "auxiliary_loss_mlp": 0.01000448,
      "balance_loss_clip": 1.00956869,
      "balance_loss_mlp": 0.99935114,
      "epoch": 0.9229844285456622,
      "flos": 70420609710720.0,
      "grad_norm": 0.7630516515822191,
      "language_loss": 0.63075006,
      "learning_rate": 6.181492060822546e-08,
      "loss": 0.6511848,
      "num_input_tokens_seen": 165991110,
      "step": 7676,
      "time_per_iteration": 3.1626782417297363
    },
    {
      "auxiliary_loss_clip": 0.01125332,
      "auxiliary_loss_mlp": 0.01025846,
      "balance_loss_clip": 1.044258,
      "balance_loss_mlp": 1.01864016,
      "epoch": 0.9231046714363014,
      "flos": 17967832796160.0,
      "grad_norm": 2.315020638544029,
      "language_loss": 0.81797123,
      "learning_rate": 6.162289727674274e-08,
      "loss": 0.83948302,
      "num_input_tokens_seen": 166008790,
      "step": 7677,
      "time_per_iteration": 2.697559356689453
    },
    {
      "auxiliary_loss_clip": 0.01124394,
      "auxiliary_loss_mlp": 0.01021826,
      "balance_loss_clip": 1.04184556,
      "balance_loss_mlp": 1.0149473,
      "epoch": 0.9232249143269404,
      "flos": 17858233422720.0,
      "grad_norm": 2.373671101906313,
      "language_loss": 0.87822837,
      "learning_rate": 6.143116799552527e-08,
      "loss": 0.89969051,
      "num_input_tokens_seen": 166025035,
      "step": 7678,
      "time_per_iteration": 2.67546010017395
    },
    {
      "auxiliary_loss_clip": 0.01109612,
      "auxiliary_loss_mlp": 0.01029821,
      "balance_loss_clip": 1.0395875,
      "balance_loss_mlp": 1.02213156,
      "epoch": 0.9233451572175795,
      "flos": 23404384903680.0,
      "grad_norm": 2.870579689841005,
      "language_loss": 0.56102002,
      "learning_rate": 6.123973279365802e-08,
      "loss": 0.58241439,
      "num_input_tokens_seen": 166044010,
      "step": 7679,
      "time_per_iteration": 2.8054609298706055
    },
    {
      "auxiliary_loss_clip": 0.01153287,
      "auxiliary_loss_mlp": 0.01029402,
      "balance_loss_clip": 1.04791617,
      "balance_loss_mlp": 1.02186167,
      "epoch": 0.9234654001082186,
      "flos": 17999326045440.0,
      "grad_norm": 2.2247208879021256,
      "language_loss": 0.77478343,
      "learning_rate": 6.10485917001824e-08,
      "loss": 0.7966103,
      "num_input_tokens_seen": 166061865,
      "step": 7680,
      "time_per_iteration": 2.6432061195373535
    },
    {
      "auxiliary_loss_clip": 0.01151575,
      "auxiliary_loss_mlp": 0.01023536,
      "balance_loss_clip": 1.04453123,
      "balance_loss_mlp": 1.01646411,
      "epoch": 0.9235856429988577,
      "flos": 24750747411840.0,
      "grad_norm": 1.6513622277918198,
      "language_loss": 0.81346339,
      "learning_rate": 6.085774474409322e-08,
      "loss": 0.8352145,
      "num_input_tokens_seen": 166082425,
      "step": 7681,
      "time_per_iteration": 2.7197794914245605
    },
    {
      "auxiliary_loss_clip": 0.01108311,
      "auxiliary_loss_mlp": 0.01024719,
      "balance_loss_clip": 1.04188335,
      "balance_loss_mlp": 1.01726806,
      "epoch": 0.9237058858894968,
      "flos": 14099894599680.0,
      "grad_norm": 1.8660828967626297,
      "language_loss": 0.69799495,
      "learning_rate": 6.066719195434267e-08,
      "loss": 0.7193253,
      "num_input_tokens_seen": 166100225,
      "step": 7682,
      "time_per_iteration": 2.6872894763946533
    },
    {
      "auxiliary_loss_clip": 0.01130102,
      "auxiliary_loss_mlp": 0.01028609,
      "balance_loss_clip": 1.04551303,
      "balance_loss_mlp": 1.02026999,
      "epoch": 0.9238261287801359,
      "flos": 28694529175680.0,
      "grad_norm": 2.1525410424748936,
      "language_loss": 0.66391337,
      "learning_rate": 6.047693335983717e-08,
      "loss": 0.6855005,
      "num_input_tokens_seen": 166122570,
      "step": 7683,
      "time_per_iteration": 2.8088231086730957
    },
    {
      "auxiliary_loss_clip": 0.01117874,
      "auxiliary_loss_mlp": 0.01023043,
      "balance_loss_clip": 1.03914702,
      "balance_loss_mlp": 1.01468682,
      "epoch": 0.923946371670775,
      "flos": 23111856541440.0,
      "grad_norm": 2.682953692345635,
      "language_loss": 0.83084726,
      "learning_rate": 6.028696898943853e-08,
      "loss": 0.85225642,
      "num_input_tokens_seen": 166141630,
      "step": 7684,
      "time_per_iteration": 4.525881767272949
    },
    {
      "auxiliary_loss_clip": 0.01108283,
      "auxiliary_loss_mlp": 0.00762788,
      "balance_loss_clip": 1.03713894,
      "balance_loss_mlp": 1.00042951,
      "epoch": 0.924066614561414,
      "flos": 21867120587520.0,
      "grad_norm": 4.2290205630755855,
      "language_loss": 0.70722741,
      "learning_rate": 6.00972988719648e-08,
      "loss": 0.72593814,
      "num_input_tokens_seen": 166159865,
      "step": 7685,
      "time_per_iteration": 2.773545742034912
    },
    {
      "auxiliary_loss_clip": 0.01147947,
      "auxiliary_loss_mlp": 0.01023068,
      "balance_loss_clip": 1.04454124,
      "balance_loss_mlp": 1.0149529,
      "epoch": 0.9241868574520532,
      "flos": 28511887495680.0,
      "grad_norm": 9.920075816987007,
      "language_loss": 0.70424032,
      "learning_rate": 5.990792303618807e-08,
      "loss": 0.72595048,
      "num_input_tokens_seen": 166179445,
      "step": 7686,
      "time_per_iteration": 2.735482692718506
    },
    {
      "auxiliary_loss_clip": 0.01142644,
      "auxiliary_loss_mlp": 0.01025198,
      "balance_loss_clip": 1.04519033,
      "balance_loss_mlp": 1.01813531,
      "epoch": 0.9243071003426923,
      "flos": 30518324282880.0,
      "grad_norm": 1.8465014165480969,
      "language_loss": 0.69742024,
      "learning_rate": 5.971884151083695e-08,
      "loss": 0.71909863,
      "num_input_tokens_seen": 166201855,
      "step": 7687,
      "time_per_iteration": 4.178483724594116
    },
    {
      "auxiliary_loss_clip": 0.01123876,
      "auxiliary_loss_mlp": 0.01022556,
      "balance_loss_clip": 1.04405761,
      "balance_loss_mlp": 1.01530814,
      "epoch": 0.9244273432333313,
      "flos": 28658331244800.0,
      "grad_norm": 1.9623339211095154,
      "language_loss": 0.7432301,
      "learning_rate": 5.9530054324595124e-08,
      "loss": 0.76469439,
      "num_input_tokens_seen": 166221970,
      "step": 7688,
      "time_per_iteration": 2.7697722911834717
    },
    {
      "auxiliary_loss_clip": 0.01047976,
      "auxiliary_loss_mlp": 0.01000435,
      "balance_loss_clip": 1.01039755,
      "balance_loss_mlp": 0.99939805,
      "epoch": 0.9245475861239704,
      "flos": 66230589237120.0,
      "grad_norm": 0.7154596585502067,
      "language_loss": 0.57475632,
      "learning_rate": 5.934156150610103e-08,
      "loss": 0.59524041,
      "num_input_tokens_seen": 166279335,
      "step": 7689,
      "time_per_iteration": 3.3235039710998535
    },
    {
      "auxiliary_loss_clip": 0.01130282,
      "auxiliary_loss_mlp": 0.01023077,
      "balance_loss_clip": 1.04247892,
      "balance_loss_mlp": 1.01479769,
      "epoch": 0.9246678290146095,
      "flos": 24239918142720.0,
      "grad_norm": 2.814715589900303,
      "language_loss": 0.79127884,
      "learning_rate": 5.915336308394914e-08,
      "loss": 0.81281239,
      "num_input_tokens_seen": 166298170,
      "step": 7690,
      "time_per_iteration": 2.7753279209136963
    },
    {
      "auxiliary_loss_clip": 0.01154649,
      "auxiliary_loss_mlp": 0.01020111,
      "balance_loss_clip": 1.04372382,
      "balance_loss_mlp": 1.01327991,
      "epoch": 0.9247880719052486,
      "flos": 18988808976000.0,
      "grad_norm": 1.8165209371151996,
      "language_loss": 0.77151883,
      "learning_rate": 5.89654590866886e-08,
      "loss": 0.79326642,
      "num_input_tokens_seen": 166317670,
      "step": 7691,
      "time_per_iteration": 2.5931270122528076
    },
    {
      "auxiliary_loss_clip": 0.0108636,
      "auxiliary_loss_mlp": 0.00763624,
      "balance_loss_clip": 1.04063153,
      "balance_loss_mlp": 1.00036383,
      "epoch": 0.9249083147958876,
      "flos": 24024095274240.0,
      "grad_norm": 1.9914322196139655,
      "language_loss": 0.88577616,
      "learning_rate": 5.877784954282483e-08,
      "loss": 0.90427601,
      "num_input_tokens_seen": 166337010,
      "step": 7692,
      "time_per_iteration": 2.849154233932495
    },
    {
      "auxiliary_loss_clip": 0.01107749,
      "auxiliary_loss_mlp": 0.01025458,
      "balance_loss_clip": 1.0368247,
      "balance_loss_mlp": 1.01711655,
      "epoch": 0.9250285576865268,
      "flos": 30773972355840.0,
      "grad_norm": 2.040179858569698,
      "language_loss": 0.72772157,
      "learning_rate": 5.8590534480817963e-08,
      "loss": 0.7490536,
      "num_input_tokens_seen": 166358735,
      "step": 7693,
      "time_per_iteration": 2.7924253940582275
    },
    {
      "auxiliary_loss_clip": 0.01143796,
      "auxiliary_loss_mlp": 0.00762411,
      "balance_loss_clip": 1.04578638,
      "balance_loss_mlp": 1.0005039,
      "epoch": 0.9251488005771659,
      "flos": 10633581348480.0,
      "grad_norm": 2.610707328010366,
      "language_loss": 0.72872078,
      "learning_rate": 5.840351392908349e-08,
      "loss": 0.74778283,
      "num_input_tokens_seen": 166374455,
      "step": 7694,
      "time_per_iteration": 3.6324615478515625
    },
    {
      "auxiliary_loss_clip": 0.01132139,
      "auxiliary_loss_mlp": 0.01031368,
      "balance_loss_clip": 1.04544258,
      "balance_loss_mlp": 1.02332759,
      "epoch": 0.9252690434678049,
      "flos": 23586416052480.0,
      "grad_norm": 2.4136599780682646,
      "language_loss": 0.70323229,
      "learning_rate": 5.821678791599205e-08,
      "loss": 0.72486734,
      "num_input_tokens_seen": 166393900,
      "step": 7695,
      "time_per_iteration": 2.8676111698150635
    },
    {
      "auxiliary_loss_clip": 0.01071891,
      "auxiliary_loss_mlp": 0.01026883,
      "balance_loss_clip": 1.03522205,
      "balance_loss_mlp": 1.01891124,
      "epoch": 0.9253892863584441,
      "flos": 21469158829440.0,
      "grad_norm": 3.1906254207198774,
      "language_loss": 0.80561972,
      "learning_rate": 5.803035646986965e-08,
      "loss": 0.82660747,
      "num_input_tokens_seen": 166413235,
      "step": 7696,
      "time_per_iteration": 2.917720317840576
    },
    {
      "auxiliary_loss_clip": 0.01130044,
      "auxiliary_loss_mlp": 0.01031982,
      "balance_loss_clip": 1.04137576,
      "balance_loss_mlp": 1.02388811,
      "epoch": 0.9255095292490831,
      "flos": 17456680304640.0,
      "grad_norm": 2.418887069617104,
      "language_loss": 0.67828882,
      "learning_rate": 5.7844219618998766e-08,
      "loss": 0.69990909,
      "num_input_tokens_seen": 166427560,
      "step": 7697,
      "time_per_iteration": 2.861776351928711
    },
    {
      "auxiliary_loss_clip": 0.01072342,
      "auxiliary_loss_mlp": 0.01025094,
      "balance_loss_clip": 1.03688431,
      "balance_loss_mlp": 1.01764965,
      "epoch": 0.9256297721397222,
      "flos": 24750675584640.0,
      "grad_norm": 1.6874332095690754,
      "language_loss": 0.71848983,
      "learning_rate": 5.765837739161505e-08,
      "loss": 0.73946416,
      "num_input_tokens_seen": 166446680,
      "step": 7698,
      "time_per_iteration": 2.984147787094116
    },
    {
      "auxiliary_loss_clip": 0.01143105,
      "auxiliary_loss_mlp": 0.01023098,
      "balance_loss_clip": 1.04453993,
      "balance_loss_mlp": 1.01586223,
      "epoch": 0.9257500150303614,
      "flos": 23112215677440.0,
      "grad_norm": 3.404075314287517,
      "language_loss": 0.74416852,
      "learning_rate": 5.7472829815911504e-08,
      "loss": 0.76583058,
      "num_input_tokens_seen": 166465505,
      "step": 7699,
      "time_per_iteration": 3.1126043796539307
    },
    {
      "auxiliary_loss_clip": 0.0112191,
      "auxiliary_loss_mlp": 0.0076201,
      "balance_loss_clip": 1.0401547,
      "balance_loss_mlp": 1.00041521,
      "epoch": 0.9258702579210004,
      "flos": 22564685687040.0,
      "grad_norm": 1.7753492730651836,
      "language_loss": 0.81501114,
      "learning_rate": 5.7287576920035164e-08,
      "loss": 0.83385032,
      "num_input_tokens_seen": 166484520,
      "step": 7700,
      "time_per_iteration": 2.758776903152466
    },
    {
      "auxiliary_loss_clip": 0.01153621,
      "auxiliary_loss_mlp": 0.01018898,
      "balance_loss_clip": 1.04429042,
      "balance_loss_mlp": 1.01145315,
      "epoch": 0.9259905008116395,
      "flos": 30004298703360.0,
      "grad_norm": 1.997883790482883,
      "language_loss": 0.76611483,
      "learning_rate": 5.7102618732088435e-08,
      "loss": 0.78784013,
      "num_input_tokens_seen": 166503850,
      "step": 7701,
      "time_per_iteration": 2.687059164047241
    },
    {
      "auxiliary_loss_clip": 0.01096664,
      "auxiliary_loss_mlp": 0.00761141,
      "balance_loss_clip": 1.04036033,
      "balance_loss_mlp": 1.00050545,
      "epoch": 0.9261107437022786,
      "flos": 24572128055040.0,
      "grad_norm": 1.632400213252303,
      "language_loss": 0.74882185,
      "learning_rate": 5.6917955280130216e-08,
      "loss": 0.76739991,
      "num_input_tokens_seen": 166525330,
      "step": 7702,
      "time_per_iteration": 2.8283770084381104
    },
    {
      "auxiliary_loss_clip": 0.0112142,
      "auxiliary_loss_mlp": 0.01024665,
      "balance_loss_clip": 1.03918076,
      "balance_loss_mlp": 1.01646376,
      "epoch": 0.9262309865929177,
      "flos": 22018448586240.0,
      "grad_norm": 2.4313834151808726,
      "language_loss": 0.72117054,
      "learning_rate": 5.6733586592172755e-08,
      "loss": 0.74263144,
      "num_input_tokens_seen": 166544825,
      "step": 7703,
      "time_per_iteration": 2.723881244659424
    },
    {
      "auxiliary_loss_clip": 0.01119241,
      "auxiliary_loss_mlp": 0.01021624,
      "balance_loss_clip": 1.03970528,
      "balance_loss_mlp": 1.01432586,
      "epoch": 0.9263512294835567,
      "flos": 20339481116160.0,
      "grad_norm": 2.090383451712382,
      "language_loss": 0.7999022,
      "learning_rate": 5.6549512696185244e-08,
      "loss": 0.82131076,
      "num_input_tokens_seen": 166563325,
      "step": 7704,
      "time_per_iteration": 2.7255699634552
    },
    {
      "auxiliary_loss_clip": 0.01125378,
      "auxiliary_loss_mlp": 0.0076224,
      "balance_loss_clip": 1.04483557,
      "balance_loss_mlp": 1.0004251,
      "epoch": 0.9264714723741959,
      "flos": 21215378263680.0,
      "grad_norm": 3.3087525736853305,
      "language_loss": 0.68326962,
      "learning_rate": 5.636573362009156e-08,
      "loss": 0.70214576,
      "num_input_tokens_seen": 166583385,
      "step": 7705,
      "time_per_iteration": 2.7402026653289795
    },
    {
      "auxiliary_loss_clip": 0.01095183,
      "auxiliary_loss_mlp": 0.00762977,
      "balance_loss_clip": 1.03785253,
      "balance_loss_mlp": 1.00040686,
      "epoch": 0.926591715264835,
      "flos": 18004964480640.0,
      "grad_norm": 2.123498252209385,
      "language_loss": 0.77226543,
      "learning_rate": 5.618224939177074e-08,
      "loss": 0.79084706,
      "num_input_tokens_seen": 166601290,
      "step": 7706,
      "time_per_iteration": 2.751391887664795
    },
    {
      "auxiliary_loss_clip": 0.01076377,
      "auxiliary_loss_mlp": 0.01023248,
      "balance_loss_clip": 1.03641701,
      "balance_loss_mlp": 1.01572919,
      "epoch": 0.926711958155474,
      "flos": 36167969825280.0,
      "grad_norm": 2.2435257574981358,
      "language_loss": 0.70483077,
      "learning_rate": 5.599906003905719e-08,
      "loss": 0.72582698,
      "num_input_tokens_seen": 166623835,
      "step": 7707,
      "time_per_iteration": 2.931312084197998
    },
    {
      "auxiliary_loss_clip": 0.01137209,
      "auxiliary_loss_mlp": 0.0102288,
      "balance_loss_clip": 1.04270148,
      "balance_loss_mlp": 1.0149467,
      "epoch": 0.9268322010461132,
      "flos": 21032736583680.0,
      "grad_norm": 2.2826329712021027,
      "language_loss": 0.81386608,
      "learning_rate": 5.581616558974023e-08,
      "loss": 0.83546698,
      "num_input_tokens_seen": 166642400,
      "step": 7708,
      "time_per_iteration": 2.6597750186920166
    },
    {
      "auxiliary_loss_clip": 0.01147173,
      "auxiliary_loss_mlp": 0.01024372,
      "balance_loss_clip": 1.0471375,
      "balance_loss_mlp": 1.01668859,
      "epoch": 0.9269524439367522,
      "flos": 22964838174720.0,
      "grad_norm": 1.8849069898650892,
      "language_loss": 0.79382491,
      "learning_rate": 5.5633566071565444e-08,
      "loss": 0.81554037,
      "num_input_tokens_seen": 166661640,
      "step": 7709,
      "time_per_iteration": 2.697882890701294
    },
    {
      "auxiliary_loss_clip": 0.01117888,
      "auxiliary_loss_mlp": 0.01021575,
      "balance_loss_clip": 1.03837907,
      "balance_loss_mlp": 1.01420188,
      "epoch": 0.9270726868273913,
      "flos": 41975551468800.0,
      "grad_norm": 2.0567208918904205,
      "language_loss": 0.71109378,
      "learning_rate": 5.5451261512232896e-08,
      "loss": 0.73248845,
      "num_input_tokens_seen": 166684320,
      "step": 7710,
      "time_per_iteration": 4.695790767669678
    },
    {
      "auxiliary_loss_clip": 0.01154538,
      "auxiliary_loss_mlp": 0.0102055,
      "balance_loss_clip": 1.04520273,
      "balance_loss_mlp": 1.01265228,
      "epoch": 0.9271929297180305,
      "flos": 19791771557760.0,
      "grad_norm": 2.423359182872409,
      "language_loss": 0.62327093,
      "learning_rate": 5.5269251939397576e-08,
      "loss": 0.6450218,
      "num_input_tokens_seen": 166703835,
      "step": 7711,
      "time_per_iteration": 2.7690513134002686
    },
    {
      "auxiliary_loss_clip": 0.01143224,
      "auxiliary_loss_mlp": 0.00762358,
      "balance_loss_clip": 1.04360723,
      "balance_loss_mlp": 1.00039577,
      "epoch": 0.9273131726086695,
      "flos": 19968343839360.0,
      "grad_norm": 2.348762756506912,
      "language_loss": 0.76834512,
      "learning_rate": 5.508753738067073e-08,
      "loss": 0.78740096,
      "num_input_tokens_seen": 166723375,
      "step": 7712,
      "time_per_iteration": 2.694146156311035
    },
    {
      "auxiliary_loss_clip": 0.01128622,
      "auxiliary_loss_mlp": 0.01022903,
      "balance_loss_clip": 1.0422256,
      "balance_loss_mlp": 1.01527929,
      "epoch": 0.9274334154993086,
      "flos": 23258587599360.0,
      "grad_norm": 2.295039780091966,
      "language_loss": 0.79309678,
      "learning_rate": 5.4906117863617875e-08,
      "loss": 0.81461203,
      "num_input_tokens_seen": 166742760,
      "step": 7713,
      "time_per_iteration": 3.7915456295013428
    },
    {
      "auxiliary_loss_clip": 0.0116659,
      "auxiliary_loss_mlp": 0.01021211,
      "balance_loss_clip": 1.04542184,
      "balance_loss_mlp": 1.0141089,
      "epoch": 0.9275536583899477,
      "flos": 31795343585280.0,
      "grad_norm": 3.094597180680298,
      "language_loss": 0.78438586,
      "learning_rate": 5.4724993415760533e-08,
      "loss": 0.8062638,
      "num_input_tokens_seen": 166761115,
      "step": 7714,
      "time_per_iteration": 2.6776621341705322
    },
    {
      "auxiliary_loss_clip": 0.01108961,
      "auxiliary_loss_mlp": 0.01026723,
      "balance_loss_clip": 1.04206276,
      "balance_loss_mlp": 1.01846504,
      "epoch": 0.9276739012805868,
      "flos": 18696998885760.0,
      "grad_norm": 5.8275530457704665,
      "language_loss": 0.7480889,
      "learning_rate": 5.454416406457496e-08,
      "loss": 0.76944572,
      "num_input_tokens_seen": 166780210,
      "step": 7715,
      "time_per_iteration": 2.6651694774627686
    },
    {
      "auxiliary_loss_clip": 0.01131479,
      "auxiliary_loss_mlp": 0.01023482,
      "balance_loss_clip": 1.04228079,
      "balance_loss_mlp": 1.01675248,
      "epoch": 0.9277941441712259,
      "flos": 13879079740800.0,
      "grad_norm": 4.756378998901257,
      "language_loss": 0.740219,
      "learning_rate": 5.436362983749299e-08,
      "loss": 0.76176864,
      "num_input_tokens_seen": 166795380,
      "step": 7716,
      "time_per_iteration": 2.7383880615234375
    },
    {
      "auxiliary_loss_clip": 0.01140601,
      "auxiliary_loss_mlp": 0.01021784,
      "balance_loss_clip": 1.04350448,
      "balance_loss_mlp": 1.01452363,
      "epoch": 0.927914387061865,
      "flos": 23258659426560.0,
      "grad_norm": 6.967462874994978,
      "language_loss": 0.64595687,
      "learning_rate": 5.418339076190137e-08,
      "loss": 0.66758072,
      "num_input_tokens_seen": 166814890,
      "step": 7717,
      "time_per_iteration": 2.7395191192626953
    },
    {
      "auxiliary_loss_clip": 0.01124982,
      "auxiliary_loss_mlp": 0.0103058,
      "balance_loss_clip": 1.04289699,
      "balance_loss_mlp": 1.02231252,
      "epoch": 0.9280346299525041,
      "flos": 18073733068800.0,
      "grad_norm": 1.8286305605795516,
      "language_loss": 0.88838232,
      "learning_rate": 5.400344686514202e-08,
      "loss": 0.90993792,
      "num_input_tokens_seen": 166832475,
      "step": 7718,
      "time_per_iteration": 2.706284523010254
    },
    {
      "auxiliary_loss_clip": 0.01142706,
      "auxiliary_loss_mlp": 0.01022188,
      "balance_loss_clip": 1.04613566,
      "balance_loss_mlp": 1.01482677,
      "epoch": 0.9281548728431431,
      "flos": 22342901160960.0,
      "grad_norm": 3.3927006741599426,
      "language_loss": 0.66797084,
      "learning_rate": 5.38237981745131e-08,
      "loss": 0.68961978,
      "num_input_tokens_seen": 166850590,
      "step": 7719,
      "time_per_iteration": 3.8272511959075928
    },
    {
      "auxiliary_loss_clip": 0.01149986,
      "auxiliary_loss_mlp": 0.01030116,
      "balance_loss_clip": 1.04423988,
      "balance_loss_mlp": 1.02254295,
      "epoch": 0.9282751157337822,
      "flos": 18843765857280.0,
      "grad_norm": 2.5186969448923073,
      "language_loss": 0.81702328,
      "learning_rate": 5.364444471726592e-08,
      "loss": 0.83882427,
      "num_input_tokens_seen": 166869795,
      "step": 7720,
      "time_per_iteration": 2.6483640670776367
    },
    {
      "auxiliary_loss_clip": 0.01146753,
      "auxiliary_loss_mlp": 0.01027537,
      "balance_loss_clip": 1.04347825,
      "balance_loss_mlp": 1.01956522,
      "epoch": 0.9283953586244214,
      "flos": 25556834476800.0,
      "grad_norm": 2.16623590233392,
      "language_loss": 0.79806173,
      "learning_rate": 5.346538652060939e-08,
      "loss": 0.81980461,
      "num_input_tokens_seen": 166891150,
      "step": 7721,
      "time_per_iteration": 2.7557599544525146
    },
    {
      "auxiliary_loss_clip": 0.01136381,
      "auxiliary_loss_mlp": 0.01030791,
      "balance_loss_clip": 1.04473102,
      "balance_loss_mlp": 1.02362084,
      "epoch": 0.9285156015150604,
      "flos": 18223480869120.0,
      "grad_norm": 2.08532631379262,
      "language_loss": 0.70613945,
      "learning_rate": 5.3286623611705994e-08,
      "loss": 0.72781116,
      "num_input_tokens_seen": 166909195,
      "step": 7722,
      "time_per_iteration": 2.6202330589294434
    },
    {
      "auxiliary_loss_clip": 0.01031911,
      "auxiliary_loss_mlp": 0.01001975,
      "balance_loss_clip": 1.010782,
      "balance_loss_mlp": 1.00103366,
      "epoch": 0.9286358444056995,
      "flos": 66400017690240.0,
      "grad_norm": 0.8174970566949777,
      "language_loss": 0.60629952,
      "learning_rate": 5.3108156017673824e-08,
      "loss": 0.62663841,
      "num_input_tokens_seen": 166970955,
      "step": 7723,
      "time_per_iteration": 3.407041549682617
    },
    {
      "auxiliary_loss_clip": 0.01123981,
      "auxiliary_loss_mlp": 0.01030116,
      "balance_loss_clip": 1.04073191,
      "balance_loss_mlp": 1.02130055,
      "epoch": 0.9287560872963386,
      "flos": 22345630594560.0,
      "grad_norm": 1.7000349970280055,
      "language_loss": 0.71966827,
      "learning_rate": 5.2929983765586775e-08,
      "loss": 0.74120927,
      "num_input_tokens_seen": 166989735,
      "step": 7724,
      "time_per_iteration": 2.8518340587615967
    },
    {
      "auxiliary_loss_clip": 0.01127605,
      "auxiliary_loss_mlp": 0.01021831,
      "balance_loss_clip": 1.04486954,
      "balance_loss_mlp": 1.01412714,
      "epoch": 0.9288763301869777,
      "flos": 25700225569920.0,
      "grad_norm": 1.89439276867645,
      "language_loss": 0.62318385,
      "learning_rate": 5.275210688247278e-08,
      "loss": 0.64467818,
      "num_input_tokens_seen": 167010060,
      "step": 7725,
      "time_per_iteration": 2.7287068367004395
    },
    {
      "auxiliary_loss_clip": 0.01093788,
      "auxiliary_loss_mlp": 0.01023378,
      "balance_loss_clip": 1.0367502,
      "balance_loss_mlp": 1.01587939,
      "epoch": 0.9289965730776167,
      "flos": 12312046028160.0,
      "grad_norm": 1.9787643451675088,
      "language_loss": 0.84947252,
      "learning_rate": 5.257452539531604e-08,
      "loss": 0.87064415,
      "num_input_tokens_seen": 167027130,
      "step": 7726,
      "time_per_iteration": 2.7210373878479004
    },
    {
      "auxiliary_loss_clip": 0.01125454,
      "auxiliary_loss_mlp": 0.01023415,
      "balance_loss_clip": 1.0384295,
      "balance_loss_mlp": 1.01583076,
      "epoch": 0.9291168159682559,
      "flos": 26685973486080.0,
      "grad_norm": 1.7543639258619088,
      "language_loss": 0.68433058,
      "learning_rate": 5.2397239331055445e-08,
      "loss": 0.70581937,
      "num_input_tokens_seen": 167049130,
      "step": 7727,
      "time_per_iteration": 2.777090549468994
    },
    {
      "auxiliary_loss_clip": 0.01131186,
      "auxiliary_loss_mlp": 0.01028287,
      "balance_loss_clip": 1.04460573,
      "balance_loss_mlp": 1.02087569,
      "epoch": 0.929237058858895,
      "flos": 14538256179840.0,
      "grad_norm": 2.2374789417276784,
      "language_loss": 0.81482112,
      "learning_rate": 5.2220248716585036e-08,
      "loss": 0.83641583,
      "num_input_tokens_seen": 167066810,
      "step": 7728,
      "time_per_iteration": 2.7761712074279785
    },
    {
      "auxiliary_loss_clip": 0.01145716,
      "auxiliary_loss_mlp": 0.01025635,
      "balance_loss_clip": 1.0444243,
      "balance_loss_mlp": 1.0183512,
      "epoch": 0.929357301749534,
      "flos": 23835456023040.0,
      "grad_norm": 2.6602355085213545,
      "language_loss": 0.75748807,
      "learning_rate": 5.204355357875445e-08,
      "loss": 0.77920163,
      "num_input_tokens_seen": 167085155,
      "step": 7729,
      "time_per_iteration": 2.7379541397094727
    },
    {
      "auxiliary_loss_clip": 0.01129352,
      "auxiliary_loss_mlp": 0.01025192,
      "balance_loss_clip": 1.04337525,
      "balance_loss_mlp": 1.01766419,
      "epoch": 0.9294775446401732,
      "flos": 12969319046400.0,
      "grad_norm": 2.109607248493187,
      "language_loss": 0.7022928,
      "learning_rate": 5.1867153944367584e-08,
      "loss": 0.72383821,
      "num_input_tokens_seen": 167101545,
      "step": 7730,
      "time_per_iteration": 2.722832679748535
    },
    {
      "auxiliary_loss_clip": 0.01132029,
      "auxiliary_loss_mlp": 0.01020599,
      "balance_loss_clip": 1.04264474,
      "balance_loss_mlp": 1.01319361,
      "epoch": 0.9295977875308122,
      "flos": 26211809024640.0,
      "grad_norm": 1.637934798961217,
      "language_loss": 0.7364924,
      "learning_rate": 5.16910498401848e-08,
      "loss": 0.75801867,
      "num_input_tokens_seen": 167120995,
      "step": 7731,
      "time_per_iteration": 2.818753719329834
    },
    {
      "auxiliary_loss_clip": 0.01124247,
      "auxiliary_loss_mlp": 0.01026286,
      "balance_loss_clip": 1.04216862,
      "balance_loss_mlp": 1.01893687,
      "epoch": 0.9297180304214513,
      "flos": 16472297105280.0,
      "grad_norm": 2.0552519523216386,
      "language_loss": 0.83403969,
      "learning_rate": 5.151524129292073e-08,
      "loss": 0.85554498,
      "num_input_tokens_seen": 167138890,
      "step": 7732,
      "time_per_iteration": 2.722790479660034
    },
    {
      "auxiliary_loss_clip": 0.0112387,
      "auxiliary_loss_mlp": 0.01022694,
      "balance_loss_clip": 1.04132926,
      "balance_loss_mlp": 1.01493943,
      "epoch": 0.9298382733120905,
      "flos": 24060436859520.0,
      "grad_norm": 4.89863364296573,
      "language_loss": 0.66631502,
      "learning_rate": 5.1339728329245155e-08,
      "loss": 0.68778062,
      "num_input_tokens_seen": 167159455,
      "step": 7733,
      "time_per_iteration": 2.77512264251709
    },
    {
      "auxiliary_loss_clip": 0.01111373,
      "auxiliary_loss_mlp": 0.01023462,
      "balance_loss_clip": 1.04188478,
      "balance_loss_mlp": 1.01508141,
      "epoch": 0.9299585162027295,
      "flos": 22127652910080.0,
      "grad_norm": 2.2992395158750396,
      "language_loss": 0.79012859,
      "learning_rate": 5.116451097578367e-08,
      "loss": 0.81147701,
      "num_input_tokens_seen": 167178495,
      "step": 7734,
      "time_per_iteration": 2.6931605339050293
    },
    {
      "auxiliary_loss_clip": 0.01123913,
      "auxiliary_loss_mlp": 0.0102235,
      "balance_loss_clip": 1.04491019,
      "balance_loss_mlp": 1.01523662,
      "epoch": 0.9300787590933686,
      "flos": 21471780522240.0,
      "grad_norm": 1.802265949465742,
      "language_loss": 0.74326456,
      "learning_rate": 5.0989589259115895e-08,
      "loss": 0.76472718,
      "num_input_tokens_seen": 167199380,
      "step": 7735,
      "time_per_iteration": 3.709296703338623
    },
    {
      "auxiliary_loss_clip": 0.011521,
      "auxiliary_loss_mlp": 0.01027067,
      "balance_loss_clip": 1.04465377,
      "balance_loss_mlp": 1.01843655,
      "epoch": 0.9301990019840077,
      "flos": 17779588594560.0,
      "grad_norm": 1.8702964309371761,
      "language_loss": 0.71516484,
      "learning_rate": 5.081496320577816e-08,
      "loss": 0.73695654,
      "num_input_tokens_seen": 167216500,
      "step": 7736,
      "time_per_iteration": 3.6901819705963135
    },
    {
      "auxiliary_loss_clip": 0.01017402,
      "auxiliary_loss_mlp": 0.01000753,
      "balance_loss_clip": 1.01119626,
      "balance_loss_mlp": 0.99966794,
      "epoch": 0.9303192448746468,
      "flos": 58896122307840.0,
      "grad_norm": 0.9616445604732344,
      "language_loss": 0.61145318,
      "learning_rate": 5.0640632842260835e-08,
      "loss": 0.63163471,
      "num_input_tokens_seen": 167276760,
      "step": 7737,
      "time_per_iteration": 3.379955768585205
    },
    {
      "auxiliary_loss_clip": 0.01087648,
      "auxiliary_loss_mlp": 0.01022481,
      "balance_loss_clip": 1.03894711,
      "balance_loss_mlp": 1.01500702,
      "epoch": 0.9304394877652858,
      "flos": 57663522172800.0,
      "grad_norm": 2.3568332519209445,
      "language_loss": 0.72878653,
      "learning_rate": 5.0466598195009426e-08,
      "loss": 0.74988782,
      "num_input_tokens_seen": 167303630,
      "step": 7738,
      "time_per_iteration": 4.057786464691162
    },
    {
      "auxiliary_loss_clip": 0.01134571,
      "auxiliary_loss_mlp": 0.01018871,
      "balance_loss_clip": 1.04287791,
      "balance_loss_mlp": 1.01156902,
      "epoch": 0.930559730655925,
      "flos": 20996143603200.0,
      "grad_norm": 1.9329614816032674,
      "language_loss": 0.70121866,
      "learning_rate": 5.0292859290425036e-08,
      "loss": 0.72275305,
      "num_input_tokens_seen": 167321500,
      "step": 7739,
      "time_per_iteration": 2.721388101577759
    },
    {
      "auxiliary_loss_clip": 0.01113051,
      "auxiliary_loss_mlp": 0.01021474,
      "balance_loss_clip": 1.04258752,
      "balance_loss_mlp": 1.01441061,
      "epoch": 0.9306799735465641,
      "flos": 23258264376960.0,
      "grad_norm": 1.8827846075858916,
      "language_loss": 0.77687883,
      "learning_rate": 5.011941615486348e-08,
      "loss": 0.79822409,
      "num_input_tokens_seen": 167340615,
      "step": 7740,
      "time_per_iteration": 2.76391339302063
    },
    {
      "auxiliary_loss_clip": 0.01097242,
      "auxiliary_loss_mlp": 0.01021307,
      "balance_loss_clip": 1.03595138,
      "balance_loss_mlp": 1.01405334,
      "epoch": 0.9308002164372031,
      "flos": 15231547560960.0,
      "grad_norm": 2.236938617037869,
      "language_loss": 0.84203362,
      "learning_rate": 4.994626881463659e-08,
      "loss": 0.86321914,
      "num_input_tokens_seen": 167356870,
      "step": 7741,
      "time_per_iteration": 2.753981828689575
    },
    {
      "auxiliary_loss_clip": 0.01152261,
      "auxiliary_loss_mlp": 0.00761396,
      "balance_loss_clip": 1.04603577,
      "balance_loss_mlp": 1.00045896,
      "epoch": 0.9309204593278423,
      "flos": 30847481539200.0,
      "grad_norm": 1.9943212598107674,
      "language_loss": 0.71563792,
      "learning_rate": 4.9773417296009814e-08,
      "loss": 0.73477447,
      "num_input_tokens_seen": 167378390,
      "step": 7742,
      "time_per_iteration": 2.686282157897949
    },
    {
      "auxiliary_loss_clip": 0.01124724,
      "auxiliary_loss_mlp": 0.01028789,
      "balance_loss_clip": 1.04830074,
      "balance_loss_mlp": 1.02035522,
      "epoch": 0.9310407022184813,
      "flos": 23037269950080.0,
      "grad_norm": 3.1567085850005276,
      "language_loss": 0.6597724,
      "learning_rate": 4.960086162520527e-08,
      "loss": 0.68130755,
      "num_input_tokens_seen": 167398480,
      "step": 7743,
      "time_per_iteration": 2.8449974060058594
    },
    {
      "auxiliary_loss_clip": 0.01118766,
      "auxiliary_loss_mlp": 0.01022824,
      "balance_loss_clip": 1.04125762,
      "balance_loss_mlp": 1.0146935,
      "epoch": 0.9311609451091204,
      "flos": 22127976132480.0,
      "grad_norm": 2.035971466304614,
      "language_loss": 0.8256188,
      "learning_rate": 4.942860182839936e-08,
      "loss": 0.84703469,
      "num_input_tokens_seen": 167416825,
      "step": 7744,
      "time_per_iteration": 2.7568726539611816
    },
    {
      "auxiliary_loss_clip": 0.01119894,
      "auxiliary_loss_mlp": 0.01028273,
      "balance_loss_clip": 1.03966308,
      "balance_loss_mlp": 1.0202384,
      "epoch": 0.9312811879997596,
      "flos": 21099206701440.0,
      "grad_norm": 1.9090804978373792,
      "language_loss": 0.79798925,
      "learning_rate": 4.925663793172341e-08,
      "loss": 0.81947088,
      "num_input_tokens_seen": 167434785,
      "step": 7745,
      "time_per_iteration": 3.640028715133667
    },
    {
      "auxiliary_loss_clip": 0.0102243,
      "auxiliary_loss_mlp": 0.01003667,
      "balance_loss_clip": 1.02142072,
      "balance_loss_mlp": 1.00264192,
      "epoch": 0.9314014308903986,
      "flos": 67148179096320.0,
      "grad_norm": 0.9100856296338732,
      "language_loss": 0.56466424,
      "learning_rate": 4.908496996126477e-08,
      "loss": 0.58492517,
      "num_input_tokens_seen": 167498245,
      "step": 7746,
      "time_per_iteration": 3.436305046081543
    },
    {
      "auxiliary_loss_clip": 0.01099531,
      "auxiliary_loss_mlp": 0.01027671,
      "balance_loss_clip": 1.04157305,
      "balance_loss_mlp": 1.01985049,
      "epoch": 0.9315216737810377,
      "flos": 22565583527040.0,
      "grad_norm": 2.022743988899823,
      "language_loss": 0.76538444,
      "learning_rate": 4.89135979430646e-08,
      "loss": 0.7866565,
      "num_input_tokens_seen": 167518290,
      "step": 7747,
      "time_per_iteration": 3.0191497802734375
    },
    {
      "auxiliary_loss_clip": 0.01153072,
      "auxiliary_loss_mlp": 0.01024217,
      "balance_loss_clip": 1.04728818,
      "balance_loss_mlp": 1.01654577,
      "epoch": 0.9316419166716768,
      "flos": 23984054588160.0,
      "grad_norm": 2.3051549807445,
      "language_loss": 0.85700983,
      "learning_rate": 4.874252190312078e-08,
      "loss": 0.87878269,
      "num_input_tokens_seen": 167538675,
      "step": 7748,
      "time_per_iteration": 2.6734838485717773
    },
    {
      "auxiliary_loss_clip": 0.01119539,
      "auxiliary_loss_mlp": 0.01022555,
      "balance_loss_clip": 1.03831863,
      "balance_loss_mlp": 1.01499748,
      "epoch": 0.9317621595623159,
      "flos": 30230464688640.0,
      "grad_norm": 1.9163523728708527,
      "language_loss": 0.6535821,
      "learning_rate": 4.857174186738477e-08,
      "loss": 0.67500305,
      "num_input_tokens_seen": 167562025,
      "step": 7749,
      "time_per_iteration": 2.8201920986175537
    },
    {
      "auxiliary_loss_clip": 0.01085747,
      "auxiliary_loss_mlp": 0.01024298,
      "balance_loss_clip": 1.04183817,
      "balance_loss_mlp": 1.01635575,
      "epoch": 0.931882402452955,
      "flos": 15742735966080.0,
      "grad_norm": 2.286803683247735,
      "language_loss": 0.72914815,
      "learning_rate": 4.840125786176408e-08,
      "loss": 0.75024861,
      "num_input_tokens_seen": 167578230,
      "step": 7750,
      "time_per_iteration": 2.762693166732788
    },
    {
      "auxiliary_loss_clip": 0.01103825,
      "auxiliary_loss_mlp": 0.01022556,
      "balance_loss_clip": 1.04493034,
      "balance_loss_mlp": 1.01571321,
      "epoch": 0.932002645343594,
      "flos": 28366521154560.0,
      "grad_norm": 1.8224129101547262,
      "language_loss": 0.77434236,
      "learning_rate": 4.823106991212067e-08,
      "loss": 0.79560614,
      "num_input_tokens_seen": 167597470,
      "step": 7751,
      "time_per_iteration": 2.8027503490448
    },
    {
      "auxiliary_loss_clip": 0.0110019,
      "auxiliary_loss_mlp": 0.01021267,
      "balance_loss_clip": 1.03909218,
      "balance_loss_mlp": 1.01442444,
      "epoch": 0.9321228882342332,
      "flos": 15341146934400.0,
      "grad_norm": 2.3711785523763313,
      "language_loss": 0.8354001,
      "learning_rate": 4.806117804427212e-08,
      "loss": 0.85661465,
      "num_input_tokens_seen": 167615405,
      "step": 7752,
      "time_per_iteration": 2.722280502319336
    },
    {
      "auxiliary_loss_clip": 0.01108674,
      "auxiliary_loss_mlp": 0.00762028,
      "balance_loss_clip": 1.04105401,
      "balance_loss_mlp": 1.00041711,
      "epoch": 0.9322431311248722,
      "flos": 17895365107200.0,
      "grad_norm": 2.086362564063483,
      "language_loss": 0.6466378,
      "learning_rate": 4.7891582283990926e-08,
      "loss": 0.66534483,
      "num_input_tokens_seen": 167634130,
      "step": 7753,
      "time_per_iteration": 2.775804281234741
    },
    {
      "auxiliary_loss_clip": 0.01109975,
      "auxiliary_loss_mlp": 0.01023833,
      "balance_loss_clip": 1.043028,
      "balance_loss_mlp": 1.01640606,
      "epoch": 0.9323633740155113,
      "flos": 24169713010560.0,
      "grad_norm": 2.9650702864535905,
      "language_loss": 0.72538328,
      "learning_rate": 4.772228265700473e-08,
      "loss": 0.74672139,
      "num_input_tokens_seen": 167654990,
      "step": 7754,
      "time_per_iteration": 2.7746353149414062
    },
    {
      "auxiliary_loss_clip": 0.01117354,
      "auxiliary_loss_mlp": 0.01021964,
      "balance_loss_clip": 1.04126847,
      "balance_loss_mlp": 1.01432562,
      "epoch": 0.9324836169061504,
      "flos": 15043482927360.0,
      "grad_norm": 4.9333235389140695,
      "language_loss": 0.76225865,
      "learning_rate": 4.75532791889961e-08,
      "loss": 0.78365183,
      "num_input_tokens_seen": 167671690,
      "step": 7755,
      "time_per_iteration": 2.6890950202941895
    },
    {
      "auxiliary_loss_clip": 0.0112788,
      "auxiliary_loss_mlp": 0.01024836,
      "balance_loss_clip": 1.04305601,
      "balance_loss_mlp": 1.01751637,
      "epoch": 0.9326038597967895,
      "flos": 18624890332800.0,
      "grad_norm": 1.976241239434444,
      "language_loss": 0.65681684,
      "learning_rate": 4.738457190560252e-08,
      "loss": 0.67834401,
      "num_input_tokens_seen": 167690800,
      "step": 7756,
      "time_per_iteration": 2.820676326751709
    },
    {
      "auxiliary_loss_clip": 0.01125365,
      "auxiliary_loss_mlp": 0.01027252,
      "balance_loss_clip": 1.04227281,
      "balance_loss_mlp": 1.01970029,
      "epoch": 0.9327241026874286,
      "flos": 18952646958720.0,
      "grad_norm": 2.086041491547538,
      "language_loss": 0.79007798,
      "learning_rate": 4.721616083241664e-08,
      "loss": 0.81160414,
      "num_input_tokens_seen": 167709055,
      "step": 7757,
      "time_per_iteration": 2.755013942718506
    },
    {
      "auxiliary_loss_clip": 0.01119007,
      "auxiliary_loss_mlp": 0.01028165,
      "balance_loss_clip": 1.04296744,
      "balance_loss_mlp": 1.01990986,
      "epoch": 0.9328443455780677,
      "flos": 29570282668800.0,
      "grad_norm": 1.8066353921632752,
      "language_loss": 0.7782321,
      "learning_rate": 4.7048045994986684e-08,
      "loss": 0.79970378,
      "num_input_tokens_seen": 167729915,
      "step": 7758,
      "time_per_iteration": 2.88997483253479
    },
    {
      "auxiliary_loss_clip": 0.01157181,
      "auxiliary_loss_mlp": 0.01025408,
      "balance_loss_clip": 1.05021632,
      "balance_loss_mlp": 1.01669383,
      "epoch": 0.9329645884687068,
      "flos": 30081722469120.0,
      "grad_norm": 2.047876638241002,
      "language_loss": 0.91095638,
      "learning_rate": 4.688022741881559e-08,
      "loss": 0.93278229,
      "num_input_tokens_seen": 167750440,
      "step": 7759,
      "time_per_iteration": 2.6903605461120605
    },
    {
      "auxiliary_loss_clip": 0.01118172,
      "auxiliary_loss_mlp": 0.0102443,
      "balance_loss_clip": 1.04153275,
      "balance_loss_mlp": 1.01733661,
      "epoch": 0.9330848313593458,
      "flos": 21867982513920.0,
      "grad_norm": 1.9589825600085278,
      "language_loss": 0.75136143,
      "learning_rate": 4.671270512936076e-08,
      "loss": 0.77278739,
      "num_input_tokens_seen": 167769600,
      "step": 7760,
      "time_per_iteration": 2.7277445793151855
    },
    {
      "auxiliary_loss_clip": 0.01085858,
      "auxiliary_loss_mlp": 0.01024543,
      "balance_loss_clip": 1.03831518,
      "balance_loss_mlp": 1.01773858,
      "epoch": 0.933205074249985,
      "flos": 22127221946880.0,
      "grad_norm": 2.091554742868412,
      "language_loss": 0.82754117,
      "learning_rate": 4.6545479152035884e-08,
      "loss": 0.84864515,
      "num_input_tokens_seen": 167788770,
      "step": 7761,
      "time_per_iteration": 3.7223141193389893
    },
    {
      "auxiliary_loss_clip": 0.01102664,
      "auxiliary_loss_mlp": 0.00761827,
      "balance_loss_clip": 1.03929317,
      "balance_loss_mlp": 1.00046802,
      "epoch": 0.9333253171406241,
      "flos": 15341254675200.0,
      "grad_norm": 2.2143979977116977,
      "language_loss": 0.76356816,
      "learning_rate": 4.637854951220821e-08,
      "loss": 0.78221309,
      "num_input_tokens_seen": 167805555,
      "step": 7762,
      "time_per_iteration": 3.7365520000457764
    },
    {
      "auxiliary_loss_clip": 0.01089891,
      "auxiliary_loss_mlp": 0.01028093,
      "balance_loss_clip": 1.04294515,
      "balance_loss_mlp": 1.02039218,
      "epoch": 0.9334455600312631,
      "flos": 15706142985600.0,
      "grad_norm": 2.068328603185799,
      "language_loss": 0.74876332,
      "learning_rate": 4.621191623520171e-08,
      "loss": 0.76994312,
      "num_input_tokens_seen": 167823985,
      "step": 7763,
      "time_per_iteration": 2.9169082641601562
    },
    {
      "auxiliary_loss_clip": 0.01120219,
      "auxiliary_loss_mlp": 0.01025448,
      "balance_loss_clip": 1.04108715,
      "balance_loss_mlp": 1.0179677,
      "epoch": 0.9335658029219023,
      "flos": 22163563532160.0,
      "grad_norm": 2.3543529550430633,
      "language_loss": 0.8465817,
      "learning_rate": 4.604557934629372e-08,
      "loss": 0.8680383,
      "num_input_tokens_seen": 167843060,
      "step": 7764,
      "time_per_iteration": 3.633563280105591
    },
    {
      "auxiliary_loss_clip": 0.01155674,
      "auxiliary_loss_mlp": 0.00761453,
      "balance_loss_clip": 1.04478598,
      "balance_loss_mlp": 1.00036907,
      "epoch": 0.9336860458125413,
      "flos": 20266833859200.0,
      "grad_norm": 2.110893382530769,
      "language_loss": 0.8040666,
      "learning_rate": 4.587953887071805e-08,
      "loss": 0.8232379,
      "num_input_tokens_seen": 167862880,
      "step": 7765,
      "time_per_iteration": 2.6328272819519043
    },
    {
      "auxiliary_loss_clip": 0.01154192,
      "auxiliary_loss_mlp": 0.00762006,
      "balance_loss_clip": 1.04485798,
      "balance_loss_mlp": 1.00044119,
      "epoch": 0.9338062887031804,
      "flos": 20919689504640.0,
      "grad_norm": 1.8782537604580092,
      "language_loss": 0.85657752,
      "learning_rate": 4.5713794833662554e-08,
      "loss": 0.87573951,
      "num_input_tokens_seen": 167882095,
      "step": 7766,
      "time_per_iteration": 2.63016939163208
    },
    {
      "auxiliary_loss_clip": 0.01119122,
      "auxiliary_loss_mlp": 0.01033013,
      "balance_loss_clip": 1.04287553,
      "balance_loss_mlp": 1.02513325,
      "epoch": 0.9339265315938196,
      "flos": 23221635482880.0,
      "grad_norm": 2.0994962012932086,
      "language_loss": 0.63785398,
      "learning_rate": 4.5548347260270236e-08,
      "loss": 0.65937537,
      "num_input_tokens_seen": 167901385,
      "step": 7767,
      "time_per_iteration": 2.689361572265625
    },
    {
      "auxiliary_loss_clip": 0.01124224,
      "auxiliary_loss_mlp": 0.01023563,
      "balance_loss_clip": 1.04080987,
      "balance_loss_mlp": 1.0163393,
      "epoch": 0.9340467744844586,
      "flos": 22820261932800.0,
      "grad_norm": 1.5816086576072033,
      "language_loss": 0.69589478,
      "learning_rate": 4.538319617564012e-08,
      "loss": 0.71737266,
      "num_input_tokens_seen": 167920405,
      "step": 7768,
      "time_per_iteration": 2.7056655883789062
    },
    {
      "auxiliary_loss_clip": 0.01141755,
      "auxiliary_loss_mlp": 0.01025925,
      "balance_loss_clip": 1.04271436,
      "balance_loss_mlp": 1.01870108,
      "epoch": 0.9341670173750977,
      "flos": 23660428026240.0,
      "grad_norm": 2.1749851808484366,
      "language_loss": 0.75016379,
      "learning_rate": 4.521834160482485e-08,
      "loss": 0.77184063,
      "num_input_tokens_seen": 167939145,
      "step": 7769,
      "time_per_iteration": 2.741490602493286
    },
    {
      "auxiliary_loss_clip": 0.0112077,
      "auxiliary_loss_mlp": 0.01020559,
      "balance_loss_clip": 1.04173255,
      "balance_loss_mlp": 1.01315641,
      "epoch": 0.9342872602657368,
      "flos": 24824256595200.0,
      "grad_norm": 1.7449438378217519,
      "language_loss": 0.82492453,
      "learning_rate": 4.5053783572832846e-08,
      "loss": 0.84633785,
      "num_input_tokens_seen": 167959325,
      "step": 7770,
      "time_per_iteration": 3.843752861022949
    },
    {
      "auxiliary_loss_clip": 0.0111606,
      "auxiliary_loss_mlp": 0.0102888,
      "balance_loss_clip": 1.04249024,
      "balance_loss_mlp": 1.02147424,
      "epoch": 0.9344075031563759,
      "flos": 25771831332480.0,
      "grad_norm": 1.6808008575332805,
      "language_loss": 0.76614499,
      "learning_rate": 4.488952210462771e-08,
      "loss": 0.78759444,
      "num_input_tokens_seen": 167979530,
      "step": 7771,
      "time_per_iteration": 2.8078372478485107
    },
    {
      "auxiliary_loss_clip": 0.01138202,
      "auxiliary_loss_mlp": 0.0102837,
      "balance_loss_clip": 1.04527378,
      "balance_loss_mlp": 1.02083635,
      "epoch": 0.9345277460470149,
      "flos": 25551303782400.0,
      "grad_norm": 2.50791963663353,
      "language_loss": 0.86348867,
      "learning_rate": 4.4725557225127495e-08,
      "loss": 0.88515443,
      "num_input_tokens_seen": 167997870,
      "step": 7772,
      "time_per_iteration": 2.894834041595459
    },
    {
      "auxiliary_loss_clip": 0.01099201,
      "auxiliary_loss_mlp": 0.01020855,
      "balance_loss_clip": 1.0381453,
      "balance_loss_mlp": 1.01369929,
      "epoch": 0.9346479889376541,
      "flos": 34313112432000.0,
      "grad_norm": 1.7055371396680807,
      "language_loss": 0.79470098,
      "learning_rate": 4.456188895920565e-08,
      "loss": 0.81590152,
      "num_input_tokens_seen": 168019625,
      "step": 7773,
      "time_per_iteration": 3.0220816135406494
    },
    {
      "auxiliary_loss_clip": 0.01128031,
      "auxiliary_loss_mlp": 0.01025357,
      "balance_loss_clip": 1.04448915,
      "balance_loss_mlp": 1.01821649,
      "epoch": 0.9347682318282932,
      "flos": 19093739581440.0,
      "grad_norm": 2.126577860340411,
      "language_loss": 0.85507834,
      "learning_rate": 4.439851733169031e-08,
      "loss": 0.87661219,
      "num_input_tokens_seen": 168037415,
      "step": 7774,
      "time_per_iteration": 2.6953072547912598
    },
    {
      "auxiliary_loss_clip": 0.01135166,
      "auxiliary_loss_mlp": 0.01022803,
      "balance_loss_clip": 1.04092538,
      "balance_loss_mlp": 1.01565611,
      "epoch": 0.9348884747189322,
      "flos": 26249587153920.0,
      "grad_norm": 2.4701690964201153,
      "language_loss": 0.6943785,
      "learning_rate": 4.4235442367365204e-08,
      "loss": 0.71595812,
      "num_input_tokens_seen": 168057725,
      "step": 7775,
      "time_per_iteration": 2.780118942260742
    },
    {
      "auxiliary_loss_clip": 0.01078001,
      "auxiliary_loss_mlp": 0.01028221,
      "balance_loss_clip": 1.03695583,
      "balance_loss_mlp": 1.0208149,
      "epoch": 0.9350087176095714,
      "flos": 18333080242560.0,
      "grad_norm": 2.1860375741984357,
      "language_loss": 0.79688346,
      "learning_rate": 4.4072664090968545e-08,
      "loss": 0.81794566,
      "num_input_tokens_seen": 168076110,
      "step": 7776,
      "time_per_iteration": 2.9078311920166016
    },
    {
      "auxiliary_loss_clip": 0.01127592,
      "auxiliary_loss_mlp": 0.0102426,
      "balance_loss_clip": 1.04227304,
      "balance_loss_mlp": 1.01691937,
      "epoch": 0.9351289605002104,
      "flos": 19318253541120.0,
      "grad_norm": 1.956092070673333,
      "language_loss": 0.84599149,
      "learning_rate": 4.391018252719347e-08,
      "loss": 0.86751002,
      "num_input_tokens_seen": 168095905,
      "step": 7777,
      "time_per_iteration": 2.966299057006836
    },
    {
      "auxiliary_loss_clip": 0.01119434,
      "auxiliary_loss_mlp": 0.01024359,
      "balance_loss_clip": 1.04432726,
      "balance_loss_mlp": 1.0168488,
      "epoch": 0.9352492033908495,
      "flos": 18799990156800.0,
      "grad_norm": 2.120989194861513,
      "language_loss": 0.69061852,
      "learning_rate": 4.374799770068849e-08,
      "loss": 0.7120564,
      "num_input_tokens_seen": 168112580,
      "step": 7778,
      "time_per_iteration": 2.759045124053955
    },
    {
      "auxiliary_loss_clip": 0.01126228,
      "auxiliary_loss_mlp": 0.01022862,
      "balance_loss_clip": 1.03741217,
      "balance_loss_mlp": 1.01556897,
      "epoch": 0.9353694462814887,
      "flos": 29530134241920.0,
      "grad_norm": 1.9581264452930467,
      "language_loss": 0.75090092,
      "learning_rate": 4.358610963605658e-08,
      "loss": 0.77239186,
      "num_input_tokens_seen": 168133030,
      "step": 7779,
      "time_per_iteration": 2.7625319957733154
    },
    {
      "auxiliary_loss_clip": 0.01099668,
      "auxiliary_loss_mlp": 0.01025556,
      "balance_loss_clip": 1.03740025,
      "balance_loss_mlp": 1.01744413,
      "epoch": 0.9354896891721277,
      "flos": 30665450390400.0,
      "grad_norm": 2.4929897557350853,
      "language_loss": 0.68978179,
      "learning_rate": 4.342451835785677e-08,
      "loss": 0.71103406,
      "num_input_tokens_seen": 168153940,
      "step": 7780,
      "time_per_iteration": 2.88458514213562
    },
    {
      "auxiliary_loss_clip": 0.01151656,
      "auxiliary_loss_mlp": 0.01026592,
      "balance_loss_clip": 1.04499269,
      "balance_loss_mlp": 1.01914155,
      "epoch": 0.9356099320627668,
      "flos": 19463907191040.0,
      "grad_norm": 1.8112025095356163,
      "language_loss": 0.75209665,
      "learning_rate": 4.3263223890601665e-08,
      "loss": 0.77387911,
      "num_input_tokens_seen": 168172650,
      "step": 7781,
      "time_per_iteration": 2.6850788593292236
    },
    {
      "auxiliary_loss_clip": 0.01079899,
      "auxiliary_loss_mlp": 0.00761457,
      "balance_loss_clip": 1.03677869,
      "balance_loss_mlp": 1.00046849,
      "epoch": 0.9357301749534058,
      "flos": 19098156954240.0,
      "grad_norm": 2.1497211074765845,
      "language_loss": 0.79300857,
      "learning_rate": 4.31022262587597e-08,
      "loss": 0.81142211,
      "num_input_tokens_seen": 168191325,
      "step": 7782,
      "time_per_iteration": 2.8972485065460205
    },
    {
      "auxiliary_loss_clip": 0.01119837,
      "auxiliary_loss_mlp": 0.0102595,
      "balance_loss_clip": 1.04275262,
      "balance_loss_mlp": 1.01759374,
      "epoch": 0.935850417844045,
      "flos": 23550361776000.0,
      "grad_norm": 2.517698419812339,
      "language_loss": 0.66234273,
      "learning_rate": 4.2941525486754225e-08,
      "loss": 0.68380064,
      "num_input_tokens_seen": 168211645,
      "step": 7783,
      "time_per_iteration": 2.9004526138305664
    },
    {
      "auxiliary_loss_clip": 0.01109835,
      "auxiliary_loss_mlp": 0.01025275,
      "balance_loss_clip": 1.04127705,
      "balance_loss_mlp": 1.01813471,
      "epoch": 0.935970660734684,
      "flos": 18588333265920.0,
      "grad_norm": 1.826009048637279,
      "language_loss": 0.79519188,
      "learning_rate": 4.278112159896286e-08,
      "loss": 0.81654298,
      "num_input_tokens_seen": 168229485,
      "step": 7784,
      "time_per_iteration": 2.805217742919922
    },
    {
      "auxiliary_loss_clip": 0.011042,
      "auxiliary_loss_mlp": 0.0102058,
      "balance_loss_clip": 1.03724074,
      "balance_loss_mlp": 1.01326621,
      "epoch": 0.9360909036253231,
      "flos": 20631255292800.0,
      "grad_norm": 1.7940245421537477,
      "language_loss": 0.67764598,
      "learning_rate": 4.2621014619719896e-08,
      "loss": 0.69889379,
      "num_input_tokens_seen": 168247250,
      "step": 7785,
      "time_per_iteration": 2.7823116779327393
    },
    {
      "auxiliary_loss_clip": 0.0103064,
      "auxiliary_loss_mlp": 0.01000931,
      "balance_loss_clip": 1.01030302,
      "balance_loss_mlp": 0.99982864,
      "epoch": 0.9362111465159623,
      "flos": 61791421052160.0,
      "grad_norm": 0.7507005130296366,
      "language_loss": 0.58550739,
      "learning_rate": 4.246120457331215e-08,
      "loss": 0.60582304,
      "num_input_tokens_seen": 168309425,
      "step": 7786,
      "time_per_iteration": 3.3575146198272705
    },
    {
      "auxiliary_loss_clip": 0.01105743,
      "auxiliary_loss_mlp": 0.01023916,
      "balance_loss_clip": 1.04490125,
      "balance_loss_mlp": 1.015944,
      "epoch": 0.9363313894066013,
      "flos": 24170395368960.0,
      "grad_norm": 2.265536993673555,
      "language_loss": 0.72073394,
      "learning_rate": 4.2301691483983325e-08,
      "loss": 0.7420305,
      "num_input_tokens_seen": 168329545,
      "step": 7787,
      "time_per_iteration": 4.1285223960876465
    },
    {
      "auxiliary_loss_clip": 0.01169975,
      "auxiliary_loss_mlp": 0.0102366,
      "balance_loss_clip": 1.04730344,
      "balance_loss_mlp": 1.01646018,
      "epoch": 0.9364516322972404,
      "flos": 20120354196480.0,
      "grad_norm": 2.014230001042915,
      "language_loss": 0.75771618,
      "learning_rate": 4.214247537593163e-08,
      "loss": 0.77965248,
      "num_input_tokens_seen": 168348795,
      "step": 7788,
      "time_per_iteration": 3.7448463439941406
    },
    {
      "auxiliary_loss_clip": 0.0111902,
      "auxiliary_loss_mlp": 0.01024892,
      "balance_loss_clip": 1.04342699,
      "balance_loss_mlp": 1.01747108,
      "epoch": 0.9365718751878795,
      "flos": 20703758895360.0,
      "grad_norm": 1.896939685147377,
      "language_loss": 0.80372012,
      "learning_rate": 4.1983556273309293e-08,
      "loss": 0.82515925,
      "num_input_tokens_seen": 168367545,
      "step": 7789,
      "time_per_iteration": 2.732887029647827
    },
    {
      "auxiliary_loss_clip": 0.01142531,
      "auxiliary_loss_mlp": 0.01025138,
      "balance_loss_clip": 1.04807973,
      "balance_loss_mlp": 1.01719844,
      "epoch": 0.9366921180785186,
      "flos": 18655270260480.0,
      "grad_norm": 2.345562211584074,
      "language_loss": 0.69229317,
      "learning_rate": 4.182493420022526e-08,
      "loss": 0.71396983,
      "num_input_tokens_seen": 168383215,
      "step": 7790,
      "time_per_iteration": 3.5061049461364746
    },
    {
      "auxiliary_loss_clip": 0.01112729,
      "auxiliary_loss_mlp": 0.00762062,
      "balance_loss_clip": 1.04127252,
      "balance_loss_mlp": 1.0003444,
      "epoch": 0.9368123609691577,
      "flos": 25774955815680.0,
      "grad_norm": 1.6189939299080844,
      "language_loss": 0.78547347,
      "learning_rate": 4.166660918074139e-08,
      "loss": 0.80422139,
      "num_input_tokens_seen": 168403120,
      "step": 7791,
      "time_per_iteration": 2.7537307739257812
    },
    {
      "auxiliary_loss_clip": 0.01119608,
      "auxiliary_loss_mlp": 0.01025791,
      "balance_loss_clip": 1.04182768,
      "balance_loss_mlp": 1.01854885,
      "epoch": 0.9369326038597968,
      "flos": 25553386771200.0,
      "grad_norm": 1.6064816824509212,
      "language_loss": 0.7350499,
      "learning_rate": 4.15085812388758e-08,
      "loss": 0.75650394,
      "num_input_tokens_seen": 168425340,
      "step": 7792,
      "time_per_iteration": 2.7436091899871826
    },
    {
      "auxiliary_loss_clip": 0.01099518,
      "auxiliary_loss_mlp": 0.01025703,
      "balance_loss_clip": 1.03860736,
      "balance_loss_mlp": 1.0173223,
      "epoch": 0.9370528467504359,
      "flos": 23220019370880.0,
      "grad_norm": 1.8678577681557862,
      "language_loss": 0.78405023,
      "learning_rate": 4.135085039860153e-08,
      "loss": 0.80530238,
      "num_input_tokens_seen": 168444740,
      "step": 7793,
      "time_per_iteration": 2.8500163555145264
    },
    {
      "auxiliary_loss_clip": 0.01144497,
      "auxiliary_loss_mlp": 0.00762549,
      "balance_loss_clip": 1.04438579,
      "balance_loss_mlp": 1.00044274,
      "epoch": 0.9371730896410749,
      "flos": 24967468120320.0,
      "grad_norm": 2.1358118794390344,
      "language_loss": 0.78999245,
      "learning_rate": 4.1193416683845906e-08,
      "loss": 0.80906284,
      "num_input_tokens_seen": 168463670,
      "step": 7794,
      "time_per_iteration": 2.995117664337158
    },
    {
      "auxiliary_loss_clip": 0.01132989,
      "auxiliary_loss_mlp": 0.00761696,
      "balance_loss_clip": 1.04280579,
      "balance_loss_mlp": 1.000489,
      "epoch": 0.9372933325317141,
      "flos": 15553091134080.0,
      "grad_norm": 2.401849999160431,
      "language_loss": 0.83672249,
      "learning_rate": 4.103628011849136e-08,
      "loss": 0.85566938,
      "num_input_tokens_seen": 168479030,
      "step": 7795,
      "time_per_iteration": 3.667625665664673
    },
    {
      "auxiliary_loss_clip": 0.0109776,
      "auxiliary_loss_mlp": 0.01024512,
      "balance_loss_clip": 1.04081082,
      "balance_loss_mlp": 1.01650679,
      "epoch": 0.9374135754223532,
      "flos": 21871861182720.0,
      "grad_norm": 1.9226924553914213,
      "language_loss": 0.75926769,
      "learning_rate": 4.0879440726375506e-08,
      "loss": 0.7804904,
      "num_input_tokens_seen": 168496815,
      "step": 7796,
      "time_per_iteration": 2.7738559246063232
    },
    {
      "auxiliary_loss_clip": 0.01115963,
      "auxiliary_loss_mlp": 0.01025363,
      "balance_loss_clip": 1.03827763,
      "balance_loss_mlp": 1.01757002,
      "epoch": 0.9375338183129922,
      "flos": 22631048064000.0,
      "grad_norm": 15.506654187437798,
      "language_loss": 0.56665123,
      "learning_rate": 4.0722898531291074e-08,
      "loss": 0.58806449,
      "num_input_tokens_seen": 168514055,
      "step": 7797,
      "time_per_iteration": 2.819617509841919
    },
    {
      "auxiliary_loss_clip": 0.01145674,
      "auxiliary_loss_mlp": 0.01022148,
      "balance_loss_clip": 1.0429461,
      "balance_loss_mlp": 1.01463759,
      "epoch": 0.9376540612036314,
      "flos": 26104292640000.0,
      "grad_norm": 2.0985397797421976,
      "language_loss": 0.76736474,
      "learning_rate": 4.0566653556985295e-08,
      "loss": 0.78904295,
      "num_input_tokens_seen": 168534600,
      "step": 7798,
      "time_per_iteration": 2.7909793853759766
    },
    {
      "auxiliary_loss_clip": 0.01158977,
      "auxiliary_loss_mlp": 0.00762644,
      "balance_loss_clip": 1.04808068,
      "balance_loss_mlp": 1.00049686,
      "epoch": 0.9377743040942704,
      "flos": 19717580016000.0,
      "grad_norm": 2.371757055500727,
      "language_loss": 0.81248248,
      "learning_rate": 4.0410705827159886e-08,
      "loss": 0.83169866,
      "num_input_tokens_seen": 168551895,
      "step": 7799,
      "time_per_iteration": 2.6599643230438232
    },
    {
      "auxiliary_loss_clip": 0.0111921,
      "auxiliary_loss_mlp": 0.01021462,
      "balance_loss_clip": 1.04086685,
      "balance_loss_mlp": 1.01400805,
      "epoch": 0.9378945469849095,
      "flos": 15267530010240.0,
      "grad_norm": 2.7186627865761914,
      "language_loss": 0.71242476,
      "learning_rate": 4.0255055365472356e-08,
      "loss": 0.73383152,
      "num_input_tokens_seen": 168569990,
      "step": 7800,
      "time_per_iteration": 2.8040101528167725
    },
    {
      "auxiliary_loss_clip": 0.01137254,
      "auxiliary_loss_mlp": 0.01023589,
      "balance_loss_clip": 1.04269505,
      "balance_loss_mlp": 1.01596832,
      "epoch": 0.9380147898755486,
      "flos": 20591394174720.0,
      "grad_norm": 2.774021278963581,
      "language_loss": 0.74712467,
      "learning_rate": 4.009970219553471e-08,
      "loss": 0.76873308,
      "num_input_tokens_seen": 168586940,
      "step": 7801,
      "time_per_iteration": 2.6798250675201416
    },
    {
      "auxiliary_loss_clip": 0.01145555,
      "auxiliary_loss_mlp": 0.00763101,
      "balance_loss_clip": 1.04461432,
      "balance_loss_mlp": 1.00044584,
      "epoch": 0.9381350327661877,
      "flos": 26281116316800.0,
      "grad_norm": 2.8469201240854676,
      "language_loss": 0.76370424,
      "learning_rate": 3.99446463409141e-08,
      "loss": 0.78279078,
      "num_input_tokens_seen": 168604795,
      "step": 7802,
      "time_per_iteration": 2.665217399597168
    },
    {
      "auxiliary_loss_clip": 0.01146029,
      "auxiliary_loss_mlp": 0.0102448,
      "balance_loss_clip": 1.04315472,
      "balance_loss_mlp": 1.01638556,
      "epoch": 0.9382552756568268,
      "flos": 23586344225280.0,
      "grad_norm": 2.4823986302059,
      "language_loss": 0.69232714,
      "learning_rate": 3.978988782513215e-08,
      "loss": 0.71403217,
      "num_input_tokens_seen": 168622290,
      "step": 7803,
      "time_per_iteration": 2.7312355041503906
    },
    {
      "auxiliary_loss_clip": 0.01127546,
      "auxiliary_loss_mlp": 0.01026061,
      "balance_loss_clip": 1.04305995,
      "balance_loss_mlp": 1.01847959,
      "epoch": 0.9383755185474659,
      "flos": 28438809275520.0,
      "grad_norm": 1.9522289640980781,
      "language_loss": 0.76644778,
      "learning_rate": 3.963542667166586e-08,
      "loss": 0.78798383,
      "num_input_tokens_seen": 168642395,
      "step": 7804,
      "time_per_iteration": 2.8198201656341553
    },
    {
      "auxiliary_loss_clip": 0.01143363,
      "auxiliary_loss_mlp": 0.01028701,
      "balance_loss_clip": 1.04797781,
      "balance_loss_mlp": 1.02089632,
      "epoch": 0.938495761438105,
      "flos": 20449583280000.0,
      "grad_norm": 1.6661655285810135,
      "language_loss": 0.68457377,
      "learning_rate": 3.9481262903946486e-08,
      "loss": 0.70629442,
      "num_input_tokens_seen": 168661840,
      "step": 7805,
      "time_per_iteration": 2.594870090484619
    },
    {
      "auxiliary_loss_clip": 0.0104614,
      "auxiliary_loss_mlp": 0.01001093,
      "balance_loss_clip": 1.00978231,
      "balance_loss_mlp": 1.00003171,
      "epoch": 0.938616004328744,
      "flos": 69302711658240.0,
      "grad_norm": 0.7749391303161651,
      "language_loss": 0.54474807,
      "learning_rate": 3.932739654536066e-08,
      "loss": 0.56522042,
      "num_input_tokens_seen": 168724540,
      "step": 7806,
      "time_per_iteration": 3.2557504177093506
    },
    {
      "auxiliary_loss_clip": 0.01074817,
      "auxiliary_loss_mlp": 0.01026254,
      "balance_loss_clip": 1.03861356,
      "balance_loss_mlp": 1.01867199,
      "epoch": 0.9387362472193832,
      "flos": 18911636605440.0,
      "grad_norm": 3.529940010866728,
      "language_loss": 0.74063814,
      "learning_rate": 3.917382761925014e-08,
      "loss": 0.76164889,
      "num_input_tokens_seen": 168740375,
      "step": 7807,
      "time_per_iteration": 2.839735269546509
    },
    {
      "auxiliary_loss_clip": 0.01076783,
      "auxiliary_loss_mlp": 0.0102932,
      "balance_loss_clip": 1.03773236,
      "balance_loss_mlp": 1.02147031,
      "epoch": 0.9388564901100223,
      "flos": 26501967089280.0,
      "grad_norm": 1.8456353229642588,
      "language_loss": 0.79423881,
      "learning_rate": 3.9020556148910754e-08,
      "loss": 0.81529987,
      "num_input_tokens_seen": 168759730,
      "step": 7808,
      "time_per_iteration": 2.813051223754883
    },
    {
      "auxiliary_loss_clip": 0.01033671,
      "auxiliary_loss_mlp": 0.01001132,
      "balance_loss_clip": 1.00935984,
      "balance_loss_mlp": 1.00017226,
      "epoch": 0.9389767330006613,
      "flos": 58941083157120.0,
      "grad_norm": 0.7123832379275747,
      "language_loss": 0.56683081,
      "learning_rate": 3.8867582157593895e-08,
      "loss": 0.58717883,
      "num_input_tokens_seen": 168813935,
      "step": 7809,
      "time_per_iteration": 3.1627204418182373
    },
    {
      "auxiliary_loss_clip": 0.01144523,
      "auxiliary_loss_mlp": 0.0102594,
      "balance_loss_clip": 1.04457593,
      "balance_loss_mlp": 1.01865911,
      "epoch": 0.9390969758913005,
      "flos": 31102554994560.0,
      "grad_norm": 1.5879896224278294,
      "language_loss": 0.76647598,
      "learning_rate": 3.871490566850544e-08,
      "loss": 0.78818059,
      "num_input_tokens_seen": 168838145,
      "step": 7810,
      "time_per_iteration": 2.7666118144989014
    },
    {
      "auxiliary_loss_clip": 0.01114322,
      "auxiliary_loss_mlp": 0.00761482,
      "balance_loss_clip": 1.0380137,
      "balance_loss_mlp": 1.00049281,
      "epoch": 0.9392172187819395,
      "flos": 22419391173120.0,
      "grad_norm": 1.8118192113381348,
      "language_loss": 0.70430803,
      "learning_rate": 3.856252670480642e-08,
      "loss": 0.72306609,
      "num_input_tokens_seen": 168856805,
      "step": 7811,
      "time_per_iteration": 2.728048086166382
    },
    {
      "auxiliary_loss_clip": 0.011212,
      "auxiliary_loss_mlp": 0.01024861,
      "balance_loss_clip": 1.0401113,
      "balance_loss_mlp": 1.01772606,
      "epoch": 0.9393374616725786,
      "flos": 19719483436800.0,
      "grad_norm": 2.12200337734317,
      "language_loss": 0.81265473,
      "learning_rate": 3.841044528961279e-08,
      "loss": 0.83411539,
      "num_input_tokens_seen": 168874600,
      "step": 7812,
      "time_per_iteration": 2.7464382648468018
    },
    {
      "auxiliary_loss_clip": 0.01077798,
      "auxiliary_loss_mlp": 0.01021966,
      "balance_loss_clip": 1.03347611,
      "balance_loss_mlp": 1.01378846,
      "epoch": 0.9394577045632178,
      "flos": 24170215800960.0,
      "grad_norm": 1.695453582962252,
      "language_loss": 0.79031098,
      "learning_rate": 3.825866144599477e-08,
      "loss": 0.81130856,
      "num_input_tokens_seen": 168893655,
      "step": 7813,
      "time_per_iteration": 3.7177155017852783
    },
    {
      "auxiliary_loss_clip": 0.01096366,
      "auxiliary_loss_mlp": 0.01024598,
      "balance_loss_clip": 1.03589201,
      "balance_loss_mlp": 1.01703358,
      "epoch": 0.9395779474538568,
      "flos": 19023929498880.0,
      "grad_norm": 2.298199652466566,
      "language_loss": 0.75534225,
      "learning_rate": 3.8107175196978145e-08,
      "loss": 0.7765519,
      "num_input_tokens_seen": 168909960,
      "step": 7814,
      "time_per_iteration": 3.703235387802124
    },
    {
      "auxiliary_loss_clip": 0.01119518,
      "auxiliary_loss_mlp": 0.01024699,
      "balance_loss_clip": 1.04487371,
      "balance_loss_mlp": 1.01737309,
      "epoch": 0.9396981903444959,
      "flos": 14319129260160.0,
      "grad_norm": 2.3335813961647336,
      "language_loss": 0.76864856,
      "learning_rate": 3.7955986565542996e-08,
      "loss": 0.79009068,
      "num_input_tokens_seen": 168928040,
      "step": 7815,
      "time_per_iteration": 2.687925100326538
    },
    {
      "auxiliary_loss_clip": 0.01150636,
      "auxiliary_loss_mlp": 0.0102831,
      "balance_loss_clip": 1.04474521,
      "balance_loss_mlp": 1.02040064,
      "epoch": 0.9398184332351349,
      "flos": 34787564202240.0,
      "grad_norm": 2.0296553387077565,
      "language_loss": 0.68666685,
      "learning_rate": 3.780509557462497e-08,
      "loss": 0.70845628,
      "num_input_tokens_seen": 168948240,
      "step": 7816,
      "time_per_iteration": 3.752211093902588
    },
    {
      "auxiliary_loss_clip": 0.01108458,
      "auxiliary_loss_mlp": 0.01024869,
      "balance_loss_clip": 1.03992701,
      "balance_loss_mlp": 1.01689422,
      "epoch": 0.9399386761257741,
      "flos": 25372253462400.0,
      "grad_norm": 1.7604958528137247,
      "language_loss": 0.75509101,
      "learning_rate": 3.765450224711375e-08,
      "loss": 0.77642423,
      "num_input_tokens_seen": 168968745,
      "step": 7817,
      "time_per_iteration": 2.90429949760437
    },
    {
      "auxiliary_loss_clip": 0.0107969,
      "auxiliary_loss_mlp": 0.00762281,
      "balance_loss_clip": 1.03727698,
      "balance_loss_mlp": 1.00038123,
      "epoch": 0.9400589190164131,
      "flos": 27304965584640.0,
      "grad_norm": 1.8671486048638493,
      "language_loss": 0.79860693,
      "learning_rate": 3.750420660585396e-08,
      "loss": 0.81702662,
      "num_input_tokens_seen": 168990685,
      "step": 7818,
      "time_per_iteration": 2.837035894393921
    },
    {
      "auxiliary_loss_clip": 0.01155271,
      "auxiliary_loss_mlp": 0.01029645,
      "balance_loss_clip": 1.04720581,
      "balance_loss_mlp": 1.02243268,
      "epoch": 0.9401791619070522,
      "flos": 23399859790080.0,
      "grad_norm": 1.9175617504562394,
      "language_loss": 0.79909086,
      "learning_rate": 3.735420867364603e-08,
      "loss": 0.82094002,
      "num_input_tokens_seen": 169011665,
      "step": 7819,
      "time_per_iteration": 2.680291175842285
    },
    {
      "auxiliary_loss_clip": 0.01147543,
      "auxiliary_loss_mlp": 0.01020727,
      "balance_loss_clip": 1.04359162,
      "balance_loss_mlp": 1.01399493,
      "epoch": 0.9402994047976914,
      "flos": 35881403120640.0,
      "grad_norm": 2.477379300719503,
      "language_loss": 0.61618292,
      "learning_rate": 3.7204508473244186e-08,
      "loss": 0.6378656,
      "num_input_tokens_seen": 169035290,
      "step": 7820,
      "time_per_iteration": 2.767477512359619
    },
    {
      "auxiliary_loss_clip": 0.01167141,
      "auxiliary_loss_mlp": 0.01021809,
      "balance_loss_clip": 1.04579139,
      "balance_loss_mlp": 1.01453412,
      "epoch": 0.9404196476883304,
      "flos": 22236821320320.0,
      "grad_norm": 1.8899726659729192,
      "language_loss": 0.69381082,
      "learning_rate": 3.7055106027357395e-08,
      "loss": 0.71570033,
      "num_input_tokens_seen": 169055155,
      "step": 7821,
      "time_per_iteration": 3.5955708026885986
    },
    {
      "auxiliary_loss_clip": 0.01062418,
      "auxiliary_loss_mlp": 0.01022888,
      "balance_loss_clip": 1.03362763,
      "balance_loss_mlp": 1.01525271,
      "epoch": 0.9405398905789695,
      "flos": 18915802583040.0,
      "grad_norm": 2.2374468351577206,
      "language_loss": 0.72388887,
      "learning_rate": 3.690600135865063e-08,
      "loss": 0.74474192,
      "num_input_tokens_seen": 169072080,
      "step": 7822,
      "time_per_iteration": 3.0065696239471436
    },
    {
      "auxiliary_loss_clip": 0.01054789,
      "auxiliary_loss_mlp": 0.01002355,
      "balance_loss_clip": 1.01059914,
      "balance_loss_mlp": 1.00138938,
      "epoch": 0.9406601334696086,
      "flos": 70274130048000.0,
      "grad_norm": 0.7772666815974292,
      "language_loss": 0.58061838,
      "learning_rate": 3.675719448974246e-08,
      "loss": 0.60118979,
      "num_input_tokens_seen": 169137170,
      "step": 7823,
      "time_per_iteration": 3.4919917583465576
    },
    {
      "auxiliary_loss_clip": 0.01160835,
      "auxiliary_loss_mlp": 0.01025487,
      "balance_loss_clip": 1.04668379,
      "balance_loss_mlp": 1.01772094,
      "epoch": 0.9407803763602477,
      "flos": 22165071903360.0,
      "grad_norm": 2.269400970173491,
      "language_loss": 0.60251051,
      "learning_rate": 3.6608685443207054e-08,
      "loss": 0.62437367,
      "num_input_tokens_seen": 169156320,
      "step": 7824,
      "time_per_iteration": 2.7110936641693115
    },
    {
      "auxiliary_loss_clip": 0.01113639,
      "auxiliary_loss_mlp": 0.01021034,
      "balance_loss_clip": 1.03850818,
      "balance_loss_mlp": 1.01392889,
      "epoch": 0.9409006192508867,
      "flos": 18879496911360.0,
      "grad_norm": 2.0378158259443064,
      "language_loss": 0.6669938,
      "learning_rate": 3.646047424157306e-08,
      "loss": 0.68834054,
      "num_input_tokens_seen": 169173295,
      "step": 7825,
      "time_per_iteration": 2.6916346549987793
    },
    {
      "auxiliary_loss_clip": 0.01094407,
      "auxiliary_loss_mlp": 0.01023629,
      "balance_loss_clip": 1.03782666,
      "balance_loss_mlp": 1.01552272,
      "epoch": 0.9410208621415259,
      "flos": 23368258800000.0,
      "grad_norm": 5.20592252440323,
      "language_loss": 0.68680459,
      "learning_rate": 3.631256090732382e-08,
      "loss": 0.70798492,
      "num_input_tokens_seen": 169193755,
      "step": 7826,
      "time_per_iteration": 2.817882776260376
    },
    {
      "auxiliary_loss_clip": 0.01113174,
      "auxiliary_loss_mlp": 0.0102486,
      "balance_loss_clip": 1.04061258,
      "balance_loss_mlp": 1.01749301,
      "epoch": 0.941141105032165,
      "flos": 22742227635840.0,
      "grad_norm": 8.800570438630531,
      "language_loss": 0.82680953,
      "learning_rate": 3.6164945462897833e-08,
      "loss": 0.84818983,
      "num_input_tokens_seen": 169213045,
      "step": 7827,
      "time_per_iteration": 2.70538330078125
    },
    {
      "auxiliary_loss_clip": 0.01109125,
      "auxiliary_loss_mlp": 0.01023797,
      "balance_loss_clip": 1.03872859,
      "balance_loss_mlp": 1.01629829,
      "epoch": 0.941261347922804,
      "flos": 20704908130560.0,
      "grad_norm": 1.8025607393846168,
      "language_loss": 0.75897419,
      "learning_rate": 3.6017627930687856e-08,
      "loss": 0.78030348,
      "num_input_tokens_seen": 169232870,
      "step": 7828,
      "time_per_iteration": 2.7360455989837646
    },
    {
      "auxiliary_loss_clip": 0.01134579,
      "auxiliary_loss_mlp": 0.0102137,
      "balance_loss_clip": 1.04201698,
      "balance_loss_mlp": 1.01366019,
      "epoch": 0.9413815908134432,
      "flos": 19421998997760.0,
      "grad_norm": 2.051161132914582,
      "language_loss": 0.76988053,
      "learning_rate": 3.587060833304267e-08,
      "loss": 0.79144007,
      "num_input_tokens_seen": 169251060,
      "step": 7829,
      "time_per_iteration": 2.6498308181762695
    },
    {
      "auxiliary_loss_clip": 0.01085226,
      "auxiliary_loss_mlp": 0.01025671,
      "balance_loss_clip": 1.03910506,
      "balance_loss_mlp": 1.01764226,
      "epoch": 0.9415018337040822,
      "flos": 17493452853120.0,
      "grad_norm": 2.286501675760331,
      "language_loss": 0.64472193,
      "learning_rate": 3.5723886692264225e-08,
      "loss": 0.66583085,
      "num_input_tokens_seen": 169268600,
      "step": 7830,
      "time_per_iteration": 2.7232909202575684
    },
    {
      "auxiliary_loss_clip": 0.01107755,
      "auxiliary_loss_mlp": 0.00761768,
      "balance_loss_clip": 1.03969514,
      "balance_loss_mlp": 1.00039542,
      "epoch": 0.9416220765947213,
      "flos": 31831613343360.0,
      "grad_norm": 2.03014171065919,
      "language_loss": 0.61928761,
      "learning_rate": 3.557746303061071e-08,
      "loss": 0.63798285,
      "num_input_tokens_seen": 169290355,
      "step": 7831,
      "time_per_iteration": 2.845623731613159
    },
    {
      "auxiliary_loss_clip": 0.01069747,
      "auxiliary_loss_mlp": 0.01025886,
      "balance_loss_clip": 1.03473759,
      "balance_loss_mlp": 1.0179224,
      "epoch": 0.9417423194853605,
      "flos": 23511973115520.0,
      "grad_norm": 1.8280614476929808,
      "language_loss": 0.72568429,
      "learning_rate": 3.543133737029391e-08,
      "loss": 0.74664056,
      "num_input_tokens_seen": 169310865,
      "step": 7832,
      "time_per_iteration": 2.767529010772705
    },
    {
      "auxiliary_loss_clip": 0.01129675,
      "auxiliary_loss_mlp": 0.01028506,
      "balance_loss_clip": 1.04951441,
      "balance_loss_mlp": 1.02047777,
      "epoch": 0.9418625623759995,
      "flos": 23915106432000.0,
      "grad_norm": 3.0922807941246626,
      "language_loss": 0.6921683,
      "learning_rate": 3.5285509733481214e-08,
      "loss": 0.71375012,
      "num_input_tokens_seen": 169330590,
      "step": 7833,
      "time_per_iteration": 2.722646474838257
    },
    {
      "auxiliary_loss_clip": 0.01103586,
      "auxiliary_loss_mlp": 0.01030363,
      "balance_loss_clip": 1.04089797,
      "balance_loss_mlp": 1.02243602,
      "epoch": 0.9419828052666386,
      "flos": 18076965292800.0,
      "grad_norm": 2.341975754509218,
      "language_loss": 0.76732433,
      "learning_rate": 3.513998014229469e-08,
      "loss": 0.7886638,
      "num_input_tokens_seen": 169349540,
      "step": 7834,
      "time_per_iteration": 2.7264437675476074
    },
    {
      "auxiliary_loss_clip": 0.01105752,
      "auxiliary_loss_mlp": 0.01024564,
      "balance_loss_clip": 1.03791261,
      "balance_loss_mlp": 1.01739049,
      "epoch": 0.9421030481572777,
      "flos": 17712328377600.0,
      "grad_norm": 2.8686252551582583,
      "language_loss": 0.86439818,
      "learning_rate": 3.499474861881069e-08,
      "loss": 0.88570136,
      "num_input_tokens_seen": 169366765,
      "step": 7835,
      "time_per_iteration": 2.757502794265747
    },
    {
      "auxiliary_loss_clip": 0.01099614,
      "auxiliary_loss_mlp": 0.01025687,
      "balance_loss_clip": 1.04209876,
      "balance_loss_mlp": 1.01821303,
      "epoch": 0.9422232910479168,
      "flos": 20194114775040.0,
      "grad_norm": 2.5797717842825936,
      "language_loss": 0.68411875,
      "learning_rate": 3.4849815185061136e-08,
      "loss": 0.70537174,
      "num_input_tokens_seen": 169386655,
      "step": 7836,
      "time_per_iteration": 2.9619851112365723
    },
    {
      "auxiliary_loss_clip": 0.01105609,
      "auxiliary_loss_mlp": 0.01024333,
      "balance_loss_clip": 1.0415957,
      "balance_loss_mlp": 1.0173564,
      "epoch": 0.9423435339385559,
      "flos": 18442571875200.0,
      "grad_norm": 1.9486665145383992,
      "language_loss": 0.7614277,
      "learning_rate": 3.470517986303223e-08,
      "loss": 0.78272712,
      "num_input_tokens_seen": 169405640,
      "step": 7837,
      "time_per_iteration": 2.7235171794891357
    },
    {
      "auxiliary_loss_clip": 0.01113199,
      "auxiliary_loss_mlp": 0.01026435,
      "balance_loss_clip": 1.04339361,
      "balance_loss_mlp": 1.01865697,
      "epoch": 0.942463776829195,
      "flos": 20080636732800.0,
      "grad_norm": 1.86389550306958,
      "language_loss": 0.79223704,
      "learning_rate": 3.4560842674664856e-08,
      "loss": 0.81363338,
      "num_input_tokens_seen": 169424155,
      "step": 7838,
      "time_per_iteration": 3.662742853164673
    },
    {
      "auxiliary_loss_clip": 0.01139649,
      "auxiliary_loss_mlp": 0.01022833,
      "balance_loss_clip": 1.04292274,
      "balance_loss_mlp": 1.01563311,
      "epoch": 0.9425840197198341,
      "flos": 22636255536000.0,
      "grad_norm": 2.5098867508024334,
      "language_loss": 0.75468755,
      "learning_rate": 3.441680364185506e-08,
      "loss": 0.77631235,
      "num_input_tokens_seen": 169444025,
      "step": 7839,
      "time_per_iteration": 2.6759140491485596
    },
    {
      "auxiliary_loss_clip": 0.01124912,
      "auxiliary_loss_mlp": 0.01031999,
      "balance_loss_clip": 1.04222727,
      "balance_loss_mlp": 1.02382159,
      "epoch": 0.9427042626104731,
      "flos": 19937892084480.0,
      "grad_norm": 2.4788296866147497,
      "language_loss": 0.74540681,
      "learning_rate": 3.427306278645314e-08,
      "loss": 0.76697588,
      "num_input_tokens_seen": 169462480,
      "step": 7840,
      "time_per_iteration": 3.5388152599334717
    },
    {
      "auxiliary_loss_clip": 0.01035046,
      "auxiliary_loss_mlp": 0.00761656,
      "balance_loss_clip": 1.02846992,
      "balance_loss_mlp": 1.00044,
      "epoch": 0.9428245055011123,
      "flos": 22856998567680.0,
      "grad_norm": 2.461491686183111,
      "language_loss": 0.73537445,
      "learning_rate": 3.4129620130264767e-08,
      "loss": 0.7533415,
      "num_input_tokens_seen": 169480840,
      "step": 7841,
      "time_per_iteration": 3.015165090560913
    },
    {
      "auxiliary_loss_clip": 0.01170004,
      "auxiliary_loss_mlp": 0.01022362,
      "balance_loss_clip": 1.04885161,
      "balance_loss_mlp": 1.01488423,
      "epoch": 0.9429447483917514,
      "flos": 20951757371520.0,
      "grad_norm": 2.393615020762769,
      "language_loss": 0.78314471,
      "learning_rate": 3.398647569505009e-08,
      "loss": 0.80506837,
      "num_input_tokens_seen": 169498265,
      "step": 7842,
      "time_per_iteration": 3.5542068481445312
    },
    {
      "auxiliary_loss_clip": 0.01133677,
      "auxiliary_loss_mlp": 0.01026133,
      "balance_loss_clip": 1.04410505,
      "balance_loss_mlp": 1.01764476,
      "epoch": 0.9430649912823904,
      "flos": 18843658116480.0,
      "grad_norm": 2.8083989224502717,
      "language_loss": 0.7490133,
      "learning_rate": 3.384362950252373e-08,
      "loss": 0.77061141,
      "num_input_tokens_seen": 169515235,
      "step": 7843,
      "time_per_iteration": 3.557448387145996
    },
    {
      "auxiliary_loss_clip": 0.01101993,
      "auxiliary_loss_mlp": 0.01023706,
      "balance_loss_clip": 1.03992057,
      "balance_loss_mlp": 1.01591587,
      "epoch": 0.9431852341730296,
      "flos": 32556038837760.0,
      "grad_norm": 3.836986278979513,
      "language_loss": 0.57130611,
      "learning_rate": 3.3701081574355473e-08,
      "loss": 0.59256309,
      "num_input_tokens_seen": 169537195,
      "step": 7844,
      "time_per_iteration": 2.943946361541748
    },
    {
      "auxiliary_loss_clip": 0.01001487,
      "auxiliary_loss_mlp": 0.00754165,
      "balance_loss_clip": 1.01066577,
      "balance_loss_mlp": 1.00083911,
      "epoch": 0.9433054770636686,
      "flos": 66904490252160.0,
      "grad_norm": 0.6409147617854382,
      "language_loss": 0.51616061,
      "learning_rate": 3.3558831932169796e-08,
      "loss": 0.53371716,
      "num_input_tokens_seen": 169605865,
      "step": 7845,
      "time_per_iteration": 3.4238173961639404
    },
    {
      "auxiliary_loss_clip": 0.01064624,
      "auxiliary_loss_mlp": 0.0102808,
      "balance_loss_clip": 1.03412056,
      "balance_loss_mlp": 1.02028978,
      "epoch": 0.9434257199543077,
      "flos": 26140346916480.0,
      "grad_norm": 2.4604276588914114,
      "language_loss": 0.88530219,
      "learning_rate": 3.341688059754588e-08,
      "loss": 0.90622926,
      "num_input_tokens_seen": 169621520,
      "step": 7846,
      "time_per_iteration": 3.864732265472412
    },
    {
      "auxiliary_loss_clip": 0.01119974,
      "auxiliary_loss_mlp": 0.01024313,
      "balance_loss_clip": 1.04070008,
      "balance_loss_mlp": 1.01710963,
      "epoch": 0.9435459628449467,
      "flos": 25003486483200.0,
      "grad_norm": 2.2440412250272668,
      "language_loss": 0.7801286,
      "learning_rate": 3.327522759201762e-08,
      "loss": 0.80157149,
      "num_input_tokens_seen": 169641390,
      "step": 7847,
      "time_per_iteration": 2.8155689239501953
    },
    {
      "auxiliary_loss_clip": 0.01098358,
      "auxiliary_loss_mlp": 0.01029296,
      "balance_loss_clip": 1.04014945,
      "balance_loss_mlp": 1.021276,
      "epoch": 0.9436662057355859,
      "flos": 22163240309760.0,
      "grad_norm": 2.236360902093543,
      "language_loss": 0.66853976,
      "learning_rate": 3.313387293707359e-08,
      "loss": 0.6898163,
      "num_input_tokens_seen": 169660095,
      "step": 7848,
      "time_per_iteration": 2.7781238555908203
    },
    {
      "auxiliary_loss_clip": 0.01147016,
      "auxiliary_loss_mlp": 0.00762804,
      "balance_loss_clip": 1.04680836,
      "balance_loss_mlp": 1.00047266,
      "epoch": 0.943786448626225,
      "flos": 20118522602880.0,
      "grad_norm": 1.8051920503785388,
      "language_loss": 0.68477803,
      "learning_rate": 3.29928166541571e-08,
      "loss": 0.70387626,
      "num_input_tokens_seen": 169679050,
      "step": 7849,
      "time_per_iteration": 2.7417914867401123
    },
    {
      "auxiliary_loss_clip": 0.01150591,
      "auxiliary_loss_mlp": 0.01024046,
      "balance_loss_clip": 1.04654145,
      "balance_loss_mlp": 1.01703691,
      "epoch": 0.943906691516864,
      "flos": 22090808534400.0,
      "grad_norm": 1.9587369078127814,
      "language_loss": 0.80755609,
      "learning_rate": 3.2852058764666346e-08,
      "loss": 0.82930243,
      "num_input_tokens_seen": 169698150,
      "step": 7850,
      "time_per_iteration": 2.6452994346618652
    },
    {
      "auxiliary_loss_clip": 0.01120573,
      "auxiliary_loss_mlp": 0.01024034,
      "balance_loss_clip": 1.04334497,
      "balance_loss_mlp": 1.01654208,
      "epoch": 0.9440269344075032,
      "flos": 35298501212160.0,
      "grad_norm": 2.0378077924036266,
      "language_loss": 0.68780708,
      "learning_rate": 3.2711599289954264e-08,
      "loss": 0.70925319,
      "num_input_tokens_seen": 169722185,
      "step": 7851,
      "time_per_iteration": 2.8091890811920166
    },
    {
      "auxiliary_loss_clip": 0.01129963,
      "auxiliary_loss_mlp": 0.01028107,
      "balance_loss_clip": 1.04182982,
      "balance_loss_mlp": 1.02040267,
      "epoch": 0.9441471772981422,
      "flos": 19238136255360.0,
      "grad_norm": 2.8559958799379794,
      "language_loss": 0.77679765,
      "learning_rate": 3.257143825132847e-08,
      "loss": 0.79837841,
      "num_input_tokens_seen": 169740355,
      "step": 7852,
      "time_per_iteration": 2.7321672439575195
    },
    {
      "auxiliary_loss_clip": 0.01122595,
      "auxiliary_loss_mlp": 0.00761451,
      "balance_loss_clip": 1.04080367,
      "balance_loss_mlp": 1.00046384,
      "epoch": 0.9442674201887813,
      "flos": 25739799379200.0,
      "grad_norm": 1.8644892339088899,
      "language_loss": 0.76064122,
      "learning_rate": 3.243157567005106e-08,
      "loss": 0.77948165,
      "num_input_tokens_seen": 169758535,
      "step": 7853,
      "time_per_iteration": 2.772968292236328
    },
    {
      "auxiliary_loss_clip": 0.01090817,
      "auxiliary_loss_mlp": 0.01029466,
      "balance_loss_clip": 1.04300106,
      "balance_loss_mlp": 1.02109122,
      "epoch": 0.9443876630794205,
      "flos": 15523321737600.0,
      "grad_norm": 2.10968548713726,
      "language_loss": 0.63748121,
      "learning_rate": 3.2292011567339296e-08,
      "loss": 0.65868413,
      "num_input_tokens_seen": 169776340,
      "step": 7854,
      "time_per_iteration": 2.834888219833374
    },
    {
      "auxiliary_loss_clip": 0.01153071,
      "auxiliary_loss_mlp": 0.01022205,
      "balance_loss_clip": 1.04609418,
      "balance_loss_mlp": 1.01463556,
      "epoch": 0.9445079059700595,
      "flos": 13400821128960.0,
      "grad_norm": 3.6513844225227463,
      "language_loss": 0.55762708,
      "learning_rate": 3.21527459643649e-08,
      "loss": 0.57937986,
      "num_input_tokens_seen": 169793225,
      "step": 7855,
      "time_per_iteration": 2.6345715522766113
    },
    {
      "auxiliary_loss_clip": 0.01096173,
      "auxiliary_loss_mlp": 0.01026697,
      "balance_loss_clip": 1.04211926,
      "balance_loss_mlp": 1.0195241,
      "epoch": 0.9446281488606986,
      "flos": 23659242877440.0,
      "grad_norm": 2.1928120396865083,
      "language_loss": 0.74360406,
      "learning_rate": 3.2013778882254536e-08,
      "loss": 0.76483274,
      "num_input_tokens_seen": 169812020,
      "step": 7856,
      "time_per_iteration": 2.7770302295684814
    },
    {
      "auxiliary_loss_clip": 0.01152115,
      "auxiliary_loss_mlp": 0.01022856,
      "balance_loss_clip": 1.04640317,
      "balance_loss_mlp": 1.01581073,
      "epoch": 0.9447483917513377,
      "flos": 25557337267200.0,
      "grad_norm": 2.0978428537544827,
      "language_loss": 0.75758314,
      "learning_rate": 3.1875110342088676e-08,
      "loss": 0.77933288,
      "num_input_tokens_seen": 169833470,
      "step": 7857,
      "time_per_iteration": 2.7872862815856934
    },
    {
      "auxiliary_loss_clip": 0.01129873,
      "auxiliary_loss_mlp": 0.01022021,
      "balance_loss_clip": 1.04049206,
      "balance_loss_mlp": 1.01452303,
      "epoch": 0.9448686346419768,
      "flos": 24535463247360.0,
      "grad_norm": 5.000697960252112,
      "language_loss": 0.65515256,
      "learning_rate": 3.1736740364904035e-08,
      "loss": 0.6766715,
      "num_input_tokens_seen": 169854000,
      "step": 7858,
      "time_per_iteration": 2.7540972232818604
    },
    {
      "auxiliary_loss_clip": 0.01141303,
      "auxiliary_loss_mlp": 0.0076199,
      "balance_loss_clip": 1.04528952,
      "balance_loss_mlp": 1.00035298,
      "epoch": 0.9449888775326158,
      "flos": 14721256995840.0,
      "grad_norm": 1.9880270143520022,
      "language_loss": 0.7734428,
      "learning_rate": 3.159866897169094e-08,
      "loss": 0.79247576,
      "num_input_tokens_seen": 169872200,
      "step": 7859,
      "time_per_iteration": 2.6692142486572266
    },
    {
      "auxiliary_loss_clip": 0.01124258,
      "auxiliary_loss_mlp": 0.00762208,
      "balance_loss_clip": 1.04246986,
      "balance_loss_mlp": 1.00044906,
      "epoch": 0.945109120423255,
      "flos": 15447873219840.0,
      "grad_norm": 1.761560178554226,
      "language_loss": 0.75758016,
      "learning_rate": 3.146089618339487e-08,
      "loss": 0.77644479,
      "num_input_tokens_seen": 169889055,
      "step": 7860,
      "time_per_iteration": 2.708678722381592
    },
    {
      "auxiliary_loss_clip": 0.01151164,
      "auxiliary_loss_mlp": 0.01024968,
      "balance_loss_clip": 1.04536867,
      "balance_loss_mlp": 1.01700473,
      "epoch": 0.9452293633138941,
      "flos": 25448097029760.0,
      "grad_norm": 1.9031909550723838,
      "language_loss": 0.68071717,
      "learning_rate": 3.132342202091554e-08,
      "loss": 0.70247841,
      "num_input_tokens_seen": 169909280,
      "step": 7861,
      "time_per_iteration": 2.6950483322143555
    },
    {
      "auxiliary_loss_clip": 0.01099755,
      "auxiliary_loss_mlp": 0.01023803,
      "balance_loss_clip": 1.0360862,
      "balance_loss_mlp": 1.0158217,
      "epoch": 0.9453496062045331,
      "flos": 21215342350080.0,
      "grad_norm": 4.505672860544062,
      "language_loss": 0.68545747,
      "learning_rate": 3.1186246505107595e-08,
      "loss": 0.70669305,
      "num_input_tokens_seen": 169928420,
      "step": 7862,
      "time_per_iteration": 2.8267576694488525
    },
    {
      "auxiliary_loss_clip": 0.01136982,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.0418961,
      "balance_loss_mlp": 1.01686406,
      "epoch": 0.9454698490951723,
      "flos": 20010898477440.0,
      "grad_norm": 2.2211712031475193,
      "language_loss": 0.83604884,
      "learning_rate": 3.104936965678084e-08,
      "loss": 0.85766256,
      "num_input_tokens_seen": 169946750,
      "step": 7863,
      "time_per_iteration": 2.7483251094818115
    },
    {
      "auxiliary_loss_clip": 0.01143288,
      "auxiliary_loss_mlp": 0.00762599,
      "balance_loss_clip": 1.04169369,
      "balance_loss_mlp": 1.00036812,
      "epoch": 0.9455900919858113,
      "flos": 21069652786560.0,
      "grad_norm": 1.8859538211120905,
      "language_loss": 0.81937408,
      "learning_rate": 3.091279149669956e-08,
      "loss": 0.83843297,
      "num_input_tokens_seen": 169965540,
      "step": 7864,
      "time_per_iteration": 3.5482029914855957
    },
    {
      "auxiliary_loss_clip": 0.0113549,
      "auxiliary_loss_mlp": 0.01023031,
      "balance_loss_clip": 1.04448926,
      "balance_loss_mlp": 1.01563442,
      "epoch": 0.9457103348764504,
      "flos": 20740854666240.0,
      "grad_norm": 1.9267563932322342,
      "language_loss": 0.73918653,
      "learning_rate": 3.0776512045581624e-08,
      "loss": 0.76077175,
      "num_input_tokens_seen": 169984330,
      "step": 7865,
      "time_per_iteration": 3.7710375785827637
    },
    {
      "auxiliary_loss_clip": 0.01114658,
      "auxiliary_loss_mlp": 0.01027285,
      "balance_loss_clip": 1.04148459,
      "balance_loss_mlp": 1.01928353,
      "epoch": 0.9458305777670896,
      "flos": 21428363957760.0,
      "grad_norm": 1.844912915904255,
      "language_loss": 0.77700174,
      "learning_rate": 3.0640531324101384e-08,
      "loss": 0.79842114,
      "num_input_tokens_seen": 170002095,
      "step": 7866,
      "time_per_iteration": 2.735133171081543
    },
    {
      "auxiliary_loss_clip": 0.01092325,
      "auxiliary_loss_mlp": 0.01027492,
      "balance_loss_clip": 1.04173088,
      "balance_loss_mlp": 1.01926053,
      "epoch": 0.9459508206577286,
      "flos": 20011185786240.0,
      "grad_norm": 2.0993363868293358,
      "language_loss": 0.75813818,
      "learning_rate": 3.0504849352886554e-08,
      "loss": 0.77933633,
      "num_input_tokens_seen": 170020240,
      "step": 7867,
      "time_per_iteration": 2.7395997047424316
    },
    {
      "auxiliary_loss_clip": 0.01157049,
      "auxiliary_loss_mlp": 0.01028513,
      "balance_loss_clip": 1.04759836,
      "balance_loss_mlp": 1.02110958,
      "epoch": 0.9460710635483677,
      "flos": 12166428291840.0,
      "grad_norm": 3.699479828583891,
      "language_loss": 0.71803844,
      "learning_rate": 3.036946615252023e-08,
      "loss": 0.73989403,
      "num_input_tokens_seen": 170035770,
      "step": 7868,
      "time_per_iteration": 3.5278756618499756
    },
    {
      "auxiliary_loss_clip": 0.01106614,
      "auxiliary_loss_mlp": 0.01026204,
      "balance_loss_clip": 1.04031205,
      "balance_loss_mlp": 1.01877117,
      "epoch": 0.9461913064390068,
      "flos": 34276196229120.0,
      "grad_norm": 2.8565808661935765,
      "language_loss": 0.67149699,
      "learning_rate": 3.0234381743539984e-08,
      "loss": 0.6928252,
      "num_input_tokens_seen": 170053385,
      "step": 7869,
      "time_per_iteration": 2.7501728534698486
    },
    {
      "auxiliary_loss_clip": 0.01114773,
      "auxiliary_loss_mlp": 0.01027755,
      "balance_loss_clip": 1.04133427,
      "balance_loss_mlp": 1.01995277,
      "epoch": 0.9463115493296459,
      "flos": 19463763536640.0,
      "grad_norm": 2.1264542526049723,
      "language_loss": 0.80023944,
      "learning_rate": 3.0099596146437863e-08,
      "loss": 0.82166481,
      "num_input_tokens_seen": 170070490,
      "step": 7870,
      "time_per_iteration": 2.606038808822632
    },
    {
      "auxiliary_loss_clip": 0.01029909,
      "auxiliary_loss_mlp": 0.01000914,
      "balance_loss_clip": 1.01078689,
      "balance_loss_mlp": 0.99999571,
      "epoch": 0.946431792220285,
      "flos": 70570824387840.0,
      "grad_norm": 0.7717982404578348,
      "language_loss": 0.60085243,
      "learning_rate": 2.996510938166086e-08,
      "loss": 0.62116063,
      "num_input_tokens_seen": 170133465,
      "step": 7871,
      "time_per_iteration": 3.3252599239349365
    },
    {
      "auxiliary_loss_clip": 0.01152602,
      "auxiliary_loss_mlp": 0.0102358,
      "balance_loss_clip": 1.04399347,
      "balance_loss_mlp": 1.01604652,
      "epoch": 0.9465520351109241,
      "flos": 18947906363520.0,
      "grad_norm": 2.306430036116706,
      "language_loss": 0.73726964,
      "learning_rate": 2.983092146960997e-08,
      "loss": 0.75903147,
      "num_input_tokens_seen": 170150810,
      "step": 7872,
      "time_per_iteration": 2.5254833698272705
    },
    {
      "auxiliary_loss_clip": 0.01102952,
      "auxiliary_loss_mlp": 0.01028528,
      "balance_loss_clip": 1.04193497,
      "balance_loss_mlp": 1.02090442,
      "epoch": 0.9466722780015632,
      "flos": 19135647774720.0,
      "grad_norm": 2.3089029418575024,
      "language_loss": 0.80434638,
      "learning_rate": 2.9697032430642256e-08,
      "loss": 0.82566124,
      "num_input_tokens_seen": 170169025,
      "step": 7873,
      "time_per_iteration": 3.7019667625427246
    },
    {
      "auxiliary_loss_clip": 0.01083355,
      "auxiliary_loss_mlp": 0.01023059,
      "balance_loss_clip": 1.03351688,
      "balance_loss_mlp": 1.01520872,
      "epoch": 0.9467925208922022,
      "flos": 17237912520960.0,
      "grad_norm": 2.497861620409902,
      "language_loss": 0.73527706,
      "learning_rate": 2.9563442285067906e-08,
      "loss": 0.75634122,
      "num_input_tokens_seen": 170186070,
      "step": 7874,
      "time_per_iteration": 2.79412579536438
    },
    {
      "auxiliary_loss_clip": 0.01130394,
      "auxiliary_loss_mlp": 0.01023693,
      "balance_loss_clip": 1.04223037,
      "balance_loss_mlp": 1.01628113,
      "epoch": 0.9469127637828414,
      "flos": 29169016859520.0,
      "grad_norm": 2.2231662523334776,
      "language_loss": 0.79536825,
      "learning_rate": 2.943015105315294e-08,
      "loss": 0.81690913,
      "num_input_tokens_seen": 170206265,
      "step": 7875,
      "time_per_iteration": 2.7316200733184814
    },
    {
      "auxiliary_loss_clip": 0.01129818,
      "auxiliary_loss_mlp": 0.0102907,
      "balance_loss_clip": 1.04306018,
      "balance_loss_mlp": 1.02084446,
      "epoch": 0.9470330066734804,
      "flos": 26030460234240.0,
      "grad_norm": 4.365482349491802,
      "language_loss": 0.66540235,
      "learning_rate": 2.929715875511718e-08,
      "loss": 0.68699127,
      "num_input_tokens_seen": 170225300,
      "step": 7876,
      "time_per_iteration": 2.811819314956665
    },
    {
      "auxiliary_loss_clip": 0.01095465,
      "auxiliary_loss_mlp": 0.01026961,
      "balance_loss_clip": 1.04000378,
      "balance_loss_mlp": 1.01903343,
      "epoch": 0.9471532495641195,
      "flos": 23440906056960.0,
      "grad_norm": 1.9519210075121585,
      "language_loss": 0.70098305,
      "learning_rate": 2.9164465411135375e-08,
      "loss": 0.72220731,
      "num_input_tokens_seen": 170245070,
      "step": 7877,
      "time_per_iteration": 2.827263832092285
    },
    {
      "auxiliary_loss_clip": 0.01142128,
      "auxiliary_loss_mlp": 0.00761969,
      "balance_loss_clip": 1.04776108,
      "balance_loss_mlp": 1.00050676,
      "epoch": 0.9472734924547586,
      "flos": 15815850099840.0,
      "grad_norm": 2.2938284605106105,
      "language_loss": 0.8078689,
      "learning_rate": 2.9032071041337426e-08,
      "loss": 0.8269099,
      "num_input_tokens_seen": 170263305,
      "step": 7878,
      "time_per_iteration": 2.767090082168579
    },
    {
      "auxiliary_loss_clip": 0.01167624,
      "auxiliary_loss_mlp": 0.01022681,
      "balance_loss_clip": 1.0472548,
      "balance_loss_mlp": 1.01518536,
      "epoch": 0.9473937353453977,
      "flos": 11181793697280.0,
      "grad_norm": 1.8290416454671985,
      "language_loss": 0.7306478,
      "learning_rate": 2.889997566580704e-08,
      "loss": 0.75255084,
      "num_input_tokens_seen": 170281460,
      "step": 7879,
      "time_per_iteration": 2.7135300636291504
    },
    {
      "auxiliary_loss_clip": 0.01128204,
      "auxiliary_loss_mlp": 0.01025949,
      "balance_loss_clip": 1.04168797,
      "balance_loss_mlp": 1.0177598,
      "epoch": 0.9475139782360368,
      "flos": 25775530433280.0,
      "grad_norm": 1.8062967270584487,
      "language_loss": 0.70409852,
      "learning_rate": 2.8768179304583086e-08,
      "loss": 0.72564,
      "num_input_tokens_seen": 170303515,
      "step": 7880,
      "time_per_iteration": 2.9123332500457764
    },
    {
      "auxiliary_loss_clip": 0.01103366,
      "auxiliary_loss_mlp": 0.00761766,
      "balance_loss_clip": 1.040133,
      "balance_loss_mlp": 1.00052118,
      "epoch": 0.9476342211266758,
      "flos": 22820046451200.0,
      "grad_norm": 1.7038024748890133,
      "language_loss": 0.73583817,
      "learning_rate": 2.8636681977659117e-08,
      "loss": 0.75448942,
      "num_input_tokens_seen": 170323165,
      "step": 7881,
      "time_per_iteration": 2.7803921699523926
    },
    {
      "auxiliary_loss_clip": 0.01161488,
      "auxiliary_loss_mlp": 0.01026069,
      "balance_loss_clip": 1.05039299,
      "balance_loss_mlp": 1.01812398,
      "epoch": 0.947754464017315,
      "flos": 20193611984640.0,
      "grad_norm": 2.2246982107560855,
      "language_loss": 0.78142518,
      "learning_rate": 2.850548370498318e-08,
      "loss": 0.80330074,
      "num_input_tokens_seen": 170341005,
      "step": 7882,
      "time_per_iteration": 2.663282871246338
    },
    {
      "auxiliary_loss_clip": 0.01151348,
      "auxiliary_loss_mlp": 0.01026538,
      "balance_loss_clip": 1.04339242,
      "balance_loss_mlp": 1.01930189,
      "epoch": 0.9478747069079541,
      "flos": 24717925359360.0,
      "grad_norm": 1.835714669848363,
      "language_loss": 0.71588492,
      "learning_rate": 2.8374584506457798e-08,
      "loss": 0.73766375,
      "num_input_tokens_seen": 170362280,
      "step": 7883,
      "time_per_iteration": 2.688249349594116
    },
    {
      "auxiliary_loss_clip": 0.01128357,
      "auxiliary_loss_mlp": 0.01024028,
      "balance_loss_clip": 1.04447627,
      "balance_loss_mlp": 1.01589251,
      "epoch": 0.9479949497985931,
      "flos": 21361355136000.0,
      "grad_norm": 3.8550606984059486,
      "language_loss": 0.67841208,
      "learning_rate": 2.824398440193998e-08,
      "loss": 0.69993591,
      "num_input_tokens_seen": 170381080,
      "step": 7884,
      "time_per_iteration": 2.7182180881500244
    },
    {
      "auxiliary_loss_clip": 0.01106661,
      "auxiliary_loss_mlp": 0.01021023,
      "balance_loss_clip": 1.04249799,
      "balance_loss_mlp": 1.01341128,
      "epoch": 0.9481151926892323,
      "flos": 18148606968960.0,
      "grad_norm": 2.0281578095011836,
      "language_loss": 0.7169106,
      "learning_rate": 2.811368341124232e-08,
      "loss": 0.73818743,
      "num_input_tokens_seen": 170400150,
      "step": 7885,
      "time_per_iteration": 2.713470220565796
    },
    {
      "auxiliary_loss_clip": 0.01170999,
      "auxiliary_loss_mlp": 0.01019307,
      "balance_loss_clip": 1.04964244,
      "balance_loss_mlp": 1.01186192,
      "epoch": 0.9482354355798713,
      "flos": 22128012046080.0,
      "grad_norm": 9.224780400061016,
      "language_loss": 0.6860432,
      "learning_rate": 2.7983681554131222e-08,
      "loss": 0.70794624,
      "num_input_tokens_seen": 170420410,
      "step": 7886,
      "time_per_iteration": 2.612142324447632
    },
    {
      "auxiliary_loss_clip": 0.01129844,
      "auxiliary_loss_mlp": 0.01029198,
      "balance_loss_clip": 1.04023468,
      "balance_loss_mlp": 1.02081776,
      "epoch": 0.9483556784705104,
      "flos": 19063072344960.0,
      "grad_norm": 2.924654366352087,
      "language_loss": 0.70255935,
      "learning_rate": 2.7853978850327365e-08,
      "loss": 0.72414982,
      "num_input_tokens_seen": 170439580,
      "step": 7887,
      "time_per_iteration": 2.709883213043213
    },
    {
      "auxiliary_loss_clip": 0.01154502,
      "auxiliary_loss_mlp": 0.01025547,
      "balance_loss_clip": 1.04757524,
      "balance_loss_mlp": 1.01796508,
      "epoch": 0.9484759213611496,
      "flos": 25777110631680.0,
      "grad_norm": 2.051156444511218,
      "language_loss": 0.86871612,
      "learning_rate": 2.7724575319507225e-08,
      "loss": 0.89051652,
      "num_input_tokens_seen": 170459290,
      "step": 7888,
      "time_per_iteration": 2.6996898651123047
    },
    {
      "auxiliary_loss_clip": 0.01121105,
      "auxiliary_loss_mlp": 0.01027269,
      "balance_loss_clip": 1.03881955,
      "balance_loss_mlp": 1.0200212,
      "epoch": 0.9485961642517886,
      "flos": 20667740532480.0,
      "grad_norm": 4.605831935931043,
      "language_loss": 0.774279,
      "learning_rate": 2.759547098130044e-08,
      "loss": 0.79576272,
      "num_input_tokens_seen": 170478020,
      "step": 7889,
      "time_per_iteration": 2.831099033355713
    },
    {
      "auxiliary_loss_clip": 0.01122887,
      "auxiliary_loss_mlp": 0.01023497,
      "balance_loss_clip": 1.04123747,
      "balance_loss_mlp": 1.01659489,
      "epoch": 0.9487164071424277,
      "flos": 22674069578880.0,
      "grad_norm": 2.9436611189319173,
      "language_loss": 0.76814401,
      "learning_rate": 2.746666585529267e-08,
      "loss": 0.78960788,
      "num_input_tokens_seen": 170498295,
      "step": 7890,
      "time_per_iteration": 3.843724489212036
    },
    {
      "auxiliary_loss_clip": 0.01122994,
      "auxiliary_loss_mlp": 0.01022295,
      "balance_loss_clip": 1.04273224,
      "balance_loss_mlp": 1.01510036,
      "epoch": 0.9488366500330668,
      "flos": 38726461716480.0,
      "grad_norm": 2.1019041195951362,
      "language_loss": 0.73983365,
      "learning_rate": 2.73381599610234e-08,
      "loss": 0.7612865,
      "num_input_tokens_seen": 170518695,
      "step": 7891,
      "time_per_iteration": 3.7621591091156006
    },
    {
      "auxiliary_loss_clip": 0.01139036,
      "auxiliary_loss_mlp": 0.01020946,
      "balance_loss_clip": 1.04316139,
      "balance_loss_mlp": 1.01299512,
      "epoch": 0.9489568929237059,
      "flos": 27890920149120.0,
      "grad_norm": 1.852316116857838,
      "language_loss": 0.71479666,
      "learning_rate": 2.7209953317987033e-08,
      "loss": 0.73639649,
      "num_input_tokens_seen": 170539735,
      "step": 7892,
      "time_per_iteration": 2.8021459579467773
    },
    {
      "auxiliary_loss_clip": 0.01135433,
      "auxiliary_loss_mlp": 0.00762533,
      "balance_loss_clip": 1.04497671,
      "balance_loss_mlp": 1.00046778,
      "epoch": 0.9490771358143449,
      "flos": 33580642291200.0,
      "grad_norm": 1.9315670774419125,
      "language_loss": 0.78029656,
      "learning_rate": 2.7082045945631793e-08,
      "loss": 0.79927617,
      "num_input_tokens_seen": 170561950,
      "step": 7893,
      "time_per_iteration": 2.800333261489868
    },
    {
      "auxiliary_loss_clip": 0.01121125,
      "auxiliary_loss_mlp": 0.00762223,
      "balance_loss_clip": 1.04076576,
      "balance_loss_mlp": 1.00042593,
      "epoch": 0.9491973787049841,
      "flos": 14793796512000.0,
      "grad_norm": 2.2196852189052056,
      "language_loss": 0.69712651,
      "learning_rate": 2.6954437863361712e-08,
      "loss": 0.71596003,
      "num_input_tokens_seen": 170579865,
      "step": 7894,
      "time_per_iteration": 3.7117655277252197
    },
    {
      "auxiliary_loss_clip": 0.01109673,
      "auxiliary_loss_mlp": 0.01024837,
      "balance_loss_clip": 1.03940034,
      "balance_loss_mlp": 1.01760125,
      "epoch": 0.9493176215956232,
      "flos": 25332535998720.0,
      "grad_norm": 2.0674441392512173,
      "language_loss": 0.70730776,
      "learning_rate": 2.6827129090534862e-08,
      "loss": 0.72865283,
      "num_input_tokens_seen": 170600165,
      "step": 7895,
      "time_per_iteration": 2.798168897628784
    },
    {
      "auxiliary_loss_clip": 0.01144853,
      "auxiliary_loss_mlp": 0.01024365,
      "balance_loss_clip": 1.04238939,
      "balance_loss_mlp": 1.01702154,
      "epoch": 0.9494378644862622,
      "flos": 21029971236480.0,
      "grad_norm": 2.534935691457149,
      "language_loss": 0.78136992,
      "learning_rate": 2.670011964646335e-08,
      "loss": 0.80306208,
      "num_input_tokens_seen": 170618845,
      "step": 7896,
      "time_per_iteration": 2.6185519695281982
    },
    {
      "auxiliary_loss_clip": 0.01113876,
      "auxiliary_loss_mlp": 0.01024459,
      "balance_loss_clip": 1.04185224,
      "balance_loss_mlp": 1.01632333,
      "epoch": 0.9495581073769014,
      "flos": 15195134148480.0,
      "grad_norm": 2.0813214767938066,
      "language_loss": 0.67915535,
      "learning_rate": 2.657340955041487e-08,
      "loss": 0.70053875,
      "num_input_tokens_seen": 170637620,
      "step": 7897,
      "time_per_iteration": 2.757967233657837
    },
    {
      "auxiliary_loss_clip": 0.01135508,
      "auxiliary_loss_mlp": 0.01027233,
      "balance_loss_clip": 1.04265976,
      "balance_loss_mlp": 1.0195266,
      "epoch": 0.9496783502675404,
      "flos": 28616566705920.0,
      "grad_norm": 2.202651969737753,
      "language_loss": 0.71495897,
      "learning_rate": 2.6446998821611167e-08,
      "loss": 0.73658639,
      "num_input_tokens_seen": 170657815,
      "step": 7898,
      "time_per_iteration": 3.6631102561950684
    },
    {
      "auxiliary_loss_clip": 0.01103779,
      "auxiliary_loss_mlp": 0.01028082,
      "balance_loss_clip": 1.03785968,
      "balance_loss_mlp": 1.02062201,
      "epoch": 0.9497985931581795,
      "flos": 14866874732160.0,
      "grad_norm": 2.315024986378149,
      "language_loss": 0.71689153,
      "learning_rate": 2.6320887479228228e-08,
      "loss": 0.73821008,
      "num_input_tokens_seen": 170674415,
      "step": 7899,
      "time_per_iteration": 2.7879111766815186
    },
    {
      "auxiliary_loss_clip": 0.01106517,
      "auxiliary_loss_mlp": 0.01032457,
      "balance_loss_clip": 1.04214919,
      "balance_loss_mlp": 1.02478623,
      "epoch": 0.9499188360488187,
      "flos": 27193319136000.0,
      "grad_norm": 2.7210594789915694,
      "language_loss": 0.72219861,
      "learning_rate": 2.619507554239786e-08,
      "loss": 0.74358833,
      "num_input_tokens_seen": 170692975,
      "step": 7900,
      "time_per_iteration": 2.7733447551727295
    },
    {
      "auxiliary_loss_clip": 0.01123132,
      "auxiliary_loss_mlp": 0.0102403,
      "balance_loss_clip": 1.04165864,
      "balance_loss_mlp": 1.0164665,
      "epoch": 0.9500390789394577,
      "flos": 24316479982080.0,
      "grad_norm": 1.9372865427428936,
      "language_loss": 0.69810045,
      "learning_rate": 2.606956303020502e-08,
      "loss": 0.71957207,
      "num_input_tokens_seen": 170713780,
      "step": 7901,
      "time_per_iteration": 2.7285304069519043
    },
    {
      "auxiliary_loss_clip": 0.01081074,
      "auxiliary_loss_mlp": 0.01021159,
      "balance_loss_clip": 1.03663111,
      "balance_loss_mlp": 1.01393819,
      "epoch": 0.9501593218300968,
      "flos": 14354752573440.0,
      "grad_norm": 3.016184260600256,
      "language_loss": 0.84359574,
      "learning_rate": 2.5944349961690036e-08,
      "loss": 0.86461818,
      "num_input_tokens_seen": 170730800,
      "step": 7902,
      "time_per_iteration": 2.7470736503601074
    },
    {
      "auxiliary_loss_clip": 0.01045629,
      "auxiliary_loss_mlp": 0.01025819,
      "balance_loss_clip": 1.03817618,
      "balance_loss_mlp": 1.01778436,
      "epoch": 0.9502795647207359,
      "flos": 38728113742080.0,
      "grad_norm": 1.8399622461099965,
      "language_loss": 0.73101795,
      "learning_rate": 2.581943635584749e-08,
      "loss": 0.75173235,
      "num_input_tokens_seen": 170753630,
      "step": 7903,
      "time_per_iteration": 3.1027050018310547
    },
    {
      "auxiliary_loss_clip": 0.01148172,
      "auxiliary_loss_mlp": 0.01018996,
      "balance_loss_clip": 1.04463935,
      "balance_loss_mlp": 1.01208222,
      "epoch": 0.950399807611375,
      "flos": 40808023799040.0,
      "grad_norm": 1.7181073118843502,
      "language_loss": 0.65479237,
      "learning_rate": 2.569482223162689e-08,
      "loss": 0.67646408,
      "num_input_tokens_seen": 170777605,
      "step": 7904,
      "time_per_iteration": 2.9439008235931396
    },
    {
      "auxiliary_loss_clip": 0.01096179,
      "auxiliary_loss_mlp": 0.01024647,
      "balance_loss_clip": 1.03744888,
      "balance_loss_mlp": 1.01664853,
      "epoch": 0.950520050502014,
      "flos": 23440403266560.0,
      "grad_norm": 1.9129808461069753,
      "language_loss": 0.72896743,
      "learning_rate": 2.5570507607932e-08,
      "loss": 0.75017571,
      "num_input_tokens_seen": 170797520,
      "step": 7905,
      "time_per_iteration": 2.831721305847168
    },
    {
      "auxiliary_loss_clip": 0.01061578,
      "auxiliary_loss_mlp": 0.01020029,
      "balance_loss_clip": 1.03354514,
      "balance_loss_mlp": 1.01239324,
      "epoch": 0.9506402933926532,
      "flos": 17783718658560.0,
      "grad_norm": 2.5435588485998726,
      "language_loss": 0.63989937,
      "learning_rate": 2.54464925036213e-08,
      "loss": 0.66071546,
      "num_input_tokens_seen": 170814810,
      "step": 7906,
      "time_per_iteration": 3.0307838916778564
    },
    {
      "auxiliary_loss_clip": 0.01093077,
      "auxiliary_loss_mlp": 0.00761783,
      "balance_loss_clip": 1.03969049,
      "balance_loss_mlp": 1.0004065,
      "epoch": 0.9507605362832923,
      "flos": 32561928668160.0,
      "grad_norm": 1.9806088083253204,
      "language_loss": 0.60879731,
      "learning_rate": 2.532277693750773e-08,
      "loss": 0.62734592,
      "num_input_tokens_seen": 170835735,
      "step": 7907,
      "time_per_iteration": 4.119615793228149
    },
    {
      "auxiliary_loss_clip": 0.01170375,
      "auxiliary_loss_mlp": 0.01025707,
      "balance_loss_clip": 1.05020261,
      "balance_loss_mlp": 1.01780903,
      "epoch": 0.9508807791739313,
      "flos": 19602054898560.0,
      "grad_norm": 2.22411800852592,
      "language_loss": 0.75954449,
      "learning_rate": 2.5199360928358948e-08,
      "loss": 0.78150529,
      "num_input_tokens_seen": 170852970,
      "step": 7908,
      "time_per_iteration": 2.6565873622894287
    },
    {
      "auxiliary_loss_clip": 0.01113578,
      "auxiliary_loss_mlp": 0.01019014,
      "balance_loss_clip": 1.04216743,
      "balance_loss_mlp": 1.01223731,
      "epoch": 0.9510010220645704,
      "flos": 21471852349440.0,
      "grad_norm": 1.7414060144400223,
      "language_loss": 0.872091,
      "learning_rate": 2.507624449489665e-08,
      "loss": 0.89341694,
      "num_input_tokens_seen": 170871600,
      "step": 7909,
      "time_per_iteration": 2.7754907608032227
    },
    {
      "auxiliary_loss_clip": 0.01125681,
      "auxiliary_loss_mlp": 0.01032074,
      "balance_loss_clip": 1.0436902,
      "balance_loss_mlp": 1.02423048,
      "epoch": 0.9511212649552095,
      "flos": 18879999701760.0,
      "grad_norm": 3.192869577046975,
      "language_loss": 0.64870727,
      "learning_rate": 2.495342765579811e-08,
      "loss": 0.67028487,
      "num_input_tokens_seen": 170890260,
      "step": 7910,
      "time_per_iteration": 2.6387102603912354
    },
    {
      "auxiliary_loss_clip": 0.0116543,
      "auxiliary_loss_mlp": 0.01022249,
      "balance_loss_clip": 1.04610169,
      "balance_loss_mlp": 1.01503694,
      "epoch": 0.9512415078458486,
      "flos": 20810521094400.0,
      "grad_norm": 2.359967189661388,
      "language_loss": 0.71366012,
      "learning_rate": 2.4830910429693984e-08,
      "loss": 0.73553693,
      "num_input_tokens_seen": 170910220,
      "step": 7911,
      "time_per_iteration": 2.6932640075683594
    },
    {
      "auxiliary_loss_clip": 0.0112338,
      "auxiliary_loss_mlp": 0.01024848,
      "balance_loss_clip": 1.04023862,
      "balance_loss_mlp": 1.01732576,
      "epoch": 0.9513617507364877,
      "flos": 18369565482240.0,
      "grad_norm": 1.967410368599867,
      "language_loss": 0.79472446,
      "learning_rate": 2.470869283517052e-08,
      "loss": 0.81620669,
      "num_input_tokens_seen": 170928255,
      "step": 7912,
      "time_per_iteration": 2.687831401824951
    },
    {
      "auxiliary_loss_clip": 0.01104457,
      "auxiliary_loss_mlp": 0.01026766,
      "balance_loss_clip": 1.03871214,
      "balance_loss_mlp": 1.01904082,
      "epoch": 0.9514819936271268,
      "flos": 25010166412800.0,
      "grad_norm": 1.8287857705954345,
      "language_loss": 0.77081847,
      "learning_rate": 2.458677489076777e-08,
      "loss": 0.79213059,
      "num_input_tokens_seen": 170949265,
      "step": 7913,
      "time_per_iteration": 2.724043846130371
    },
    {
      "auxiliary_loss_clip": 0.01068787,
      "auxiliary_loss_mlp": 0.01023141,
      "balance_loss_clip": 1.03464115,
      "balance_loss_mlp": 1.0158515,
      "epoch": 0.9516022365177659,
      "flos": 18662129758080.0,
      "grad_norm": 1.7255452473336237,
      "language_loss": 0.83119261,
      "learning_rate": 2.446515661498072e-08,
      "loss": 0.85211194,
      "num_input_tokens_seen": 170968595,
      "step": 7914,
      "time_per_iteration": 2.8229281902313232
    },
    {
      "auxiliary_loss_clip": 0.01121513,
      "auxiliary_loss_mlp": 0.01028439,
      "balance_loss_clip": 1.04214501,
      "balance_loss_mlp": 1.02161729,
      "epoch": 0.9517224794084049,
      "flos": 25372109808000.0,
      "grad_norm": 2.204233610928725,
      "language_loss": 0.7477957,
      "learning_rate": 2.434383802625861e-08,
      "loss": 0.76929522,
      "num_input_tokens_seen": 170987550,
      "step": 7915,
      "time_per_iteration": 3.345874786376953
    },
    {
      "auxiliary_loss_clip": 0.01128444,
      "auxiliary_loss_mlp": 0.01022277,
      "balance_loss_clip": 1.04291439,
      "balance_loss_mlp": 1.01500571,
      "epoch": 0.9518427222990441,
      "flos": 21470918595840.0,
      "grad_norm": 7.592916308233226,
      "language_loss": 0.73864043,
      "learning_rate": 2.4222819143005168e-08,
      "loss": 0.76014769,
      "num_input_tokens_seen": 171007145,
      "step": 7916,
      "time_per_iteration": 4.570230960845947
    },
    {
      "auxiliary_loss_clip": 0.01126282,
      "auxiliary_loss_mlp": 0.01027665,
      "balance_loss_clip": 1.04418516,
      "balance_loss_mlp": 1.02005959,
      "epoch": 0.9519629651896832,
      "flos": 21033634423680.0,
      "grad_norm": 1.881435753355056,
      "language_loss": 0.81341654,
      "learning_rate": 2.4102099983579706e-08,
      "loss": 0.83495599,
      "num_input_tokens_seen": 171026295,
      "step": 7917,
      "time_per_iteration": 2.7723963260650635
    },
    {
      "auxiliary_loss_clip": 0.0114713,
      "auxiliary_loss_mlp": 0.01023214,
      "balance_loss_clip": 1.04546642,
      "balance_loss_mlp": 1.01445246,
      "epoch": 0.9520832080803222,
      "flos": 21689219502720.0,
      "grad_norm": 2.3716508575000126,
      "language_loss": 0.77075952,
      "learning_rate": 2.3981680566294236e-08,
      "loss": 0.79246294,
      "num_input_tokens_seen": 171045895,
      "step": 7918,
      "time_per_iteration": 2.6551172733306885
    },
    {
      "auxiliary_loss_clip": 0.01124836,
      "auxiliary_loss_mlp": 0.0102586,
      "balance_loss_clip": 1.04194808,
      "balance_loss_mlp": 1.01873684,
      "epoch": 0.9522034509709614,
      "flos": 23145289125120.0,
      "grad_norm": 3.4079183906459667,
      "language_loss": 0.73383969,
      "learning_rate": 2.3861560909416822e-08,
      "loss": 0.75534666,
      "num_input_tokens_seen": 171065445,
      "step": 7919,
      "time_per_iteration": 3.6701014041900635
    },
    {
      "auxiliary_loss_clip": 0.01138642,
      "auxiliary_loss_mlp": 0.0102764,
      "balance_loss_clip": 1.04727221,
      "balance_loss_mlp": 1.02061844,
      "epoch": 0.9523236938616004,
      "flos": 24679428958080.0,
      "grad_norm": 1.8265432391535081,
      "language_loss": 0.82578748,
      "learning_rate": 2.3741741031169325e-08,
      "loss": 0.84745032,
      "num_input_tokens_seen": 171085015,
      "step": 7920,
      "time_per_iteration": 2.7059524059295654
    },
    {
      "auxiliary_loss_clip": 0.01102029,
      "auxiliary_loss_mlp": 0.01027244,
      "balance_loss_clip": 1.04038525,
      "balance_loss_mlp": 1.01941824,
      "epoch": 0.9524439367522395,
      "flos": 22672309812480.0,
      "grad_norm": 1.8818034882075354,
      "language_loss": 0.71647763,
      "learning_rate": 2.3622220949728544e-08,
      "loss": 0.73777038,
      "num_input_tokens_seen": 171103900,
      "step": 7921,
      "time_per_iteration": 2.873263120651245
    },
    {
      "auxiliary_loss_clip": 0.01115645,
      "auxiliary_loss_mlp": 0.0076233,
      "balance_loss_clip": 1.04244375,
      "balance_loss_mlp": 1.00049162,
      "epoch": 0.9525641796428787,
      "flos": 34055525024640.0,
      "grad_norm": 3.0858685215658506,
      "language_loss": 0.61371303,
      "learning_rate": 2.3503000683225526e-08,
      "loss": 0.63249278,
      "num_input_tokens_seen": 171121615,
      "step": 7922,
      "time_per_iteration": 2.9232678413391113
    },
    {
      "auxiliary_loss_clip": 0.01088534,
      "auxiliary_loss_mlp": 0.0102787,
      "balance_loss_clip": 1.03696072,
      "balance_loss_mlp": 1.01977611,
      "epoch": 0.9526844225335177,
      "flos": 16727083251840.0,
      "grad_norm": 2.172326230785647,
      "language_loss": 0.84505701,
      "learning_rate": 2.3384080249745585e-08,
      "loss": 0.86622101,
      "num_input_tokens_seen": 171139505,
      "step": 7923,
      "time_per_iteration": 3.6030850410461426
    },
    {
      "auxiliary_loss_clip": 0.01127105,
      "auxiliary_loss_mlp": 0.01023026,
      "balance_loss_clip": 1.04223824,
      "balance_loss_mlp": 1.01571846,
      "epoch": 0.9528046654241568,
      "flos": 36939367330560.0,
      "grad_norm": 2.444320309506902,
      "language_loss": 0.82856059,
      "learning_rate": 2.3265459667329178e-08,
      "loss": 0.85006189,
      "num_input_tokens_seen": 171158995,
      "step": 7924,
      "time_per_iteration": 2.7804901599884033
    },
    {
      "auxiliary_loss_clip": 0.01136483,
      "auxiliary_loss_mlp": 0.01027309,
      "balance_loss_clip": 1.04419029,
      "balance_loss_mlp": 1.01981986,
      "epoch": 0.9529249083147959,
      "flos": 18255010032000.0,
      "grad_norm": 4.049914734720597,
      "language_loss": 0.86711478,
      "learning_rate": 2.31471389539708e-08,
      "loss": 0.88875264,
      "num_input_tokens_seen": 171176120,
      "step": 7925,
      "time_per_iteration": 2.6500000953674316
    },
    {
      "auxiliary_loss_clip": 0.01119239,
      "auxiliary_loss_mlp": 0.01025474,
      "balance_loss_clip": 1.04204607,
      "balance_loss_mlp": 1.01796663,
      "epoch": 0.953045151205435,
      "flos": 28658438985600.0,
      "grad_norm": 3.3168361214988744,
      "language_loss": 0.72952068,
      "learning_rate": 2.3029118127619872e-08,
      "loss": 0.75096774,
      "num_input_tokens_seen": 171195835,
      "step": 7926,
      "time_per_iteration": 2.7633814811706543
    },
    {
      "auxiliary_loss_clip": 0.01096939,
      "auxiliary_loss_mlp": 0.01025004,
      "balance_loss_clip": 1.04068577,
      "balance_loss_mlp": 1.01746392,
      "epoch": 0.953165394096074,
      "flos": 21835232288640.0,
      "grad_norm": 2.8009297171153524,
      "language_loss": 0.8697747,
      "learning_rate": 2.2911397206179628e-08,
      "loss": 0.89099413,
      "num_input_tokens_seen": 171212585,
      "step": 7927,
      "time_per_iteration": 2.7645394802093506
    },
    {
      "auxiliary_loss_clip": 0.01073333,
      "auxiliary_loss_mlp": 0.01025365,
      "balance_loss_clip": 1.03742933,
      "balance_loss_mlp": 1.01835489,
      "epoch": 0.9532856369867132,
      "flos": 19975059682560.0,
      "grad_norm": 3.8331092802446802,
      "language_loss": 0.62592775,
      "learning_rate": 2.279397620750845e-08,
      "loss": 0.64691472,
      "num_input_tokens_seen": 171231630,
      "step": 7928,
      "time_per_iteration": 2.789640426635742
    },
    {
      "auxiliary_loss_clip": 0.01059,
      "auxiliary_loss_mlp": 0.01028532,
      "balance_loss_clip": 1.0317384,
      "balance_loss_mlp": 1.02084327,
      "epoch": 0.9534058798773523,
      "flos": 15049588239360.0,
      "grad_norm": 2.2750370514421543,
      "language_loss": 0.78867698,
      "learning_rate": 2.2676855149419195e-08,
      "loss": 0.80955231,
      "num_input_tokens_seen": 171248800,
      "step": 7929,
      "time_per_iteration": 2.9532525539398193
    },
    {
      "auxiliary_loss_clip": 0.01096234,
      "auxiliary_loss_mlp": 0.00761796,
      "balance_loss_clip": 1.04131448,
      "balance_loss_mlp": 1.00051999,
      "epoch": 0.9535261227679913,
      "flos": 17602800831360.0,
      "grad_norm": 2.466863386910152,
      "language_loss": 0.75170267,
      "learning_rate": 2.2560034049678988e-08,
      "loss": 0.77028298,
      "num_input_tokens_seen": 171263150,
      "step": 7930,
      "time_per_iteration": 3.0106828212738037
    },
    {
      "auxiliary_loss_clip": 0.01139332,
      "auxiliary_loss_mlp": 0.01028291,
      "balance_loss_clip": 1.04381895,
      "balance_loss_mlp": 1.02012491,
      "epoch": 0.9536463656586305,
      "flos": 23142954741120.0,
      "grad_norm": 1.6837641648727733,
      "language_loss": 0.75246692,
      "learning_rate": 2.2443512926008988e-08,
      "loss": 0.77414316,
      "num_input_tokens_seen": 171282480,
      "step": 7931,
      "time_per_iteration": 2.748466730117798
    },
    {
      "auxiliary_loss_clip": 0.01109831,
      "auxiliary_loss_mlp": 0.01026066,
      "balance_loss_clip": 1.03890681,
      "balance_loss_mlp": 1.01843023,
      "epoch": 0.9537666085492695,
      "flos": 18625033987200.0,
      "grad_norm": 2.3560378574845573,
      "language_loss": 0.70221496,
      "learning_rate": 2.2327291796085946e-08,
      "loss": 0.72357386,
      "num_input_tokens_seen": 171300840,
      "step": 7932,
      "time_per_iteration": 2.689805507659912
    },
    {
      "auxiliary_loss_clip": 0.01147895,
      "auxiliary_loss_mlp": 0.01024615,
      "balance_loss_clip": 1.04477274,
      "balance_loss_mlp": 1.01697922,
      "epoch": 0.9538868514399086,
      "flos": 18989347680000.0,
      "grad_norm": 2.542858256581709,
      "language_loss": 0.77756417,
      "learning_rate": 2.2211370677540197e-08,
      "loss": 0.79928923,
      "num_input_tokens_seen": 171317365,
      "step": 7933,
      "time_per_iteration": 2.748317003250122
    },
    {
      "auxiliary_loss_clip": 0.01063688,
      "auxiliary_loss_mlp": 0.00762474,
      "balance_loss_clip": 1.03688252,
      "balance_loss_mlp": 1.00040102,
      "epoch": 0.9540070943305478,
      "flos": 16800556521600.0,
      "grad_norm": 2.3258763628729286,
      "language_loss": 0.78554577,
      "learning_rate": 2.2095749587957012e-08,
      "loss": 0.80380738,
      "num_input_tokens_seen": 171335270,
      "step": 7934,
      "time_per_iteration": 2.8218793869018555
    },
    {
      "auxiliary_loss_clip": 0.01096102,
      "auxiliary_loss_mlp": 0.01030613,
      "balance_loss_clip": 1.04207349,
      "balance_loss_mlp": 1.02261448,
      "epoch": 0.9541273372211868,
      "flos": 20156911263360.0,
      "grad_norm": 1.765811772682769,
      "language_loss": 0.69874644,
      "learning_rate": 2.1980428544876138e-08,
      "loss": 0.72001356,
      "num_input_tokens_seen": 171353910,
      "step": 7935,
      "time_per_iteration": 2.7117397785186768
    },
    {
      "auxiliary_loss_clip": 0.01087807,
      "auxiliary_loss_mlp": 0.01023696,
      "balance_loss_clip": 1.03560448,
      "balance_loss_mlp": 1.01635265,
      "epoch": 0.9542475801118259,
      "flos": 26725511381760.0,
      "grad_norm": 20.591608285744627,
      "language_loss": 0.74302673,
      "learning_rate": 2.1865407565791584e-08,
      "loss": 0.76414174,
      "num_input_tokens_seen": 171375480,
      "step": 7936,
      "time_per_iteration": 2.8714377880096436
    },
    {
      "auxiliary_loss_clip": 0.01134805,
      "auxiliary_loss_mlp": 0.01025786,
      "balance_loss_clip": 1.04143715,
      "balance_loss_mlp": 1.01847553,
      "epoch": 0.954367823002465,
      "flos": 23330911633920.0,
      "grad_norm": 1.9900047522051736,
      "language_loss": 0.7743175,
      "learning_rate": 2.175068666815183e-08,
      "loss": 0.79592341,
      "num_input_tokens_seen": 171396320,
      "step": 7937,
      "time_per_iteration": 2.658999443054199
    },
    {
      "auxiliary_loss_clip": 0.01137418,
      "auxiliary_loss_mlp": 0.01023523,
      "balance_loss_clip": 1.0468111,
      "balance_loss_mlp": 1.01590252,
      "epoch": 0.9544880658931041,
      "flos": 14902713527040.0,
      "grad_norm": 2.4131960004228117,
      "language_loss": 0.79047048,
      "learning_rate": 2.163626586935985e-08,
      "loss": 0.81207985,
      "num_input_tokens_seen": 171412860,
      "step": 7938,
      "time_per_iteration": 2.704216957092285
    },
    {
      "auxiliary_loss_clip": 0.01109736,
      "auxiliary_loss_mlp": 0.01025364,
      "balance_loss_clip": 1.04083133,
      "balance_loss_mlp": 1.01774645,
      "epoch": 0.9546083087837431,
      "flos": 29095902725760.0,
      "grad_norm": 3.3131923093532105,
      "language_loss": 0.6319198,
      "learning_rate": 2.1522145186773755e-08,
      "loss": 0.65327078,
      "num_input_tokens_seen": 171431780,
      "step": 7939,
      "time_per_iteration": 2.7984414100646973
    },
    {
      "auxiliary_loss_clip": 0.01092693,
      "auxiliary_loss_mlp": 0.01026881,
      "balance_loss_clip": 1.0389756,
      "balance_loss_mlp": 1.01991034,
      "epoch": 0.9547285516743822,
      "flos": 21142335957120.0,
      "grad_norm": 2.1860859946130455,
      "language_loss": 0.85902226,
      "learning_rate": 2.140832463770481e-08,
      "loss": 0.88021803,
      "num_input_tokens_seen": 171450975,
      "step": 7940,
      "time_per_iteration": 2.7835936546325684
    },
    {
      "auxiliary_loss_clip": 0.01108444,
      "auxiliary_loss_mlp": 0.01034726,
      "balance_loss_clip": 1.04575372,
      "balance_loss_mlp": 1.02676845,
      "epoch": 0.9548487945650214,
      "flos": 27490157130240.0,
      "grad_norm": 1.9901336301551287,
      "language_loss": 0.75785834,
      "learning_rate": 2.129480423941987e-08,
      "loss": 0.77928996,
      "num_input_tokens_seen": 171467645,
      "step": 7941,
      "time_per_iteration": 2.754699468612671
    },
    {
      "auxiliary_loss_clip": 0.01138174,
      "auxiliary_loss_mlp": 0.01022744,
      "balance_loss_clip": 1.04513669,
      "balance_loss_mlp": 1.01499844,
      "epoch": 0.9549690374556604,
      "flos": 22273198819200.0,
      "grad_norm": 1.7671410707952475,
      "language_loss": 0.8040266,
      "learning_rate": 2.1181584009140052e-08,
      "loss": 0.82563567,
      "num_input_tokens_seen": 171487185,
      "step": 7942,
      "time_per_iteration": 5.253152847290039
    },
    {
      "auxiliary_loss_clip": 0.01082015,
      "auxiliary_loss_mlp": 0.01016561,
      "balance_loss_clip": 1.03387225,
      "balance_loss_mlp": 1.00905359,
      "epoch": 0.9550892803462995,
      "flos": 17595294888960.0,
      "grad_norm": 5.04380379625138,
      "language_loss": 0.84140182,
      "learning_rate": 2.10686639640405e-08,
      "loss": 0.86238754,
      "num_input_tokens_seen": 171501275,
      "step": 7943,
      "time_per_iteration": 2.9845738410949707
    },
    {
      "auxiliary_loss_clip": 0.01138753,
      "auxiliary_loss_mlp": 0.01023463,
      "balance_loss_clip": 1.04198575,
      "balance_loss_mlp": 1.01616764,
      "epoch": 0.9552095232369386,
      "flos": 24353144789760.0,
      "grad_norm": 1.7445358111702436,
      "language_loss": 0.81143671,
      "learning_rate": 2.0956044121251294e-08,
      "loss": 0.83305889,
      "num_input_tokens_seen": 171520060,
      "step": 7944,
      "time_per_iteration": 2.7053773403167725
    },
    {
      "auxiliary_loss_clip": 0.01123415,
      "auxiliary_loss_mlp": 0.01028802,
      "balance_loss_clip": 1.04076993,
      "balance_loss_mlp": 1.020576,
      "epoch": 0.9553297661275777,
      "flos": 22746860490240.0,
      "grad_norm": 1.6661212950134319,
      "language_loss": 0.80892342,
      "learning_rate": 2.084372449785654e-08,
      "loss": 0.83044553,
      "num_input_tokens_seen": 171539895,
      "step": 7945,
      "time_per_iteration": 3.662910223007202
    },
    {
      "auxiliary_loss_clip": 0.01156207,
      "auxiliary_loss_mlp": 0.01025662,
      "balance_loss_clip": 1.04787469,
      "balance_loss_mlp": 1.01852143,
      "epoch": 0.9554500090182168,
      "flos": 15413866018560.0,
      "grad_norm": 4.59318039176615,
      "language_loss": 0.68867034,
      "learning_rate": 2.0731705110895282e-08,
      "loss": 0.71048903,
      "num_input_tokens_seen": 171557385,
      "step": 7946,
      "time_per_iteration": 2.6491591930389404
    },
    {
      "auxiliary_loss_clip": 0.0110775,
      "auxiliary_loss_mlp": 0.01028853,
      "balance_loss_clip": 1.04117656,
      "balance_loss_mlp": 1.02076495,
      "epoch": 0.9555702519088559,
      "flos": 23513517400320.0,
      "grad_norm": 1.7536091348496643,
      "language_loss": 0.86681736,
      "learning_rate": 2.0619985977360587e-08,
      "loss": 0.88818336,
      "num_input_tokens_seen": 171575705,
      "step": 7947,
      "time_per_iteration": 2.8062825202941895
    },
    {
      "auxiliary_loss_clip": 0.01167078,
      "auxiliary_loss_mlp": 0.01018717,
      "balance_loss_clip": 1.04801416,
      "balance_loss_mlp": 1.01232111,
      "epoch": 0.955690494799495,
      "flos": 22962072827520.0,
      "grad_norm": 1.8669146431251673,
      "language_loss": 0.77014172,
      "learning_rate": 2.0508567114200237e-08,
      "loss": 0.79199958,
      "num_input_tokens_seen": 171595620,
      "step": 7948,
      "time_per_iteration": 2.638887405395508
    },
    {
      "auxiliary_loss_clip": 0.01126703,
      "auxiliary_loss_mlp": 0.01024241,
      "balance_loss_clip": 1.04370522,
      "balance_loss_mlp": 1.01692188,
      "epoch": 0.955810737690134,
      "flos": 26031250333440.0,
      "grad_norm": 2.204102249095545,
      "language_loss": 0.78675216,
      "learning_rate": 2.0397448538316485e-08,
      "loss": 0.80826157,
      "num_input_tokens_seen": 171616660,
      "step": 7949,
      "time_per_iteration": 3.744006872177124
    },
    {
      "auxiliary_loss_clip": 0.01112317,
      "auxiliary_loss_mlp": 0.01023988,
      "balance_loss_clip": 1.03925443,
      "balance_loss_mlp": 1.01645386,
      "epoch": 0.9559309805807732,
      "flos": 20849951249280.0,
      "grad_norm": 5.682294287814617,
      "language_loss": 0.66541481,
      "learning_rate": 2.028663026656563e-08,
      "loss": 0.68677789,
      "num_input_tokens_seen": 171635515,
      "step": 7950,
      "time_per_iteration": 2.7885684967041016
    },
    {
      "auxiliary_loss_clip": 0.0113791,
      "auxiliary_loss_mlp": 0.01022631,
      "balance_loss_clip": 1.0419271,
      "balance_loss_mlp": 1.01525807,
      "epoch": 0.9560512234714122,
      "flos": 21578219498880.0,
      "grad_norm": 1.9382104443235169,
      "language_loss": 0.71767652,
      "learning_rate": 2.0176112315758885e-08,
      "loss": 0.73928189,
      "num_input_tokens_seen": 171653305,
      "step": 7951,
      "time_per_iteration": 2.6981167793273926
    },
    {
      "auxiliary_loss_clip": 0.01121705,
      "auxiliary_loss_mlp": 0.01025144,
      "balance_loss_clip": 1.04269934,
      "balance_loss_mlp": 1.01714468,
      "epoch": 0.9561714663620513,
      "flos": 17450144029440.0,
      "grad_norm": 2.3893320999509684,
      "language_loss": 0.69122708,
      "learning_rate": 2.0065894702661957e-08,
      "loss": 0.71269554,
      "num_input_tokens_seen": 171669980,
      "step": 7952,
      "time_per_iteration": 2.6432344913482666
    },
    {
      "auxiliary_loss_clip": 0.01108275,
      "auxiliary_loss_mlp": 0.0076285,
      "balance_loss_clip": 1.04184055,
      "balance_loss_mlp": 1.00038695,
      "epoch": 0.9562917092526905,
      "flos": 26098510550400.0,
      "grad_norm": 2.1247892818081313,
      "language_loss": 0.78012252,
      "learning_rate": 1.9955977443994577e-08,
      "loss": 0.79883373,
      "num_input_tokens_seen": 171689970,
      "step": 7953,
      "time_per_iteration": 2.7798683643341064
    },
    {
      "auxiliary_loss_clip": 0.01163739,
      "auxiliary_loss_mlp": 0.01027336,
      "balance_loss_clip": 1.04918861,
      "balance_loss_mlp": 1.01875281,
      "epoch": 0.9564119521433295,
      "flos": 24096742531200.0,
      "grad_norm": 2.179043502528967,
      "language_loss": 0.62320811,
      "learning_rate": 1.9846360556430965e-08,
      "loss": 0.64511883,
      "num_input_tokens_seen": 171708270,
      "step": 7954,
      "time_per_iteration": 2.7088990211486816
    },
    {
      "auxiliary_loss_clip": 0.01139738,
      "auxiliary_loss_mlp": 0.01019442,
      "balance_loss_clip": 1.0418191,
      "balance_loss_mlp": 1.01213706,
      "epoch": 0.9565321950339686,
      "flos": 32008903896960.0,
      "grad_norm": 5.977623600782517,
      "language_loss": 0.61181968,
      "learning_rate": 1.973704405660004e-08,
      "loss": 0.63341147,
      "num_input_tokens_seen": 171729385,
      "step": 7955,
      "time_per_iteration": 2.7530412673950195
    },
    {
      "auxiliary_loss_clip": 0.01085718,
      "auxiliary_loss_mlp": 0.01026578,
      "balance_loss_clip": 1.03399694,
      "balance_loss_mlp": 1.01919293,
      "epoch": 0.9566524379246077,
      "flos": 23588642695680.0,
      "grad_norm": 1.5525208629408596,
      "language_loss": 0.7824856,
      "learning_rate": 1.9628027961085203e-08,
      "loss": 0.80360854,
      "num_input_tokens_seen": 171752615,
      "step": 7956,
      "time_per_iteration": 2.8410871028900146
    },
    {
      "auxiliary_loss_clip": 0.01106151,
      "auxiliary_loss_mlp": 0.01022013,
      "balance_loss_clip": 1.04152083,
      "balance_loss_mlp": 1.0141151,
      "epoch": 0.9567726808152468,
      "flos": 38067716240640.0,
      "grad_norm": 2.9995718297336493,
      "language_loss": 0.83535016,
      "learning_rate": 1.9519312286423894e-08,
      "loss": 0.85663176,
      "num_input_tokens_seen": 171775810,
      "step": 7957,
      "time_per_iteration": 2.860395908355713
    },
    {
      "auxiliary_loss_clip": 0.01106257,
      "auxiliary_loss_mlp": 0.01023014,
      "balance_loss_clip": 1.04194379,
      "balance_loss_mlp": 1.01536131,
      "epoch": 0.9568929237058859,
      "flos": 22744059229440.0,
      "grad_norm": 1.7589250825585154,
      "language_loss": 0.77484691,
      "learning_rate": 1.9410897049108255e-08,
      "loss": 0.7961396,
      "num_input_tokens_seen": 171795090,
      "step": 7958,
      "time_per_iteration": 2.768087148666382
    },
    {
      "auxiliary_loss_clip": 0.01108324,
      "auxiliary_loss_mlp": 0.01026043,
      "balance_loss_clip": 1.04281831,
      "balance_loss_mlp": 1.0171442,
      "epoch": 0.957013166596525,
      "flos": 23841633162240.0,
      "grad_norm": 2.1059107690288554,
      "language_loss": 0.91353518,
      "learning_rate": 1.9302782265584905e-08,
      "loss": 0.93487883,
      "num_input_tokens_seen": 171815755,
      "step": 7959,
      "time_per_iteration": 2.773733615875244
    },
    {
      "auxiliary_loss_clip": 0.0115226,
      "auxiliary_loss_mlp": 0.01020116,
      "balance_loss_clip": 1.0441047,
      "balance_loss_mlp": 1.01304114,
      "epoch": 0.9571334094871641,
      "flos": 17639286071040.0,
      "grad_norm": 3.370193082253107,
      "language_loss": 0.87088513,
      "learning_rate": 1.9194967952254282e-08,
      "loss": 0.89260888,
      "num_input_tokens_seen": 171834330,
      "step": 7960,
      "time_per_iteration": 2.6634631156921387
    },
    {
      "auxiliary_loss_clip": 0.01088243,
      "auxiliary_loss_mlp": 0.01020735,
      "balance_loss_clip": 1.04091847,
      "balance_loss_mlp": 1.01339173,
      "epoch": 0.9572536523778031,
      "flos": 15369623441280.0,
      "grad_norm": 4.394488912901993,
      "language_loss": 0.8098219,
      "learning_rate": 1.9087454125472635e-08,
      "loss": 0.83091176,
      "num_input_tokens_seen": 171848805,
      "step": 7961,
      "time_per_iteration": 2.94343638420105
    },
    {
      "auxiliary_loss_clip": 0.0108516,
      "auxiliary_loss_mlp": 0.01030991,
      "balance_loss_clip": 1.03848386,
      "balance_loss_mlp": 1.02295041,
      "epoch": 0.9573738952684423,
      "flos": 24969838417920.0,
      "grad_norm": 1.8952557170316464,
      "language_loss": 0.78652066,
      "learning_rate": 1.8980240801548696e-08,
      "loss": 0.80768216,
      "num_input_tokens_seen": 171867995,
      "step": 7962,
      "time_per_iteration": 3.257641553878784
    },
    {
      "auxiliary_loss_clip": 0.01170346,
      "auxiliary_loss_mlp": 0.01022703,
      "balance_loss_clip": 1.04913044,
      "balance_loss_mlp": 1.01505589,
      "epoch": 0.9574941381590814,
      "flos": 25769461034880.0,
      "grad_norm": 1.8700629955652055,
      "language_loss": 0.74105209,
      "learning_rate": 1.8873327996747458e-08,
      "loss": 0.76298261,
      "num_input_tokens_seen": 171886495,
      "step": 7963,
      "time_per_iteration": 3.3988606929779053
    },
    {
      "auxiliary_loss_clip": 0.01091039,
      "auxiliary_loss_mlp": 0.00761989,
      "balance_loss_clip": 1.03682017,
      "balance_loss_mlp": 1.00042367,
      "epoch": 0.9576143810497204,
      "flos": 32307178435200.0,
      "grad_norm": 2.0765145445587656,
      "language_loss": 0.66318274,
      "learning_rate": 1.8766715727287053e-08,
      "loss": 0.68171299,
      "num_input_tokens_seen": 171908200,
      "step": 7964,
      "time_per_iteration": 2.955719232559204
    },
    {
      "auxiliary_loss_clip": 0.01112156,
      "auxiliary_loss_mlp": 0.00762126,
      "balance_loss_clip": 1.04234719,
      "balance_loss_mlp": 1.00045395,
      "epoch": 0.9577346239403596,
      "flos": 27745733376000.0,
      "grad_norm": 2.0584443451530547,
      "language_loss": 0.79396534,
      "learning_rate": 1.8660404009340546e-08,
      "loss": 0.81270814,
      "num_input_tokens_seen": 171928650,
      "step": 7965,
      "time_per_iteration": 2.913443088531494
    },
    {
      "auxiliary_loss_clip": 0.01015635,
      "auxiliary_loss_mlp": 0.01000837,
      "balance_loss_clip": 1.00951362,
      "balance_loss_mlp": 0.99982375,
      "epoch": 0.9578548668309986,
      "flos": 57468313710720.0,
      "grad_norm": 0.8774486367780325,
      "language_loss": 0.59563303,
      "learning_rate": 1.8554392859035485e-08,
      "loss": 0.61579776,
      "num_input_tokens_seen": 171986400,
      "step": 7966,
      "time_per_iteration": 3.3471975326538086
    },
    {
      "auxiliary_loss_clip": 0.01115546,
      "auxiliary_loss_mlp": 0.01025169,
      "balance_loss_clip": 1.0401051,
      "balance_loss_mlp": 1.01672292,
      "epoch": 0.9579751097216377,
      "flos": 19756040503680.0,
      "grad_norm": 1.7825931783419668,
      "language_loss": 0.79184103,
      "learning_rate": 1.8448682292453444e-08,
      "loss": 0.81324822,
      "num_input_tokens_seen": 172005475,
      "step": 7967,
      "time_per_iteration": 2.7345080375671387
    },
    {
      "auxiliary_loss_clip": 0.01129752,
      "auxiliary_loss_mlp": 0.00762366,
      "balance_loss_clip": 1.0466553,
      "balance_loss_mlp": 1.00055122,
      "epoch": 0.9580953526122769,
      "flos": 18041270152320.0,
      "grad_norm": 1.8008175379613605,
      "language_loss": 0.65915513,
      "learning_rate": 1.8343272325631154e-08,
      "loss": 0.67807633,
      "num_input_tokens_seen": 172024420,
      "step": 7968,
      "time_per_iteration": 5.162299394607544
    },
    {
      "auxiliary_loss_clip": 0.01114528,
      "auxiliary_loss_mlp": 0.01024862,
      "balance_loss_clip": 1.04207218,
      "balance_loss_mlp": 1.01759028,
      "epoch": 0.9582155955029159,
      "flos": 24270154416000.0,
      "grad_norm": 4.145658828320684,
      "language_loss": 0.78344727,
      "learning_rate": 1.8238162974558492e-08,
      "loss": 0.8048411,
      "num_input_tokens_seen": 172038350,
      "step": 7969,
      "time_per_iteration": 2.8118860721588135
    },
    {
      "auxiliary_loss_clip": 0.01152754,
      "auxiliary_loss_mlp": 0.01029492,
      "balance_loss_clip": 1.04816103,
      "balance_loss_mlp": 1.02244961,
      "epoch": 0.958335838393555,
      "flos": 22783309816320.0,
      "grad_norm": 2.5188101171769968,
      "language_loss": 0.74851787,
      "learning_rate": 1.8133354255181144e-08,
      "loss": 0.77034032,
      "num_input_tokens_seen": 172058665,
      "step": 7970,
      "time_per_iteration": 2.656244993209839
    },
    {
      "auxiliary_loss_clip": 0.0113201,
      "auxiliary_loss_mlp": 0.01025276,
      "balance_loss_clip": 1.044011,
      "balance_loss_mlp": 1.01758742,
      "epoch": 0.958456081284194,
      "flos": 16911484698240.0,
      "grad_norm": 3.9430999408118215,
      "language_loss": 0.74581313,
      "learning_rate": 1.802884618339795e-08,
      "loss": 0.76738608,
      "num_input_tokens_seen": 172077470,
      "step": 7971,
      "time_per_iteration": 3.572366952896118
    },
    {
      "auxiliary_loss_clip": 0.01103356,
      "auxiliary_loss_mlp": 0.01027172,
      "balance_loss_clip": 1.04219484,
      "balance_loss_mlp": 1.01944757,
      "epoch": 0.9585763241748332,
      "flos": 19974951941760.0,
      "grad_norm": 2.4250758624046824,
      "language_loss": 0.81309235,
      "learning_rate": 1.7924638775062894e-08,
      "loss": 0.83439755,
      "num_input_tokens_seen": 172096590,
      "step": 7972,
      "time_per_iteration": 2.757429599761963
    },
    {
      "auxiliary_loss_clip": 0.01101232,
      "auxiliary_loss_mlp": 0.01022252,
      "balance_loss_clip": 1.0372895,
      "balance_loss_mlp": 1.01451492,
      "epoch": 0.9586965670654722,
      "flos": 21395649646080.0,
      "grad_norm": 1.7796179607273126,
      "language_loss": 0.82143915,
      "learning_rate": 1.7820732045984444e-08,
      "loss": 0.84267402,
      "num_input_tokens_seen": 172116735,
      "step": 7973,
      "time_per_iteration": 2.810936450958252
    },
    {
      "auxiliary_loss_clip": 0.01133036,
      "auxiliary_loss_mlp": 0.01023963,
      "balance_loss_clip": 1.04402041,
      "balance_loss_mlp": 1.01647639,
      "epoch": 0.9588168099561113,
      "flos": 21435115714560.0,
      "grad_norm": 3.5856694264998423,
      "language_loss": 0.74229932,
      "learning_rate": 1.7717126011924655e-08,
      "loss": 0.76386929,
      "num_input_tokens_seen": 172138320,
      "step": 7974,
      "time_per_iteration": 2.8430428504943848
    },
    {
      "auxiliary_loss_clip": 0.01120513,
      "auxiliary_loss_mlp": 0.01026086,
      "balance_loss_clip": 1.03928483,
      "balance_loss_mlp": 1.01856947,
      "epoch": 0.9589370528467505,
      "flos": 11763761852160.0,
      "grad_norm": 3.0787119669118757,
      "language_loss": 0.76250184,
      "learning_rate": 1.7613820688600957e-08,
      "loss": 0.78396785,
      "num_input_tokens_seen": 172154225,
      "step": 7975,
      "time_per_iteration": 3.690044641494751
    },
    {
      "auxiliary_loss_clip": 0.01155385,
      "auxiliary_loss_mlp": 0.01029897,
      "balance_loss_clip": 1.04622388,
      "balance_loss_mlp": 1.02255321,
      "epoch": 0.9590572957373895,
      "flos": 23441516588160.0,
      "grad_norm": 1.9973116295772948,
      "language_loss": 0.78495133,
      "learning_rate": 1.7510816091684588e-08,
      "loss": 0.80680418,
      "num_input_tokens_seen": 172174150,
      "step": 7976,
      "time_per_iteration": 2.696505308151245
    },
    {
      "auxiliary_loss_clip": 0.01132518,
      "auxiliary_loss_mlp": 0.01022939,
      "balance_loss_clip": 1.04190636,
      "balance_loss_mlp": 1.0143857,
      "epoch": 0.9591775386280286,
      "flos": 22528272274560.0,
      "grad_norm": 2.803286753818879,
      "language_loss": 0.79216611,
      "learning_rate": 1.740811223680083e-08,
      "loss": 0.8137207,
      "num_input_tokens_seen": 172191005,
      "step": 7977,
      "time_per_iteration": 2.755993127822876
    },
    {
      "auxiliary_loss_clip": 0.01076473,
      "auxiliary_loss_mlp": 0.01021949,
      "balance_loss_clip": 1.03971887,
      "balance_loss_mlp": 1.01469564,
      "epoch": 0.9592977815186677,
      "flos": 18186959715840.0,
      "grad_norm": 2.056055832888113,
      "language_loss": 0.74177361,
      "learning_rate": 1.7305709139530334e-08,
      "loss": 0.76275778,
      "num_input_tokens_seen": 172209785,
      "step": 7978,
      "time_per_iteration": 2.826500177383423
    },
    {
      "auxiliary_loss_clip": 0.01134894,
      "auxiliary_loss_mlp": 0.01022586,
      "balance_loss_clip": 1.04153371,
      "balance_loss_mlp": 1.0146172,
      "epoch": 0.9594180244093068,
      "flos": 16537797555840.0,
      "grad_norm": 2.415051972048675,
      "language_loss": 0.74323475,
      "learning_rate": 1.7203606815407334e-08,
      "loss": 0.76480961,
      "num_input_tokens_seen": 172224380,
      "step": 7979,
      "time_per_iteration": 2.6885757446289062
    },
    {
      "auxiliary_loss_clip": 0.01141384,
      "auxiliary_loss_mlp": 0.01022958,
      "balance_loss_clip": 1.04836631,
      "balance_loss_mlp": 1.01541829,
      "epoch": 0.9595382672999458,
      "flos": 20554334317440.0,
      "grad_norm": 18.31167207003161,
      "language_loss": 0.79758167,
      "learning_rate": 1.7101805279920557e-08,
      "loss": 0.81922507,
      "num_input_tokens_seen": 172242540,
      "step": 7980,
      "time_per_iteration": 2.6544456481933594
    },
    {
      "auxiliary_loss_clip": 0.01131408,
      "auxiliary_loss_mlp": 0.01024844,
      "balance_loss_clip": 1.0420233,
      "balance_loss_mlp": 1.01652932,
      "epoch": 0.959658510190585,
      "flos": 22638266697600.0,
      "grad_norm": 3.4665663716331823,
      "language_loss": 0.80930483,
      "learning_rate": 1.7000304548513643e-08,
      "loss": 0.83086741,
      "num_input_tokens_seen": 172262645,
      "step": 7981,
      "time_per_iteration": 2.75713849067688
    },
    {
      "auxiliary_loss_clip": 0.01071282,
      "auxiliary_loss_mlp": 0.01024953,
      "balance_loss_clip": 1.03421688,
      "balance_loss_mlp": 1.01690078,
      "epoch": 0.9597787530812241,
      "flos": 19135252725120.0,
      "grad_norm": 2.3190288593335753,
      "language_loss": 0.83027327,
      "learning_rate": 1.6899104636583394e-08,
      "loss": 0.85123563,
      "num_input_tokens_seen": 172280695,
      "step": 7982,
      "time_per_iteration": 2.8475165367126465
    },
    {
      "auxiliary_loss_clip": 0.01024845,
      "auxiliary_loss_mlp": 0.01000731,
      "balance_loss_clip": 1.00971365,
      "balance_loss_mlp": 0.9998011,
      "epoch": 0.9598989959718631,
      "flos": 60098124055680.0,
      "grad_norm": 0.7925091440540193,
      "language_loss": 0.61887419,
      "learning_rate": 1.6798205559482638e-08,
      "loss": 0.63913,
      "num_input_tokens_seen": 172343075,
      "step": 7983,
      "time_per_iteration": 3.712358236312866
    },
    {
      "auxiliary_loss_clip": 0.01120878,
      "auxiliary_loss_mlp": 0.01024768,
      "balance_loss_clip": 1.04039061,
      "balance_loss_mlp": 1.01696837,
      "epoch": 0.9600192388625023,
      "flos": 20886795624960.0,
      "grad_norm": 6.199573226282304,
      "language_loss": 0.76747644,
      "learning_rate": 1.669760733251713e-08,
      "loss": 0.78893292,
      "num_input_tokens_seen": 172361950,
      "step": 7984,
      "time_per_iteration": 2.723687171936035
    },
    {
      "auxiliary_loss_clip": 0.01085386,
      "auxiliary_loss_mlp": 0.00761363,
      "balance_loss_clip": 1.03731859,
      "balance_loss_mlp": 1.00038481,
      "epoch": 0.9601394817531413,
      "flos": 20445740524800.0,
      "grad_norm": 2.7471454833315625,
      "language_loss": 0.82647157,
      "learning_rate": 1.659730997094755e-08,
      "loss": 0.84493911,
      "num_input_tokens_seen": 172380440,
      "step": 7985,
      "time_per_iteration": 2.8551571369171143
    },
    {
      "auxiliary_loss_clip": 0.01111971,
      "auxiliary_loss_mlp": 0.01022607,
      "balance_loss_clip": 1.03943062,
      "balance_loss_mlp": 1.01565409,
      "epoch": 0.9602597246437804,
      "flos": 21507152440320.0,
      "grad_norm": 2.2214128644977227,
      "language_loss": 0.62434447,
      "learning_rate": 1.6497313489989283e-08,
      "loss": 0.64569026,
      "num_input_tokens_seen": 172400265,
      "step": 7986,
      "time_per_iteration": 2.8197619915008545
    },
    {
      "auxiliary_loss_clip": 0.01090496,
      "auxiliary_loss_mlp": 0.01024352,
      "balance_loss_clip": 1.03505754,
      "balance_loss_mlp": 1.01679409,
      "epoch": 0.9603799675344196,
      "flos": 29935099152000.0,
      "grad_norm": 2.10364988491677,
      "language_loss": 0.69806117,
      "learning_rate": 1.639761790481131e-08,
      "loss": 0.71920967,
      "num_input_tokens_seen": 172421145,
      "step": 7987,
      "time_per_iteration": 2.8990492820739746
    },
    {
      "auxiliary_loss_clip": 0.01124283,
      "auxiliary_loss_mlp": 0.0102295,
      "balance_loss_clip": 1.04317474,
      "balance_loss_mlp": 1.0158813,
      "epoch": 0.9605002104250586,
      "flos": 28001525103360.0,
      "grad_norm": 2.2262069843492163,
      "language_loss": 0.79447329,
      "learning_rate": 1.6298223230537754e-08,
      "loss": 0.81594563,
      "num_input_tokens_seen": 172438945,
      "step": 7988,
      "time_per_iteration": 2.870847463607788
    },
    {
      "auxiliary_loss_clip": 0.0115544,
      "auxiliary_loss_mlp": 0.01021863,
      "balance_loss_clip": 1.04636312,
      "balance_loss_mlp": 1.01423621,
      "epoch": 0.9606204533156977,
      "flos": 35590490870400.0,
      "grad_norm": 2.430090941442822,
      "language_loss": 0.69829172,
      "learning_rate": 1.619912948224611e-08,
      "loss": 0.72006476,
      "num_input_tokens_seen": 172460150,
      "step": 7989,
      "time_per_iteration": 2.7971391677856445
    },
    {
      "auxiliary_loss_clip": 0.01101986,
      "auxiliary_loss_mlp": 0.01024208,
      "balance_loss_clip": 1.04227543,
      "balance_loss_mlp": 1.01622653,
      "epoch": 0.9607406962063368,
      "flos": 26574614346240.0,
      "grad_norm": 2.615209795436606,
      "language_loss": 0.60873914,
      "learning_rate": 1.6100336674969682e-08,
      "loss": 0.63000113,
      "num_input_tokens_seen": 172478990,
      "step": 7990,
      "time_per_iteration": 2.802196502685547
    },
    {
      "auxiliary_loss_clip": 0.01069182,
      "auxiliary_loss_mlp": 0.00761773,
      "balance_loss_clip": 1.03778625,
      "balance_loss_mlp": 1.00044763,
      "epoch": 0.9608609390969759,
      "flos": 25331781813120.0,
      "grad_norm": 2.897018712411053,
      "language_loss": 0.76235908,
      "learning_rate": 1.600184482369449e-08,
      "loss": 0.78066862,
      "num_input_tokens_seen": 172498905,
      "step": 7991,
      "time_per_iteration": 2.8633852005004883
    },
    {
      "auxiliary_loss_clip": 0.01136455,
      "auxiliary_loss_mlp": 0.00762516,
      "balance_loss_clip": 1.04481125,
      "balance_loss_mlp": 1.00041819,
      "epoch": 0.960981181987615,
      "flos": 21069114082560.0,
      "grad_norm": 2.729814013838655,
      "language_loss": 0.89385337,
      "learning_rate": 1.5903653943362126e-08,
      "loss": 0.91284311,
      "num_input_tokens_seen": 172517900,
      "step": 7992,
      "time_per_iteration": 2.7481515407562256
    },
    {
      "auxiliary_loss_clip": 0.01114118,
      "auxiliary_loss_mlp": 0.01016456,
      "balance_loss_clip": 1.03786969,
      "balance_loss_mlp": 1.00880325,
      "epoch": 0.9611014248782541,
      "flos": 17823256554240.0,
      "grad_norm": 2.16105086030297,
      "language_loss": 0.76762468,
      "learning_rate": 1.580576404886802e-08,
      "loss": 0.78893042,
      "num_input_tokens_seen": 172536430,
      "step": 7993,
      "time_per_iteration": 3.6688265800476074
    },
    {
      "auxiliary_loss_clip": 0.01108331,
      "auxiliary_loss_mlp": 0.01024263,
      "balance_loss_clip": 1.04088211,
      "balance_loss_mlp": 1.01728058,
      "epoch": 0.9612216677688932,
      "flos": 19354631040000.0,
      "grad_norm": 2.699887322674217,
      "language_loss": 0.79765713,
      "learning_rate": 1.570817515506162e-08,
      "loss": 0.81898308,
      "num_input_tokens_seen": 172555120,
      "step": 7994,
      "time_per_iteration": 3.9050934314727783
    },
    {
      "auxiliary_loss_clip": 0.01139601,
      "auxiliary_loss_mlp": 0.00761507,
      "balance_loss_clip": 1.04445159,
      "balance_loss_mlp": 1.00043583,
      "epoch": 0.9613419106595322,
      "flos": 15808739207040.0,
      "grad_norm": 2.030842381203786,
      "language_loss": 0.81381267,
      "learning_rate": 1.561088727674753e-08,
      "loss": 0.83282375,
      "num_input_tokens_seen": 172569330,
      "step": 7995,
      "time_per_iteration": 2.6803815364837646
    },
    {
      "auxiliary_loss_clip": 0.01144732,
      "auxiliary_loss_mlp": 0.01026378,
      "balance_loss_clip": 1.04486299,
      "balance_loss_mlp": 1.01771188,
      "epoch": 0.9614621535501714,
      "flos": 25702488126720.0,
      "grad_norm": 2.431886387737323,
      "language_loss": 0.71063256,
      "learning_rate": 1.551390042868417e-08,
      "loss": 0.73234367,
      "num_input_tokens_seen": 172591100,
      "step": 7996,
      "time_per_iteration": 4.376043319702148
    },
    {
      "auxiliary_loss_clip": 0.01099161,
      "auxiliary_loss_mlp": 0.01022112,
      "balance_loss_clip": 1.03955972,
      "balance_loss_mlp": 1.01447058,
      "epoch": 0.9615823964408104,
      "flos": 17819054663040.0,
      "grad_norm": 1.9944875697632989,
      "language_loss": 0.70930344,
      "learning_rate": 1.5417214625584207e-08,
      "loss": 0.7305162,
      "num_input_tokens_seen": 172608755,
      "step": 7997,
      "time_per_iteration": 2.8955085277557373
    },
    {
      "auxiliary_loss_clip": 0.01148415,
      "auxiliary_loss_mlp": 0.01025216,
      "balance_loss_clip": 1.04540253,
      "balance_loss_mlp": 1.01728535,
      "epoch": 0.9617026393314495,
      "flos": 20190020624640.0,
      "grad_norm": 7.279844644551098,
      "language_loss": 0.85459685,
      "learning_rate": 1.5320829882114806e-08,
      "loss": 0.87633318,
      "num_input_tokens_seen": 172626830,
      "step": 7998,
      "time_per_iteration": 2.6686933040618896
    },
    {
      "auxiliary_loss_clip": 0.01122078,
      "auxiliary_loss_mlp": 0.00762261,
      "balance_loss_clip": 1.03758097,
      "balance_loss_mlp": 1.00044203,
      "epoch": 0.9618228822220887,
      "flos": 20267013427200.0,
      "grad_norm": 1.865280238061075,
      "language_loss": 0.7884196,
      "learning_rate": 1.5224746212897378e-08,
      "loss": 0.80726302,
      "num_input_tokens_seen": 172646125,
      "step": 7999,
      "time_per_iteration": 2.7708027362823486
    },
    {
      "auxiliary_loss_clip": 0.01124725,
      "auxiliary_loss_mlp": 0.00761836,
      "balance_loss_clip": 1.04026866,
      "balance_loss_mlp": 1.00041068,
      "epoch": 0.9619431251127277,
      "flos": 21031300039680.0,
      "grad_norm": 1.6810498245635161,
      "language_loss": 0.7771579,
      "learning_rate": 1.512896363250804e-08,
      "loss": 0.79602355,
      "num_input_tokens_seen": 172666235,
      "step": 8000,
      "time_per_iteration": 2.698246717453003
    },
    {
      "auxiliary_loss_clip": 0.01143044,
      "auxiliary_loss_mlp": 0.01024136,
      "balance_loss_clip": 1.04468846,
      "balance_loss_mlp": 1.01676321,
      "epoch": 0.9620633680033668,
      "flos": 22382654538240.0,
      "grad_norm": 1.9746342306017415,
      "language_loss": 0.75681186,
      "learning_rate": 1.503348215547673e-08,
      "loss": 0.77848363,
      "num_input_tokens_seen": 172687325,
      "step": 8001,
      "time_per_iteration": 3.7089626789093018
    },
    {
      "auxiliary_loss_clip": 0.01123287,
      "auxiliary_loss_mlp": 0.01023054,
      "balance_loss_clip": 1.04119921,
      "balance_loss_mlp": 1.01557398,
      "epoch": 0.962183610894006,
      "flos": 18471730740480.0,
      "grad_norm": 2.6254418898063916,
      "language_loss": 0.8066023,
      "learning_rate": 1.4938301796288078e-08,
      "loss": 0.82806575,
      "num_input_tokens_seen": 172703895,
      "step": 8002,
      "time_per_iteration": 2.7297744750976562
    },
    {
      "auxiliary_loss_clip": 0.01154189,
      "auxiliary_loss_mlp": 0.01025641,
      "balance_loss_clip": 1.04601574,
      "balance_loss_mlp": 1.01770163,
      "epoch": 0.962303853784645,
      "flos": 18435245500800.0,
      "grad_norm": 3.318416897273892,
      "language_loss": 0.81862384,
      "learning_rate": 1.4843422569380537e-08,
      "loss": 0.84042215,
      "num_input_tokens_seen": 172720650,
      "step": 8003,
      "time_per_iteration": 2.697007656097412
    },
    {
      "auxiliary_loss_clip": 0.01126981,
      "auxiliary_loss_mlp": 0.01019059,
      "balance_loss_clip": 1.04324806,
      "balance_loss_mlp": 1.01173413,
      "epoch": 0.9624240966752841,
      "flos": 26391074826240.0,
      "grad_norm": 1.9979914659343434,
      "language_loss": 0.82528758,
      "learning_rate": 1.4748844489147483e-08,
      "loss": 0.84674799,
      "num_input_tokens_seen": 172737640,
      "step": 8004,
      "time_per_iteration": 2.783081293106079
    },
    {
      "auxiliary_loss_clip": 0.01151173,
      "auxiliary_loss_mlp": 0.01021089,
      "balance_loss_clip": 1.0427506,
      "balance_loss_mlp": 1.01383781,
      "epoch": 0.9625443395659231,
      "flos": 14647675985280.0,
      "grad_norm": 1.862024017339593,
      "language_loss": 0.71287441,
      "learning_rate": 1.4654567569936326e-08,
      "loss": 0.73459697,
      "num_input_tokens_seen": 172755215,
      "step": 8005,
      "time_per_iteration": 2.6065118312835693
    },
    {
      "auxiliary_loss_clip": 0.01158583,
      "auxiliary_loss_mlp": 0.01026236,
      "balance_loss_clip": 1.04743958,
      "balance_loss_mlp": 1.01812029,
      "epoch": 0.9626645824565623,
      "flos": 18367626147840.0,
      "grad_norm": 2.2284286565529325,
      "language_loss": 0.83423609,
      "learning_rate": 1.456059182604874e-08,
      "loss": 0.85608429,
      "num_input_tokens_seen": 172774020,
      "step": 8006,
      "time_per_iteration": 2.6685261726379395
    },
    {
      "auxiliary_loss_clip": 0.01155547,
      "auxiliary_loss_mlp": 0.01023674,
      "balance_loss_clip": 1.04482961,
      "balance_loss_mlp": 1.01553249,
      "epoch": 0.9627848253472013,
      "flos": 16580424021120.0,
      "grad_norm": 1.9734448193729277,
      "language_loss": 0.76534319,
      "learning_rate": 1.4466917271740653e-08,
      "loss": 0.78713536,
      "num_input_tokens_seen": 172792220,
      "step": 8007,
      "time_per_iteration": 2.6003456115722656
    },
    {
      "auxiliary_loss_clip": 0.01139118,
      "auxiliary_loss_mlp": 0.01028434,
      "balance_loss_clip": 1.04450071,
      "balance_loss_mlp": 1.01957059,
      "epoch": 0.9629050682378404,
      "flos": 20886867452160.0,
      "grad_norm": 2.928863926685771,
      "language_loss": 0.6795274,
      "learning_rate": 1.4373543921222697e-08,
      "loss": 0.70120287,
      "num_input_tokens_seen": 172811805,
      "step": 8008,
      "time_per_iteration": 2.724123239517212
    },
    {
      "auxiliary_loss_clip": 0.0110625,
      "auxiliary_loss_mlp": 0.01025224,
      "balance_loss_clip": 1.04079449,
      "balance_loss_mlp": 1.01778507,
      "epoch": 0.9630253111284796,
      "flos": 17019252478080.0,
      "grad_norm": 2.5872404381232212,
      "language_loss": 0.78339136,
      "learning_rate": 1.428047178865932e-08,
      "loss": 0.8047061,
      "num_input_tokens_seen": 172828595,
      "step": 8009,
      "time_per_iteration": 2.7561838626861572
    },
    {
      "auxiliary_loss_clip": 0.01116611,
      "auxiliary_loss_mlp": 0.0102002,
      "balance_loss_clip": 1.040797,
      "balance_loss_mlp": 1.01278651,
      "epoch": 0.9631455540191186,
      "flos": 20338942412160.0,
      "grad_norm": 1.8217644635949453,
      "language_loss": 0.74985218,
      "learning_rate": 1.4187700888169451e-08,
      "loss": 0.77121854,
      "num_input_tokens_seen": 172847770,
      "step": 8010,
      "time_per_iteration": 2.7077066898345947
    },
    {
      "auxiliary_loss_clip": 0.01045417,
      "auxiliary_loss_mlp": 0.01001259,
      "balance_loss_clip": 1.0124867,
      "balance_loss_mlp": 1.00036526,
      "epoch": 0.9632657969097577,
      "flos": 65956700033280.0,
      "grad_norm": 0.7568946750531425,
      "language_loss": 0.56985503,
      "learning_rate": 1.40952312338265e-08,
      "loss": 0.59032178,
      "num_input_tokens_seen": 172912415,
      "step": 8011,
      "time_per_iteration": 3.3194832801818848
    },
    {
      "auxiliary_loss_clip": 0.01115605,
      "auxiliary_loss_mlp": 0.01022503,
      "balance_loss_clip": 1.04027534,
      "balance_loss_mlp": 1.01536489,
      "epoch": 0.9633860398003968,
      "flos": 44419523823360.0,
      "grad_norm": 2.414434487370435,
      "language_loss": 0.68822813,
      "learning_rate": 1.4003062839657909e-08,
      "loss": 0.70960921,
      "num_input_tokens_seen": 172934895,
      "step": 8012,
      "time_per_iteration": 2.924306869506836
    },
    {
      "auxiliary_loss_clip": 0.0115428,
      "auxiliary_loss_mlp": 0.01024191,
      "balance_loss_clip": 1.04421151,
      "balance_loss_mlp": 1.01714301,
      "epoch": 0.9635062826910359,
      "flos": 24827704300800.0,
      "grad_norm": 2.2400632456855236,
      "language_loss": 0.80005836,
      "learning_rate": 1.391119571964583e-08,
      "loss": 0.82184309,
      "num_input_tokens_seen": 172955835,
      "step": 8013,
      "time_per_iteration": 2.7167670726776123
    },
    {
      "auxiliary_loss_clip": 0.01128219,
      "auxiliary_loss_mlp": 0.01025033,
      "balance_loss_clip": 1.04341245,
      "balance_loss_mlp": 1.01786232,
      "epoch": 0.9636265255816749,
      "flos": 15961360095360.0,
      "grad_norm": 2.411903754697974,
      "language_loss": 0.72816104,
      "learning_rate": 1.3819629887726225e-08,
      "loss": 0.74969351,
      "num_input_tokens_seen": 172973925,
      "step": 8014,
      "time_per_iteration": 2.677516222000122
    },
    {
      "auxiliary_loss_clip": 0.01129938,
      "auxiliary_loss_mlp": 0.01024137,
      "balance_loss_clip": 1.04605365,
      "balance_loss_mlp": 1.01653719,
      "epoch": 0.9637467684723141,
      "flos": 22601781457920.0,
      "grad_norm": 2.3252685862232636,
      "language_loss": 0.76177394,
      "learning_rate": 1.3728365357789317e-08,
      "loss": 0.7833147,
      "num_input_tokens_seen": 172993290,
      "step": 8015,
      "time_per_iteration": 2.8443615436553955
    },
    {
      "auxiliary_loss_clip": 0.01100217,
      "auxiliary_loss_mlp": 0.01026343,
      "balance_loss_clip": 1.04170251,
      "balance_loss_mlp": 1.01774228,
      "epoch": 0.9638670113629532,
      "flos": 17565812801280.0,
      "grad_norm": 2.483553790595973,
      "language_loss": 0.76290375,
      "learning_rate": 1.3637402143680254e-08,
      "loss": 0.78416932,
      "num_input_tokens_seen": 173008190,
      "step": 8016,
      "time_per_iteration": 2.7703769207000732
    },
    {
      "auxiliary_loss_clip": 0.01026738,
      "auxiliary_loss_mlp": 0.01000349,
      "balance_loss_clip": 1.01068878,
      "balance_loss_mlp": 0.99945492,
      "epoch": 0.9639872542535922,
      "flos": 55072139379840.0,
      "grad_norm": 0.7234702393016952,
      "language_loss": 0.55021745,
      "learning_rate": 1.3546740259197998e-08,
      "loss": 0.57048833,
      "num_input_tokens_seen": 173061000,
      "step": 8017,
      "time_per_iteration": 3.260359764099121
    },
    {
      "auxiliary_loss_clip": 0.01098766,
      "auxiliary_loss_mlp": 0.01027125,
      "balance_loss_clip": 1.03935337,
      "balance_loss_mlp": 1.01934671,
      "epoch": 0.9641074971442314,
      "flos": 24134484746880.0,
      "grad_norm": 2.5741029927322425,
      "language_loss": 0.70148897,
      "learning_rate": 1.3456379718095989e-08,
      "loss": 0.72274786,
      "num_input_tokens_seen": 173081415,
      "step": 8018,
      "time_per_iteration": 2.790693759918213
    },
    {
      "auxiliary_loss_clip": 0.01046363,
      "auxiliary_loss_mlp": 0.01000304,
      "balance_loss_clip": 1.01118958,
      "balance_loss_mlp": 0.99939805,
      "epoch": 0.9642277400348704,
      "flos": 66747416077440.0,
      "grad_norm": 0.8608585080200593,
      "language_loss": 0.6198647,
      "learning_rate": 1.3366320534081487e-08,
      "loss": 0.64033139,
      "num_input_tokens_seen": 173144095,
      "step": 8019,
      "time_per_iteration": 4.249652147293091
    },
    {
      "auxiliary_loss_clip": 0.01100277,
      "auxiliary_loss_mlp": 0.00762455,
      "balance_loss_clip": 1.03919339,
      "balance_loss_mlp": 1.00040889,
      "epoch": 0.9643479829255095,
      "flos": 30920272450560.0,
      "grad_norm": 2.444245176225366,
      "language_loss": 0.75785625,
      "learning_rate": 1.3276562720816675e-08,
      "loss": 0.77648354,
      "num_input_tokens_seen": 173165605,
      "step": 8020,
      "time_per_iteration": 3.7398829460144043
    },
    {
      "auxiliary_loss_clip": 0.01126499,
      "auxiliary_loss_mlp": 0.01024472,
      "balance_loss_clip": 1.04378688,
      "balance_loss_mlp": 1.01652062,
      "epoch": 0.9644682258161487,
      "flos": 20048245643520.0,
      "grad_norm": 2.426395075197953,
      "language_loss": 0.82978976,
      "learning_rate": 1.3187106291917549e-08,
      "loss": 0.85129941,
      "num_input_tokens_seen": 173182595,
      "step": 8021,
      "time_per_iteration": 3.682817220687866
    },
    {
      "auxiliary_loss_clip": 0.01113979,
      "auxiliary_loss_mlp": 0.01022978,
      "balance_loss_clip": 1.04026377,
      "balance_loss_mlp": 1.01570892,
      "epoch": 0.9645884687067877,
      "flos": 21178713456000.0,
      "grad_norm": 2.3105185378054705,
      "language_loss": 0.70865542,
      "learning_rate": 1.309795126095503e-08,
      "loss": 0.73002499,
      "num_input_tokens_seen": 173200895,
      "step": 8022,
      "time_per_iteration": 2.756834030151367
    },
    {
      "auxiliary_loss_clip": 0.01134424,
      "auxiliary_loss_mlp": 0.01019381,
      "balance_loss_clip": 1.04108405,
      "balance_loss_mlp": 1.01212978,
      "epoch": 0.9647087115974268,
      "flos": 18945967029120.0,
      "grad_norm": 2.257383090835471,
      "language_loss": 0.80490714,
      "learning_rate": 1.3009097641453192e-08,
      "loss": 0.82644522,
      "num_input_tokens_seen": 173218745,
      "step": 8023,
      "time_per_iteration": 2.7580299377441406
    },
    {
      "auxiliary_loss_clip": 0.01128462,
      "auxiliary_loss_mlp": 0.01027412,
      "balance_loss_clip": 1.04150462,
      "balance_loss_mlp": 1.01991379,
      "epoch": 0.9648289544880659,
      "flos": 16545088016640.0,
      "grad_norm": 2.0515505823926126,
      "language_loss": 0.76054776,
      "learning_rate": 1.2920545446891474e-08,
      "loss": 0.78210652,
      "num_input_tokens_seen": 173235465,
      "step": 8024,
      "time_per_iteration": 2.7893168926239014
    },
    {
      "auxiliary_loss_clip": 0.01171217,
      "auxiliary_loss_mlp": 0.01028362,
      "balance_loss_clip": 1.04951096,
      "balance_loss_mlp": 1.02088785,
      "epoch": 0.964949197378705,
      "flos": 24057527857920.0,
      "grad_norm": 2.405367235769762,
      "language_loss": 0.70753556,
      "learning_rate": 1.2832294690703127e-08,
      "loss": 0.72953135,
      "num_input_tokens_seen": 173254440,
      "step": 8025,
      "time_per_iteration": 2.6692495346069336
    },
    {
      "auxiliary_loss_clip": 0.01160179,
      "auxiliary_loss_mlp": 0.01021815,
      "balance_loss_clip": 1.04709971,
      "balance_loss_mlp": 1.01406074,
      "epoch": 0.965069440269344,
      "flos": 23365565280000.0,
      "grad_norm": 2.0891119075369255,
      "language_loss": 0.77391428,
      "learning_rate": 1.2744345386275668e-08,
      "loss": 0.79573423,
      "num_input_tokens_seen": 173273980,
      "step": 8026,
      "time_per_iteration": 3.6276588439941406
    },
    {
      "auxiliary_loss_clip": 0.01120093,
      "auxiliary_loss_mlp": 0.01026055,
      "balance_loss_clip": 1.0447185,
      "balance_loss_mlp": 1.01809788,
      "epoch": 0.9651896831599832,
      "flos": 25374875155200.0,
      "grad_norm": 1.612948486141217,
      "language_loss": 0.78543115,
      "learning_rate": 1.265669754695109e-08,
      "loss": 0.80689263,
      "num_input_tokens_seen": 173293550,
      "step": 8027,
      "time_per_iteration": 2.7871158123016357
    },
    {
      "auxiliary_loss_clip": 0.01130939,
      "auxiliary_loss_mlp": 0.01022424,
      "balance_loss_clip": 1.04047203,
      "balance_loss_mlp": 1.01435649,
      "epoch": 0.9653099260506223,
      "flos": 22272875596800.0,
      "grad_norm": 2.0471671645338,
      "language_loss": 0.82017207,
      "learning_rate": 1.2569351186025201e-08,
      "loss": 0.84170568,
      "num_input_tokens_seen": 173312005,
      "step": 8028,
      "time_per_iteration": 2.666443109512329
    },
    {
      "auxiliary_loss_clip": 0.01130037,
      "auxiliary_loss_mlp": 0.01024481,
      "balance_loss_clip": 1.03922629,
      "balance_loss_mlp": 1.01753688,
      "epoch": 0.9654301689412613,
      "flos": 26760847386240.0,
      "grad_norm": 2.0320881262118298,
      "language_loss": 0.75525254,
      "learning_rate": 1.2482306316748737e-08,
      "loss": 0.77679771,
      "num_input_tokens_seen": 173332450,
      "step": 8029,
      "time_per_iteration": 2.7358508110046387
    },
    {
      "auxiliary_loss_clip": 0.01106812,
      "auxiliary_loss_mlp": 0.00762535,
      "balance_loss_clip": 1.03771102,
      "balance_loss_mlp": 1.00038791,
      "epoch": 0.9655504118319005,
      "flos": 17412689122560.0,
      "grad_norm": 2.228476116228172,
      "language_loss": 0.78533131,
      "learning_rate": 1.2395562952326021e-08,
      "loss": 0.80402476,
      "num_input_tokens_seen": 173349610,
      "step": 8030,
      "time_per_iteration": 2.7675859928131104
    },
    {
      "auxiliary_loss_clip": 0.01053753,
      "auxiliary_loss_mlp": 0.0102785,
      "balance_loss_clip": 1.0389359,
      "balance_loss_mlp": 1.0191834,
      "epoch": 0.9656706547225395,
      "flos": 22126970551680.0,
      "grad_norm": 2.1304024784432785,
      "language_loss": 0.8103385,
      "learning_rate": 1.2309121105916309e-08,
      "loss": 0.83115453,
      "num_input_tokens_seen": 173367900,
      "step": 8031,
      "time_per_iteration": 2.9912023544311523
    },
    {
      "auxiliary_loss_clip": 0.01150054,
      "auxiliary_loss_mlp": 0.01024569,
      "balance_loss_clip": 1.04338121,
      "balance_loss_mlp": 1.01694,
      "epoch": 0.9657908976131786,
      "flos": 37049289926400.0,
      "grad_norm": 2.1815176704694053,
      "language_loss": 0.69179678,
      "learning_rate": 1.222298079063222e-08,
      "loss": 0.713543,
      "num_input_tokens_seen": 173389040,
      "step": 8032,
      "time_per_iteration": 3.309072494506836
    },
    {
      "auxiliary_loss_clip": 0.0109369,
      "auxiliary_loss_mlp": 0.01020945,
      "balance_loss_clip": 1.03753543,
      "balance_loss_mlp": 1.014117,
      "epoch": 0.9659111405038178,
      "flos": 24389809597440.0,
      "grad_norm": 3.2031781435108635,
      "language_loss": 0.72343934,
      "learning_rate": 1.2137142019541524e-08,
      "loss": 0.74458569,
      "num_input_tokens_seen": 173407595,
      "step": 8033,
      "time_per_iteration": 2.9077444076538086
    },
    {
      "auxiliary_loss_clip": 0.01129237,
      "auxiliary_loss_mlp": 0.01023937,
      "balance_loss_clip": 1.04538035,
      "balance_loss_mlp": 1.01699615,
      "epoch": 0.9660313833944568,
      "flos": 25009412227200.0,
      "grad_norm": 2.1386743874822454,
      "language_loss": 0.73556745,
      "learning_rate": 1.2051604805666027e-08,
      "loss": 0.75709927,
      "num_input_tokens_seen": 173424720,
      "step": 8034,
      "time_per_iteration": 2.74832820892334
    },
    {
      "auxiliary_loss_clip": 0.01133696,
      "auxiliary_loss_mlp": 0.01021991,
      "balance_loss_clip": 1.04432309,
      "balance_loss_mlp": 1.01454687,
      "epoch": 0.9661516262850959,
      "flos": 11801575895040.0,
      "grad_norm": 2.138831830016502,
      "language_loss": 0.78544796,
      "learning_rate": 1.196636916198135e-08,
      "loss": 0.80700487,
      "num_input_tokens_seen": 173442260,
      "step": 8035,
      "time_per_iteration": 2.7889111042022705
    },
    {
      "auxiliary_loss_clip": 0.01157601,
      "auxiliary_loss_mlp": 0.01024768,
      "balance_loss_clip": 1.04479218,
      "balance_loss_mlp": 1.01698911,
      "epoch": 0.9662718691757349,
      "flos": 20047778766720.0,
      "grad_norm": 2.1466154027086293,
      "language_loss": 0.76783937,
      "learning_rate": 1.1881435101418036e-08,
      "loss": 0.78966302,
      "num_input_tokens_seen": 173461675,
      "step": 8036,
      "time_per_iteration": 2.6282174587249756
    },
    {
      "auxiliary_loss_clip": 0.00980255,
      "auxiliary_loss_mlp": 0.01001543,
      "balance_loss_clip": 1.01280379,
      "balance_loss_mlp": 1.00051236,
      "epoch": 0.9663921120663741,
      "flos": 68027703517440.0,
      "grad_norm": 0.725869949581116,
      "language_loss": 0.65523154,
      "learning_rate": 1.1796802636860003e-08,
      "loss": 0.67504954,
      "num_input_tokens_seen": 173530205,
      "step": 8037,
      "time_per_iteration": 4.025680065155029
    },
    {
      "auxiliary_loss_clip": 0.01075403,
      "auxiliary_loss_mlp": 0.01022363,
      "balance_loss_clip": 1.03761768,
      "balance_loss_mlp": 1.014781,
      "epoch": 0.9665123549570132,
      "flos": 26322916769280.0,
      "grad_norm": 2.2438789336142078,
      "language_loss": 0.73638421,
      "learning_rate": 1.1712471781146316e-08,
      "loss": 0.75736189,
      "num_input_tokens_seen": 173549540,
      "step": 8038,
      "time_per_iteration": 4.05248236656189
    },
    {
      "auxiliary_loss_clip": 0.01122193,
      "auxiliary_loss_mlp": 0.01022761,
      "balance_loss_clip": 1.04180574,
      "balance_loss_mlp": 1.01539969,
      "epoch": 0.9666325978476522,
      "flos": 43941121557120.0,
      "grad_norm": 2.246277520669249,
      "language_loss": 0.66798496,
      "learning_rate": 1.1628442547069628e-08,
      "loss": 0.68943453,
      "num_input_tokens_seen": 173571740,
      "step": 8039,
      "time_per_iteration": 3.32426381111145
    },
    {
      "auxiliary_loss_clip": 0.01120017,
      "auxiliary_loss_mlp": 0.01022024,
      "balance_loss_clip": 1.03978682,
      "balance_loss_mlp": 1.01437652,
      "epoch": 0.9667528407382914,
      "flos": 21543422198400.0,
      "grad_norm": 2.204254309217907,
      "language_loss": 0.77291131,
      "learning_rate": 1.1544714947377521e-08,
      "loss": 0.79433167,
      "num_input_tokens_seen": 173589425,
      "step": 8040,
      "time_per_iteration": 2.7604403495788574
    },
    {
      "auxiliary_loss_clip": 0.0110325,
      "auxiliary_loss_mlp": 0.01023637,
      "balance_loss_clip": 1.04457974,
      "balance_loss_mlp": 1.01571548,
      "epoch": 0.9668730836289304,
      "flos": 23878585278720.0,
      "grad_norm": 2.5727448701928424,
      "language_loss": 0.70499802,
      "learning_rate": 1.1461288994770945e-08,
      "loss": 0.72626686,
      "num_input_tokens_seen": 173608500,
      "step": 8041,
      "time_per_iteration": 2.9358572959899902
    },
    {
      "auxiliary_loss_clip": 0.01132433,
      "auxiliary_loss_mlp": 0.01025578,
      "balance_loss_clip": 1.04139125,
      "balance_loss_mlp": 1.01757956,
      "epoch": 0.9669933265195695,
      "flos": 28293011971200.0,
      "grad_norm": 2.7189743301332427,
      "language_loss": 0.77234364,
      "learning_rate": 1.1378164701906002e-08,
      "loss": 0.79392368,
      "num_input_tokens_seen": 173630265,
      "step": 8042,
      "time_per_iteration": 2.7528235912323
    },
    {
      "auxiliary_loss_clip": 0.0110243,
      "auxiliary_loss_mlp": 0.01027871,
      "balance_loss_clip": 1.04049277,
      "balance_loss_mlp": 1.01979518,
      "epoch": 0.9671135694102087,
      "flos": 22454763091200.0,
      "grad_norm": 1.726462978367233,
      "language_loss": 0.66475892,
      "learning_rate": 1.1295342081392156e-08,
      "loss": 0.68606186,
      "num_input_tokens_seen": 173649625,
      "step": 8043,
      "time_per_iteration": 2.784113883972168
    },
    {
      "auxiliary_loss_clip": 0.01135265,
      "auxiliary_loss_mlp": 0.01023275,
      "balance_loss_clip": 1.04525864,
      "balance_loss_mlp": 1.01613104,
      "epoch": 0.9672338123008477,
      "flos": 20155941596160.0,
      "grad_norm": 1.9693705957835446,
      "language_loss": 0.69746065,
      "learning_rate": 1.1212821145793804e-08,
      "loss": 0.719046,
      "num_input_tokens_seen": 173669240,
      "step": 8044,
      "time_per_iteration": 2.653991460800171
    },
    {
      "auxiliary_loss_clip": 0.01152355,
      "auxiliary_loss_mlp": 0.01023712,
      "balance_loss_clip": 1.04430592,
      "balance_loss_mlp": 1.01625538,
      "epoch": 0.9673540551914868,
      "flos": 16977487939200.0,
      "grad_norm": 2.368979896482645,
      "language_loss": 0.78964531,
      "learning_rate": 1.1130601907629156e-08,
      "loss": 0.81140602,
      "num_input_tokens_seen": 173686970,
      "step": 8045,
      "time_per_iteration": 3.4919486045837402
    },
    {
      "auxiliary_loss_clip": 0.01032795,
      "auxiliary_loss_mlp": 0.00999748,
      "balance_loss_clip": 1.00825953,
      "balance_loss_mlp": 0.99872267,
      "epoch": 0.9674742980821259,
      "flos": 61892903952000.0,
      "grad_norm": 0.8269827950691125,
      "language_loss": 0.64770615,
      "learning_rate": 1.1048684379370899e-08,
      "loss": 0.66803157,
      "num_input_tokens_seen": 173747655,
      "step": 8046,
      "time_per_iteration": 4.134878873825073
    },
    {
      "auxiliary_loss_clip": 0.01154697,
      "auxiliary_loss_mlp": 0.01024334,
      "balance_loss_clip": 1.04581308,
      "balance_loss_mlp": 1.01701736,
      "epoch": 0.967594540972765,
      "flos": 18697824898560.0,
      "grad_norm": 9.07101138799354,
      "language_loss": 0.74865663,
      "learning_rate": 1.0967068573445759e-08,
      "loss": 0.7704469,
      "num_input_tokens_seen": 173765140,
      "step": 8047,
      "time_per_iteration": 3.5541985034942627
    },
    {
      "auxiliary_loss_clip": 0.01139117,
      "auxiliary_loss_mlp": 0.01023564,
      "balance_loss_clip": 1.04262793,
      "balance_loss_mlp": 1.01597083,
      "epoch": 0.967714783863404,
      "flos": 20777411733120.0,
      "grad_norm": 1.972652822175121,
      "language_loss": 0.65356159,
      "learning_rate": 1.0885754502234945e-08,
      "loss": 0.67518842,
      "num_input_tokens_seen": 173784800,
      "step": 8048,
      "time_per_iteration": 2.6983163356781006
    },
    {
      "auxiliary_loss_clip": 0.011165,
      "auxiliary_loss_mlp": 0.01020372,
      "balance_loss_clip": 1.04095221,
      "balance_loss_mlp": 1.01236153,
      "epoch": 0.9678350267540432,
      "flos": 23185473465600.0,
      "grad_norm": 1.9107999853899398,
      "language_loss": 0.78083378,
      "learning_rate": 1.08047421780737e-08,
      "loss": 0.80220252,
      "num_input_tokens_seen": 173803990,
      "step": 8049,
      "time_per_iteration": 2.769676923751831
    },
    {
      "auxiliary_loss_clip": 0.01144902,
      "auxiliary_loss_mlp": 0.01022662,
      "balance_loss_clip": 1.04613638,
      "balance_loss_mlp": 1.01526785,
      "epoch": 0.9679552696446823,
      "flos": 21726063878400.0,
      "grad_norm": 5.702395961198445,
      "language_loss": 0.73930788,
      "learning_rate": 1.0724031613251305e-08,
      "loss": 0.76098353,
      "num_input_tokens_seen": 173821890,
      "step": 8050,
      "time_per_iteration": 2.7213003635406494
    },
    {
      "auxiliary_loss_clip": 0.01121017,
      "auxiliary_loss_mlp": 0.01027228,
      "balance_loss_clip": 1.04104543,
      "balance_loss_mlp": 1.01905584,
      "epoch": 0.9680755125353213,
      "flos": 26869046129280.0,
      "grad_norm": 3.530681702367444,
      "language_loss": 0.66672719,
      "learning_rate": 1.0643622820011744e-08,
      "loss": 0.68820959,
      "num_input_tokens_seen": 173842945,
      "step": 8051,
      "time_per_iteration": 2.8018312454223633
    },
    {
      "auxiliary_loss_clip": 0.01095922,
      "auxiliary_loss_mlp": 0.01028767,
      "balance_loss_clip": 1.04013693,
      "balance_loss_mlp": 1.02000546,
      "epoch": 0.9681957554259605,
      "flos": 28325008010880.0,
      "grad_norm": 2.638765000671085,
      "language_loss": 0.68444741,
      "learning_rate": 1.0563515810552814e-08,
      "loss": 0.70569426,
      "num_input_tokens_seen": 173859915,
      "step": 8052,
      "time_per_iteration": 3.777534246444702
    },
    {
      "auxiliary_loss_clip": 0.01110275,
      "auxiliary_loss_mlp": 0.01025836,
      "balance_loss_clip": 1.04535413,
      "balance_loss_mlp": 1.01751471,
      "epoch": 0.9683159983165995,
      "flos": 20557674282240.0,
      "grad_norm": 1.7572870228576825,
      "language_loss": 0.73133147,
      "learning_rate": 1.0483710597026795e-08,
      "loss": 0.75269258,
      "num_input_tokens_seen": 173879775,
      "step": 8053,
      "time_per_iteration": 2.706684112548828
    },
    {
      "auxiliary_loss_clip": 0.01120411,
      "auxiliary_loss_mlp": 0.01024951,
      "balance_loss_clip": 1.04168367,
      "balance_loss_mlp": 1.01756573,
      "epoch": 0.9684362412072386,
      "flos": 24207958016640.0,
      "grad_norm": 2.983015631499851,
      "language_loss": 0.73808968,
      "learning_rate": 1.0404207191540227e-08,
      "loss": 0.75954324,
      "num_input_tokens_seen": 173900230,
      "step": 8054,
      "time_per_iteration": 2.7722012996673584
    },
    {
      "auxiliary_loss_clip": 0.0112487,
      "auxiliary_loss_mlp": 0.01022937,
      "balance_loss_clip": 1.04184449,
      "balance_loss_mlp": 1.01499772,
      "epoch": 0.9685564840978778,
      "flos": 22346241125760.0,
      "grad_norm": 2.5901618522049565,
      "language_loss": 0.74960339,
      "learning_rate": 1.0325005606153236e-08,
      "loss": 0.77108139,
      "num_input_tokens_seen": 173919690,
      "step": 8055,
      "time_per_iteration": 2.6964664459228516
    },
    {
      "auxiliary_loss_clip": 0.01126929,
      "auxiliary_loss_mlp": 0.01026241,
      "balance_loss_clip": 1.04203403,
      "balance_loss_mlp": 1.01852798,
      "epoch": 0.9686767269885168,
      "flos": 14386389477120.0,
      "grad_norm": 2.4568008393062906,
      "language_loss": 0.79390275,
      "learning_rate": 1.0246105852881104e-08,
      "loss": 0.81543446,
      "num_input_tokens_seen": 173934790,
      "step": 8056,
      "time_per_iteration": 2.6937034130096436
    },
    {
      "auxiliary_loss_clip": 0.0114213,
      "auxiliary_loss_mlp": 0.01031625,
      "balance_loss_clip": 1.04355586,
      "balance_loss_mlp": 1.02408481,
      "epoch": 0.9687969698791559,
      "flos": 21287630471040.0,
      "grad_norm": 2.165017412313263,
      "language_loss": 0.78687644,
      "learning_rate": 1.0167507943692476e-08,
      "loss": 0.80861402,
      "num_input_tokens_seen": 173953875,
      "step": 8057,
      "time_per_iteration": 2.6803324222564697
    },
    {
      "auxiliary_loss_clip": 0.01160209,
      "auxiliary_loss_mlp": 0.01027916,
      "balance_loss_clip": 1.04760492,
      "balance_loss_mlp": 1.01967275,
      "epoch": 0.968917212769795,
      "flos": 19828328624640.0,
      "grad_norm": 3.5043303046523464,
      "language_loss": 0.71812761,
      "learning_rate": 1.008921189051093e-08,
      "loss": 0.74000883,
      "num_input_tokens_seen": 173971220,
      "step": 8058,
      "time_per_iteration": 2.6979763507843018
    },
    {
      "auxiliary_loss_clip": 0.01115838,
      "auxiliary_loss_mlp": 0.01026327,
      "balance_loss_clip": 1.04244494,
      "balance_loss_mlp": 1.01775587,
      "epoch": 0.9690374556604341,
      "flos": 21681749473920.0,
      "grad_norm": 2.0769391713825267,
      "language_loss": 0.77329952,
      "learning_rate": 1.0011217705213848e-08,
      "loss": 0.79472119,
      "num_input_tokens_seen": 173989095,
      "step": 8059,
      "time_per_iteration": 2.756071090698242
    },
    {
      "auxiliary_loss_clip": 0.01126028,
      "auxiliary_loss_mlp": 0.01022096,
      "balance_loss_clip": 1.042243,
      "balance_loss_mlp": 1.01524496,
      "epoch": 0.9691576985510731,
      "flos": 32635437851520.0,
      "grad_norm": 1.8487434260953992,
      "language_loss": 0.74604785,
      "learning_rate": 9.933525399632658e-09,
      "loss": 0.76752913,
      "num_input_tokens_seen": 174007330,
      "step": 8060,
      "time_per_iteration": 2.7619311809539795
    },
    {
      "auxiliary_loss_clip": 0.01114994,
      "auxiliary_loss_mlp": 0.01022476,
      "balance_loss_clip": 1.04098749,
      "balance_loss_mlp": 1.01451826,
      "epoch": 0.9692779414417123,
      "flos": 35663174040960.0,
      "grad_norm": 2.000334629878782,
      "language_loss": 0.64787954,
      "learning_rate": 9.856134985553488e-09,
      "loss": 0.66925418,
      "num_input_tokens_seen": 174027055,
      "step": 8061,
      "time_per_iteration": 3.0130441188812256
    },
    {
      "auxiliary_loss_clip": 0.01107858,
      "auxiliary_loss_mlp": 0.01026325,
      "balance_loss_clip": 1.04072976,
      "balance_loss_mlp": 1.01872504,
      "epoch": 0.9693981843323514,
      "flos": 28366952117760.0,
      "grad_norm": 1.8656375007859018,
      "language_loss": 0.73680234,
      "learning_rate": 9.77904647471628e-09,
      "loss": 0.75814414,
      "num_input_tokens_seen": 174050235,
      "step": 8062,
      "time_per_iteration": 2.850700855255127
    },
    {
      "auxiliary_loss_clip": 0.01123327,
      "auxiliary_loss_mlp": 0.01031282,
      "balance_loss_clip": 1.04151297,
      "balance_loss_mlp": 1.02328861,
      "epoch": 0.9695184272229904,
      "flos": 23622865378560.0,
      "grad_norm": 1.8776180289691764,
      "language_loss": 0.73846066,
      "learning_rate": 9.702259878815454e-09,
      "loss": 0.76000679,
      "num_input_tokens_seen": 174070560,
      "step": 8063,
      "time_per_iteration": 2.8012499809265137
    },
    {
      "auxiliary_loss_clip": 0.01076842,
      "auxiliary_loss_mlp": 0.01026459,
      "balance_loss_clip": 1.04162109,
      "balance_loss_mlp": 1.01834655,
      "epoch": 0.9696386701136296,
      "flos": 23294677789440.0,
      "grad_norm": 2.149387267906729,
      "language_loss": 0.74751711,
      "learning_rate": 9.625775209499254e-09,
      "loss": 0.76855016,
      "num_input_tokens_seen": 174090565,
      "step": 8064,
      "time_per_iteration": 3.0808029174804688
    },
    {
      "auxiliary_loss_clip": 0.01108017,
      "auxiliary_loss_mlp": 0.00762229,
      "balance_loss_clip": 1.04160511,
      "balance_loss_mlp": 1.00050259,
      "epoch": 0.9697589130042686,
      "flos": 15121876360320.0,
      "grad_norm": 2.1021074976328604,
      "language_loss": 0.73669535,
      "learning_rate": 9.549592478370172e-09,
      "loss": 0.7553978,
      "num_input_tokens_seen": 174108745,
      "step": 8065,
      "time_per_iteration": 3.193836212158203
    },
    {
      "auxiliary_loss_clip": 0.01087335,
      "auxiliary_loss_mlp": 0.01020977,
      "balance_loss_clip": 1.03416049,
      "balance_loss_mlp": 1.01349926,
      "epoch": 0.9698791558949077,
      "flos": 18879532824960.0,
      "grad_norm": 2.8203762326256188,
      "language_loss": 0.79246032,
      "learning_rate": 9.473711696985632e-09,
      "loss": 0.81354344,
      "num_input_tokens_seen": 174128075,
      "step": 8066,
      "time_per_iteration": 2.8342983722686768
    },
    {
      "auxiliary_loss_clip": 0.01098614,
      "auxiliary_loss_mlp": 0.01026086,
      "balance_loss_clip": 1.03573751,
      "balance_loss_mlp": 1.01831961,
      "epoch": 0.9699993987855468,
      "flos": 17931455297280.0,
      "grad_norm": 2.116956672243187,
      "language_loss": 0.76067543,
      "learning_rate": 9.398132876856201e-09,
      "loss": 0.7819224,
      "num_input_tokens_seen": 174147040,
      "step": 8067,
      "time_per_iteration": 2.714081048965454
    },
    {
      "auxiliary_loss_clip": 0.01026373,
      "auxiliary_loss_mlp": 0.01000456,
      "balance_loss_clip": 1.01099586,
      "balance_loss_mlp": 0.99948484,
      "epoch": 0.9701196416761859,
      "flos": 67182186297600.0,
      "grad_norm": 0.7798162891375813,
      "language_loss": 0.60801494,
      "learning_rate": 9.322856029447379e-09,
      "loss": 0.62828326,
      "num_input_tokens_seen": 174208225,
      "step": 8068,
      "time_per_iteration": 3.211106538772583
    },
    {
      "auxiliary_loss_clip": 0.01079878,
      "auxiliary_loss_mlp": 0.01025434,
      "balance_loss_clip": 1.03695631,
      "balance_loss_mlp": 1.01792407,
      "epoch": 0.970239884566825,
      "flos": 24277804012800.0,
      "grad_norm": 2.0641151884634636,
      "language_loss": 0.80422997,
      "learning_rate": 9.247881166178695e-09,
      "loss": 0.82528311,
      "num_input_tokens_seen": 174226935,
      "step": 8069,
      "time_per_iteration": 2.8204727172851562
    },
    {
      "auxiliary_loss_clip": 0.01113855,
      "auxiliary_loss_mlp": 0.01024119,
      "balance_loss_clip": 1.04050601,
      "balance_loss_mlp": 1.01638269,
      "epoch": 0.970360127457464,
      "flos": 25301689194240.0,
      "grad_norm": 7.546747453595161,
      "language_loss": 0.76558018,
      "learning_rate": 9.173208298423274e-09,
      "loss": 0.78695989,
      "num_input_tokens_seen": 174248140,
      "step": 8070,
      "time_per_iteration": 2.7769339084625244
    },
    {
      "auxiliary_loss_clip": 0.01138983,
      "auxiliary_loss_mlp": 0.010234,
      "balance_loss_clip": 1.04370761,
      "balance_loss_mlp": 1.01598787,
      "epoch": 0.9704803703481032,
      "flos": 29572473398400.0,
      "grad_norm": 1.6680149217086968,
      "language_loss": 0.7631461,
      "learning_rate": 9.09883743750961e-09,
      "loss": 0.78476995,
      "num_input_tokens_seen": 174271030,
      "step": 8071,
      "time_per_iteration": 4.885892868041992
    },
    {
      "auxiliary_loss_clip": 0.01148704,
      "auxiliary_loss_mlp": 0.010238,
      "balance_loss_clip": 1.04340959,
      "balance_loss_mlp": 1.0160867,
      "epoch": 0.9706006132387422,
      "flos": 17380046638080.0,
      "grad_norm": 1.6747877783139,
      "language_loss": 0.83859563,
      "learning_rate": 9.024768594719124e-09,
      "loss": 0.86032069,
      "num_input_tokens_seen": 174289410,
      "step": 8072,
      "time_per_iteration": 3.5692150592803955
    },
    {
      "auxiliary_loss_clip": 0.01113756,
      "auxiliary_loss_mlp": 0.01024665,
      "balance_loss_clip": 1.04358387,
      "balance_loss_mlp": 1.01676691,
      "epoch": 0.9707208561293813,
      "flos": 18186421011840.0,
      "grad_norm": 2.2136236393031474,
      "language_loss": 0.72869074,
      "learning_rate": 8.95100178128816e-09,
      "loss": 0.75007492,
      "num_input_tokens_seen": 174308550,
      "step": 8073,
      "time_per_iteration": 2.7498819828033447
    },
    {
      "auxiliary_loss_clip": 0.01136882,
      "auxiliary_loss_mlp": 0.01026223,
      "balance_loss_clip": 1.0465951,
      "balance_loss_mlp": 1.01868272,
      "epoch": 0.9708410990200205,
      "flos": 31248388212480.0,
      "grad_norm": 1.9060828998871384,
      "language_loss": 0.69934374,
      "learning_rate": 8.877537008407321e-09,
      "loss": 0.72097474,
      "num_input_tokens_seen": 174328600,
      "step": 8074,
      "time_per_iteration": 2.767714738845825
    },
    {
      "auxiliary_loss_clip": 0.0116842,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.04790187,
      "balance_loss_mlp": 1.01756597,
      "epoch": 0.9709613419106595,
      "flos": 30554450386560.0,
      "grad_norm": 1.6503408508742663,
      "language_loss": 0.68680799,
      "learning_rate": 8.804374287221028e-09,
      "loss": 0.7087422,
      "num_input_tokens_seen": 174349835,
      "step": 8075,
      "time_per_iteration": 2.62650465965271
    },
    {
      "auxiliary_loss_clip": 0.0113906,
      "auxiliary_loss_mlp": 0.01020199,
      "balance_loss_clip": 1.04275799,
      "balance_loss_mlp": 1.01308799,
      "epoch": 0.9710815848012986,
      "flos": 23730166281600.0,
      "grad_norm": 1.6716594643496359,
      "language_loss": 0.84346962,
      "learning_rate": 8.731513628827958e-09,
      "loss": 0.86506224,
      "num_input_tokens_seen": 174369200,
      "step": 8076,
      "time_per_iteration": 2.7090446949005127
    },
    {
      "auxiliary_loss_clip": 0.01158137,
      "auxiliary_loss_mlp": 0.00762046,
      "balance_loss_clip": 1.04622793,
      "balance_loss_mlp": 1.00043595,
      "epoch": 0.9712018276919377,
      "flos": 23761875012480.0,
      "grad_norm": 1.861294466324686,
      "language_loss": 0.8268097,
      "learning_rate": 8.658955044280825e-09,
      "loss": 0.84601152,
      "num_input_tokens_seen": 174388125,
      "step": 8077,
      "time_per_iteration": 2.6913111209869385
    },
    {
      "auxiliary_loss_clip": 0.01123545,
      "auxiliary_loss_mlp": 0.01024719,
      "balance_loss_clip": 1.03951764,
      "balance_loss_mlp": 1.01710796,
      "epoch": 0.9713220705825768,
      "flos": 23330983461120.0,
      "grad_norm": 1.7220909324581757,
      "language_loss": 0.77727193,
      "learning_rate": 8.586698544587268e-09,
      "loss": 0.79875457,
      "num_input_tokens_seen": 174409735,
      "step": 8078,
      "time_per_iteration": 3.678645133972168
    },
    {
      "auxiliary_loss_clip": 0.01141379,
      "auxiliary_loss_mlp": 0.01019134,
      "balance_loss_clip": 1.04508901,
      "balance_loss_mlp": 1.01184726,
      "epoch": 0.9714423134732159,
      "flos": 22200946611840.0,
      "grad_norm": 2.9771066065106693,
      "language_loss": 0.73912597,
      "learning_rate": 8.514744140707853e-09,
      "loss": 0.7607311,
      "num_input_tokens_seen": 174428875,
      "step": 8079,
      "time_per_iteration": 2.685206413269043
    },
    {
      "auxiliary_loss_clip": 0.01149616,
      "auxiliary_loss_mlp": 0.01023613,
      "balance_loss_clip": 1.04542518,
      "balance_loss_mlp": 1.01628184,
      "epoch": 0.971562556363855,
      "flos": 20229917656320.0,
      "grad_norm": 1.730515160245731,
      "language_loss": 0.76170576,
      "learning_rate": 8.443091843558515e-09,
      "loss": 0.78343797,
      "num_input_tokens_seen": 174447960,
      "step": 8080,
      "time_per_iteration": 2.615691900253296
    },
    {
      "auxiliary_loss_clip": 0.01089446,
      "auxiliary_loss_mlp": 0.01023728,
      "balance_loss_clip": 1.03887248,
      "balance_loss_mlp": 1.01538968,
      "epoch": 0.9716827992544941,
      "flos": 24970197553920.0,
      "grad_norm": 2.3728759968617745,
      "language_loss": 0.65232027,
      "learning_rate": 8.37174166400878e-09,
      "loss": 0.67345202,
      "num_input_tokens_seen": 174463535,
      "step": 8081,
      "time_per_iteration": 2.8649325370788574
    },
    {
      "auxiliary_loss_clip": 0.01131572,
      "auxiliary_loss_mlp": 0.01031618,
      "balance_loss_clip": 1.0471431,
      "balance_loss_mlp": 1.02347636,
      "epoch": 0.9718030421451331,
      "flos": 24681476033280.0,
      "grad_norm": 2.0012105500538575,
      "language_loss": 0.85147405,
      "learning_rate": 8.300693612881992e-09,
      "loss": 0.873106,
      "num_input_tokens_seen": 174483600,
      "step": 8082,
      "time_per_iteration": 2.8293261528015137
    },
    {
      "auxiliary_loss_clip": 0.01088341,
      "auxiliary_loss_mlp": 0.01021779,
      "balance_loss_clip": 1.03981185,
      "balance_loss_mlp": 1.01417923,
      "epoch": 0.9719232850357723,
      "flos": 22090700793600.0,
      "grad_norm": 2.280306358112638,
      "language_loss": 0.8132571,
      "learning_rate": 8.22994770095664e-09,
      "loss": 0.83435827,
      "num_input_tokens_seen": 174502175,
      "step": 8083,
      "time_per_iteration": 2.81709885597229
    },
    {
      "auxiliary_loss_clip": 0.01127999,
      "auxiliary_loss_mlp": 0.01031169,
      "balance_loss_clip": 1.04677641,
      "balance_loss_mlp": 1.02263904,
      "epoch": 0.9720435279264114,
      "flos": 23656908493440.0,
      "grad_norm": 5.4747453548246785,
      "language_loss": 0.75383139,
      "learning_rate": 8.159503938964585e-09,
      "loss": 0.77542305,
      "num_input_tokens_seen": 174519495,
      "step": 8084,
      "time_per_iteration": 2.719034433364868
    },
    {
      "auxiliary_loss_clip": 0.0110164,
      "auxiliary_loss_mlp": 0.01025677,
      "balance_loss_clip": 1.03921103,
      "balance_loss_mlp": 1.01865888,
      "epoch": 0.9721637708170504,
      "flos": 28365910623360.0,
      "grad_norm": 2.1725635893552258,
      "language_loss": 0.70382392,
      "learning_rate": 8.089362337592164e-09,
      "loss": 0.72509706,
      "num_input_tokens_seen": 174543120,
      "step": 8085,
      "time_per_iteration": 2.812962770462036
    },
    {
      "auxiliary_loss_clip": 0.01147493,
      "auxiliary_loss_mlp": 0.01022509,
      "balance_loss_clip": 1.04400802,
      "balance_loss_mlp": 1.01485014,
      "epoch": 0.9722840137076896,
      "flos": 29130807767040.0,
      "grad_norm": 1.7098686217168737,
      "language_loss": 0.72472084,
      "learning_rate": 8.019522907479536e-09,
      "loss": 0.7464208,
      "num_input_tokens_seen": 174563480,
      "step": 8086,
      "time_per_iteration": 2.7373428344726562
    },
    {
      "auxiliary_loss_clip": 0.0112426,
      "auxiliary_loss_mlp": 0.01027993,
      "balance_loss_clip": 1.04371262,
      "balance_loss_mlp": 1.01994026,
      "epoch": 0.9724042565983286,
      "flos": 19243954258560.0,
      "grad_norm": 2.062618908321677,
      "language_loss": 0.77464294,
      "learning_rate": 7.949985659221558e-09,
      "loss": 0.79616547,
      "num_input_tokens_seen": 174580745,
      "step": 8087,
      "time_per_iteration": 2.6814920902252197
    },
    {
      "auxiliary_loss_clip": 0.0111443,
      "auxiliary_loss_mlp": 0.01025002,
      "balance_loss_clip": 1.04189706,
      "balance_loss_mlp": 1.01769471,
      "epoch": 0.9725244994889677,
      "flos": 23039676161280.0,
      "grad_norm": 1.9584331012499832,
      "language_loss": 0.79089808,
      "learning_rate": 7.880750603366904e-09,
      "loss": 0.81229246,
      "num_input_tokens_seen": 174599615,
      "step": 8088,
      "time_per_iteration": 2.73047137260437
    },
    {
      "auxiliary_loss_clip": 0.01124015,
      "auxiliary_loss_mlp": 0.01027442,
      "balance_loss_clip": 1.04324222,
      "balance_loss_mlp": 1.01908553,
      "epoch": 0.9726447423796069,
      "flos": 23367468700800.0,
      "grad_norm": 1.6995427810734838,
      "language_loss": 0.7930125,
      "learning_rate": 7.811817750418282e-09,
      "loss": 0.81452703,
      "num_input_tokens_seen": 174618375,
      "step": 8089,
      "time_per_iteration": 2.81733775138855
    },
    {
      "auxiliary_loss_clip": 0.01155721,
      "auxiliary_loss_mlp": 0.01027906,
      "balance_loss_clip": 1.04749584,
      "balance_loss_mlp": 1.0194838,
      "epoch": 0.9727649852702459,
      "flos": 26541648639360.0,
      "grad_norm": 2.389140689187668,
      "language_loss": 0.8047787,
      "learning_rate": 7.743187110833105e-09,
      "loss": 0.82661498,
      "num_input_tokens_seen": 174641135,
      "step": 8090,
      "time_per_iteration": 2.7638306617736816
    },
    {
      "auxiliary_loss_clip": 0.01133302,
      "auxiliary_loss_mlp": 0.01024924,
      "balance_loss_clip": 1.04172862,
      "balance_loss_mlp": 1.01722288,
      "epoch": 0.972885228160885,
      "flos": 20522338277760.0,
      "grad_norm": 1.6231310830268415,
      "language_loss": 0.80640692,
      "learning_rate": 7.674858695022602e-09,
      "loss": 0.82798922,
      "num_input_tokens_seen": 174659490,
      "step": 8091,
      "time_per_iteration": 2.636728525161743
    },
    {
      "auxiliary_loss_clip": 0.01124154,
      "auxiliary_loss_mlp": 0.00762589,
      "balance_loss_clip": 1.0424639,
      "balance_loss_mlp": 1.00048232,
      "epoch": 0.9730054710515241,
      "flos": 17566064196480.0,
      "grad_norm": 2.942812228575113,
      "language_loss": 0.76167518,
      "learning_rate": 7.606832513351591e-09,
      "loss": 0.78054267,
      "num_input_tokens_seen": 174677440,
      "step": 8092,
      "time_per_iteration": 2.7393887042999268
    },
    {
      "auxiliary_loss_clip": 0.01029358,
      "auxiliary_loss_mlp": 0.01007389,
      "balance_loss_clip": 1.01636899,
      "balance_loss_mlp": 1.00618458,
      "epoch": 0.9731257139421632,
      "flos": 68972010117120.0,
      "grad_norm": 0.8467562786446068,
      "language_loss": 0.63932741,
      "learning_rate": 7.539108576140264e-09,
      "loss": 0.65969491,
      "num_input_tokens_seen": 174741550,
      "step": 8093,
      "time_per_iteration": 3.316434144973755
    },
    {
      "auxiliary_loss_clip": 0.01078054,
      "auxiliary_loss_mlp": 0.0102046,
      "balance_loss_clip": 1.03430033,
      "balance_loss_mlp": 1.01297915,
      "epoch": 0.9732459568328022,
      "flos": 18478841633280.0,
      "grad_norm": 4.851717751356968,
      "language_loss": 0.7018609,
      "learning_rate": 7.471686893661732e-09,
      "loss": 0.72284603,
      "num_input_tokens_seen": 174759845,
      "step": 8094,
      "time_per_iteration": 2.8283913135528564
    },
    {
      "auxiliary_loss_clip": 0.01074606,
      "auxiliary_loss_mlp": 0.00762336,
      "balance_loss_clip": 1.04005194,
      "balance_loss_mlp": 1.00055814,
      "epoch": 0.9733661997234414,
      "flos": 20883886623360.0,
      "grad_norm": 2.0181580451362064,
      "language_loss": 0.64335287,
      "learning_rate": 7.4045674761442636e-09,
      "loss": 0.6617223,
      "num_input_tokens_seen": 174777175,
      "step": 8095,
      "time_per_iteration": 2.8243703842163086
    },
    {
      "auxiliary_loss_clip": 0.01168381,
      "auxiliary_loss_mlp": 0.010304,
      "balance_loss_clip": 1.04789257,
      "balance_loss_mlp": 1.02310753,
      "epoch": 0.9734864426140805,
      "flos": 23766795175680.0,
      "grad_norm": 2.1020409506631186,
      "language_loss": 0.74453628,
      "learning_rate": 7.337750333769488e-09,
      "loss": 0.76652408,
      "num_input_tokens_seen": 174796980,
      "step": 8096,
      "time_per_iteration": 3.498701572418213
    },
    {
      "auxiliary_loss_clip": 0.01061226,
      "auxiliary_loss_mlp": 0.01020763,
      "balance_loss_clip": 1.03443789,
      "balance_loss_mlp": 1.01245356,
      "epoch": 0.9736066855047195,
      "flos": 35042422176000.0,
      "grad_norm": 4.864505570709128,
      "language_loss": 0.73125792,
      "learning_rate": 7.2712354766737425e-09,
      "loss": 0.75207782,
      "num_input_tokens_seen": 174817310,
      "step": 8097,
      "time_per_iteration": 4.004864692687988
    },
    {
      "auxiliary_loss_clip": 0.01149882,
      "auxiliary_loss_mlp": 0.00762489,
      "balance_loss_clip": 1.04769421,
      "balance_loss_mlp": 1.00042653,
      "epoch": 0.9737269283953586,
      "flos": 20410620001920.0,
      "grad_norm": 1.839604531768323,
      "language_loss": 0.81078041,
      "learning_rate": 7.2050229149469565e-09,
      "loss": 0.82990408,
      "num_input_tokens_seen": 174837320,
      "step": 8098,
      "time_per_iteration": 4.241528511047363
    },
    {
      "auxiliary_loss_clip": 0.01120874,
      "auxiliary_loss_mlp": 0.01020648,
      "balance_loss_clip": 1.04065585,
      "balance_loss_mlp": 1.0127151,
      "epoch": 0.9738471712859977,
      "flos": 28911680847360.0,
      "grad_norm": 1.8303995321925528,
      "language_loss": 0.63582993,
      "learning_rate": 7.139112658633984e-09,
      "loss": 0.6572451,
      "num_input_tokens_seen": 174857470,
      "step": 8099,
      "time_per_iteration": 2.8671023845672607
    },
    {
      "auxiliary_loss_clip": 0.01135007,
      "auxiliary_loss_mlp": 0.01025296,
      "balance_loss_clip": 1.04245162,
      "balance_loss_mlp": 1.01705849,
      "epoch": 0.9739674141766368,
      "flos": 27782326356480.0,
      "grad_norm": 3.4510503430372874,
      "language_loss": 0.70406014,
      "learning_rate": 7.073504717733048e-09,
      "loss": 0.72566319,
      "num_input_tokens_seen": 174877035,
      "step": 8100,
      "time_per_iteration": 2.7676193714141846
    },
    {
      "auxiliary_loss_clip": 0.01004498,
      "auxiliary_loss_mlp": 0.01001024,
      "balance_loss_clip": 1.01113188,
      "balance_loss_mlp": 1.00010633,
      "epoch": 0.9740876570672758,
      "flos": 68863057188480.0,
      "grad_norm": 0.8854116340350127,
      "language_loss": 0.5709697,
      "learning_rate": 7.008199102196855e-09,
      "loss": 0.59102494,
      "num_input_tokens_seen": 174938460,
      "step": 8101,
      "time_per_iteration": 3.739116907119751
    },
    {
      "auxiliary_loss_clip": 0.0102916,
      "auxiliary_loss_mlp": 0.01000747,
      "balance_loss_clip": 1.00991189,
      "balance_loss_mlp": 0.99976367,
      "epoch": 0.974207899957915,
      "flos": 58236622646400.0,
      "grad_norm": 2.649569654168084,
      "language_loss": 0.58979785,
      "learning_rate": 6.9431958219321464e-09,
      "loss": 0.61009687,
      "num_input_tokens_seen": 174994625,
      "step": 8102,
      "time_per_iteration": 4.315386056900024
    },
    {
      "auxiliary_loss_clip": 0.01124142,
      "auxiliary_loss_mlp": 0.01030165,
      "balance_loss_clip": 1.04462886,
      "balance_loss_mlp": 1.02255929,
      "epoch": 0.9743281428485541,
      "flos": 22600057605120.0,
      "grad_norm": 1.6118981106985848,
      "language_loss": 0.77779251,
      "learning_rate": 6.878494886800146e-09,
      "loss": 0.79933554,
      "num_input_tokens_seen": 175015400,
      "step": 8103,
      "time_per_iteration": 2.8354105949401855
    },
    {
      "auxiliary_loss_clip": 0.01095791,
      "auxiliary_loss_mlp": 0.00762293,
      "balance_loss_clip": 1.03762138,
      "balance_loss_mlp": 1.00049925,
      "epoch": 0.9744483857391931,
      "flos": 20008815488640.0,
      "grad_norm": 1.8841628159406765,
      "language_loss": 0.76155925,
      "learning_rate": 6.814096306615669e-09,
      "loss": 0.78014004,
      "num_input_tokens_seen": 175033540,
      "step": 8104,
      "time_per_iteration": 3.9121599197387695
    },
    {
      "auxiliary_loss_clip": 0.01072429,
      "auxiliary_loss_mlp": 0.01027911,
      "balance_loss_clip": 1.03771579,
      "balance_loss_mlp": 1.01972139,
      "epoch": 0.9745686286298323,
      "flos": 17675268520320.0,
      "grad_norm": 2.364747773610803,
      "language_loss": 0.65330935,
      "learning_rate": 6.750000091148011e-09,
      "loss": 0.67431271,
      "num_input_tokens_seen": 175050835,
      "step": 8105,
      "time_per_iteration": 2.901024580001831
    },
    {
      "auxiliary_loss_clip": 0.01153687,
      "auxiliary_loss_mlp": 0.01024461,
      "balance_loss_clip": 1.04663599,
      "balance_loss_mlp": 1.01646757,
      "epoch": 0.9746888715204713,
      "flos": 29460252332160.0,
      "grad_norm": 1.807175417940045,
      "language_loss": 0.72630006,
      "learning_rate": 6.686206250120729e-09,
      "loss": 0.74808156,
      "num_input_tokens_seen": 175072330,
      "step": 8106,
      "time_per_iteration": 2.8778233528137207
    },
    {
      "auxiliary_loss_clip": 0.01109587,
      "auxiliary_loss_mlp": 0.01025427,
      "balance_loss_clip": 1.04332161,
      "balance_loss_mlp": 1.01776767,
      "epoch": 0.9748091144111104,
      "flos": 18479308510080.0,
      "grad_norm": 1.9363357282085176,
      "language_loss": 0.75033879,
      "learning_rate": 6.622714793210749e-09,
      "loss": 0.77168894,
      "num_input_tokens_seen": 175091250,
      "step": 8107,
      "time_per_iteration": 2.8396310806274414
    },
    {
      "auxiliary_loss_clip": 0.01133811,
      "auxiliary_loss_mlp": 0.01026411,
      "balance_loss_clip": 1.04276371,
      "balance_loss_mlp": 1.0188477,
      "epoch": 0.9749293573017496,
      "flos": 20665154753280.0,
      "grad_norm": 2.0361316263548166,
      "language_loss": 0.79049945,
      "learning_rate": 6.559525730050364e-09,
      "loss": 0.8121016,
      "num_input_tokens_seen": 175111350,
      "step": 8108,
      "time_per_iteration": 2.6731998920440674
    },
    {
      "auxiliary_loss_clip": 0.01143093,
      "auxiliary_loss_mlp": 0.01029566,
      "balance_loss_clip": 1.04563928,
      "balance_loss_mlp": 1.02182961,
      "epoch": 0.9750496001923886,
      "flos": 18478590238080.0,
      "grad_norm": 2.315230549779988,
      "language_loss": 0.761455,
      "learning_rate": 6.496639070224574e-09,
      "loss": 0.78318167,
      "num_input_tokens_seen": 175129835,
      "step": 8109,
      "time_per_iteration": 2.6875152587890625
    },
    {
      "auxiliary_loss_clip": 0.01083661,
      "auxiliary_loss_mlp": 0.0102238,
      "balance_loss_clip": 1.03497756,
      "balance_loss_mlp": 1.01529026,
      "epoch": 0.9751698430830277,
      "flos": 19572967860480.0,
      "grad_norm": 2.4411553523134297,
      "language_loss": 0.83502483,
      "learning_rate": 6.4340548232739714e-09,
      "loss": 0.8560853,
      "num_input_tokens_seen": 175146035,
      "step": 8110,
      "time_per_iteration": 2.7979676723480225
    },
    {
      "auxiliary_loss_clip": 0.01074194,
      "auxiliary_loss_mlp": 0.01022828,
      "balance_loss_clip": 1.0401268,
      "balance_loss_mlp": 1.01496339,
      "epoch": 0.9752900859736668,
      "flos": 23550325862400.0,
      "grad_norm": 2.7864893860070423,
      "language_loss": 0.79709041,
      "learning_rate": 6.371772998692071e-09,
      "loss": 0.81806058,
      "num_input_tokens_seen": 175165290,
      "step": 8111,
      "time_per_iteration": 2.8644509315490723
    },
    {
      "auxiliary_loss_clip": 0.01156126,
      "auxiliary_loss_mlp": 0.01023529,
      "balance_loss_clip": 1.04349458,
      "balance_loss_mlp": 1.01595891,
      "epoch": 0.9754103288643059,
      "flos": 20303211358080.0,
      "grad_norm": 3.467384543805792,
      "language_loss": 0.6488651,
      "learning_rate": 6.309793605927094e-09,
      "loss": 0.67066163,
      "num_input_tokens_seen": 175183610,
      "step": 8112,
      "time_per_iteration": 2.702648162841797
    },
    {
      "auxiliary_loss_clip": 0.01064222,
      "auxiliary_loss_mlp": 0.01023962,
      "balance_loss_clip": 1.03433204,
      "balance_loss_mlp": 1.01651406,
      "epoch": 0.975530571754945,
      "flos": 19350680544000.0,
      "grad_norm": 1.9010346427688307,
      "language_loss": 0.80351639,
      "learning_rate": 6.248116654381297e-09,
      "loss": 0.82439816,
      "num_input_tokens_seen": 175202080,
      "step": 8113,
      "time_per_iteration": 3.02219820022583
    },
    {
      "auxiliary_loss_clip": 0.01125761,
      "auxiliary_loss_mlp": 0.01021453,
      "balance_loss_clip": 1.04305339,
      "balance_loss_mlp": 1.01493859,
      "epoch": 0.9756508146455841,
      "flos": 23583399310080.0,
      "grad_norm": 1.7037555877496862,
      "language_loss": 0.7295177,
      "learning_rate": 6.186742153410751e-09,
      "loss": 0.75098991,
      "num_input_tokens_seen": 175221575,
      "step": 8114,
      "time_per_iteration": 3.295132637023926
    },
    {
      "auxiliary_loss_clip": 0.01138158,
      "auxiliary_loss_mlp": 0.0102494,
      "balance_loss_clip": 1.04430711,
      "balance_loss_mlp": 1.01692879,
      "epoch": 0.9757710575362232,
      "flos": 22966921163520.0,
      "grad_norm": 2.401730007015984,
      "language_loss": 0.87301683,
      "learning_rate": 6.125670112326453e-09,
      "loss": 0.89464778,
      "num_input_tokens_seen": 175240835,
      "step": 8115,
      "time_per_iteration": 2.7482476234436035
    },
    {
      "auxiliary_loss_clip": 0.01141107,
      "auxiliary_loss_mlp": 0.01029856,
      "balance_loss_clip": 1.04210448,
      "balance_loss_mlp": 1.0218513,
      "epoch": 0.9758913004268622,
      "flos": 27966009530880.0,
      "grad_norm": 3.8313567312373404,
      "language_loss": 0.70050389,
      "learning_rate": 6.064900540392548e-09,
      "loss": 0.72221351,
      "num_input_tokens_seen": 175262930,
      "step": 8116,
      "time_per_iteration": 2.8394463062286377
    },
    {
      "auxiliary_loss_clip": 0.01137834,
      "auxiliary_loss_mlp": 0.00760895,
      "balance_loss_clip": 1.04512811,
      "balance_loss_mlp": 1.00059915,
      "epoch": 0.9760115433175014,
      "flos": 22200156512640.0,
      "grad_norm": 4.598452702230232,
      "language_loss": 0.78922617,
      "learning_rate": 6.0044334468278835e-09,
      "loss": 0.80821347,
      "num_input_tokens_seen": 175282275,
      "step": 8117,
      "time_per_iteration": 2.915431022644043
    },
    {
      "auxiliary_loss_clip": 0.01158633,
      "auxiliary_loss_mlp": 0.01020076,
      "balance_loss_clip": 1.04606867,
      "balance_loss_mlp": 1.01216638,
      "epoch": 0.9761317862081405,
      "flos": 26250736389120.0,
      "grad_norm": 1.7471146376422215,
      "language_loss": 0.71599317,
      "learning_rate": 5.944268840805345e-09,
      "loss": 0.73778027,
      "num_input_tokens_seen": 175303020,
      "step": 8118,
      "time_per_iteration": 2.824903726577759
    },
    {
      "auxiliary_loss_clip": 0.01140635,
      "auxiliary_loss_mlp": 0.01022882,
      "balance_loss_clip": 1.04614615,
      "balance_loss_mlp": 1.01574147,
      "epoch": 0.9762520290987795,
      "flos": 26575440359040.0,
      "grad_norm": 4.140749478630509,
      "language_loss": 0.64410818,
      "learning_rate": 5.88440673145163e-09,
      "loss": 0.66574335,
      "num_input_tokens_seen": 175324070,
      "step": 8119,
      "time_per_iteration": 2.6564579010009766
    },
    {
      "auxiliary_loss_clip": 0.01138817,
      "auxiliary_loss_mlp": 0.01023288,
      "balance_loss_clip": 1.0462786,
      "balance_loss_mlp": 1.01528287,
      "epoch": 0.9763722719894187,
      "flos": 18005036307840.0,
      "grad_norm": 2.2878767723721545,
      "language_loss": 0.8276093,
      "learning_rate": 5.824847127848142e-09,
      "loss": 0.84923041,
      "num_input_tokens_seen": 175342595,
      "step": 8120,
      "time_per_iteration": 2.7057063579559326
    },
    {
      "auxiliary_loss_clip": 0.01149606,
      "auxiliary_loss_mlp": 0.01025741,
      "balance_loss_clip": 1.04434252,
      "balance_loss_mlp": 1.01857638,
      "epoch": 0.9764925148800577,
      "flos": 22455660931200.0,
      "grad_norm": 4.9459498648256,
      "language_loss": 0.79379904,
      "learning_rate": 5.765590039029433e-09,
      "loss": 0.81555253,
      "num_input_tokens_seen": 175361915,
      "step": 8121,
      "time_per_iteration": 2.6403636932373047
    },
    {
      "auxiliary_loss_clip": 0.01116728,
      "auxiliary_loss_mlp": 0.01027848,
      "balance_loss_clip": 1.04520345,
      "balance_loss_mlp": 1.02031374,
      "epoch": 0.9766127577706968,
      "flos": 36757084786560.0,
      "grad_norm": 1.7755461207818821,
      "language_loss": 0.70951235,
      "learning_rate": 5.706635473985422e-09,
      "loss": 0.73095804,
      "num_input_tokens_seen": 175385785,
      "step": 8122,
      "time_per_iteration": 3.7809078693389893
    },
    {
      "auxiliary_loss_clip": 0.01122869,
      "auxiliary_loss_mlp": 0.01025592,
      "balance_loss_clip": 1.03885055,
      "balance_loss_mlp": 1.01817691,
      "epoch": 0.976733000661336,
      "flos": 22309971367680.0,
      "grad_norm": 1.8897637362578104,
      "language_loss": 0.85147572,
      "learning_rate": 5.6479834416591764e-09,
      "loss": 0.87296033,
      "num_input_tokens_seen": 175405145,
      "step": 8123,
      "time_per_iteration": 3.579012393951416
    },
    {
      "auxiliary_loss_clip": 0.01126412,
      "auxiliary_loss_mlp": 0.01024058,
      "balance_loss_clip": 1.04213941,
      "balance_loss_mlp": 1.01605296,
      "epoch": 0.976853243551975,
      "flos": 25810938264960.0,
      "grad_norm": 2.1896918891401134,
      "language_loss": 0.68347132,
      "learning_rate": 5.589633950947803e-09,
      "loss": 0.70497596,
      "num_input_tokens_seen": 175422645,
      "step": 8124,
      "time_per_iteration": 3.654958486557007
    },
    {
      "auxiliary_loss_clip": 0.01158597,
      "auxiliary_loss_mlp": 0.0102476,
      "balance_loss_clip": 1.04630291,
      "balance_loss_mlp": 1.0167973,
      "epoch": 0.9769734864426141,
      "flos": 21397445326080.0,
      "grad_norm": 2.35524408242944,
      "language_loss": 0.69785976,
      "learning_rate": 5.5315870107035535e-09,
      "loss": 0.71969336,
      "num_input_tokens_seen": 175440695,
      "step": 8125,
      "time_per_iteration": 2.732046127319336
    },
    {
      "auxiliary_loss_clip": 0.0109501,
      "auxiliary_loss_mlp": 0.01025301,
      "balance_loss_clip": 1.03804648,
      "balance_loss_mlp": 1.01765728,
      "epoch": 0.9770937293332532,
      "flos": 13990977584640.0,
      "grad_norm": 1.95131414451376,
      "language_loss": 0.78581858,
      "learning_rate": 5.473842629731607e-09,
      "loss": 0.80702162,
      "num_input_tokens_seen": 175459195,
      "step": 8126,
      "time_per_iteration": 2.8069489002227783
    },
    {
      "auxiliary_loss_clip": 0.01120644,
      "auxiliary_loss_mlp": 0.01028315,
      "balance_loss_clip": 1.04633451,
      "balance_loss_mlp": 1.01964235,
      "epoch": 0.9772139722238923,
      "flos": 17931994001280.0,
      "grad_norm": 2.111043882116364,
      "language_loss": 0.77697974,
      "learning_rate": 5.416400816792066e-09,
      "loss": 0.79846931,
      "num_input_tokens_seen": 175476710,
      "step": 8127,
      "time_per_iteration": 2.8009586334228516
    },
    {
      "auxiliary_loss_clip": 0.0110312,
      "auxiliary_loss_mlp": 0.01026094,
      "balance_loss_clip": 1.04030764,
      "balance_loss_mlp": 1.01860809,
      "epoch": 0.9773342151145313,
      "flos": 20446171488000.0,
      "grad_norm": 3.501461624980779,
      "language_loss": 0.78771782,
      "learning_rate": 5.359261580598407e-09,
      "loss": 0.80900991,
      "num_input_tokens_seen": 175492550,
      "step": 8128,
      "time_per_iteration": 2.8158228397369385
    },
    {
      "auxiliary_loss_clip": 0.010899,
      "auxiliary_loss_mlp": 0.01033163,
      "balance_loss_clip": 1.04035616,
      "balance_loss_mlp": 1.02452636,
      "epoch": 0.9774544580051704,
      "flos": 11837306949120.0,
      "grad_norm": 2.3919465161665916,
      "language_loss": 0.78139514,
      "learning_rate": 5.302424929819027e-09,
      "loss": 0.80262578,
      "num_input_tokens_seen": 175506560,
      "step": 8129,
      "time_per_iteration": 2.769054412841797
    },
    {
      "auxiliary_loss_clip": 0.01145108,
      "auxiliary_loss_mlp": 0.01023776,
      "balance_loss_clip": 1.04152501,
      "balance_loss_mlp": 1.01614046,
      "epoch": 0.9775747008958096,
      "flos": 13479932833920.0,
      "grad_norm": 2.234525580099156,
      "language_loss": 0.73201728,
      "learning_rate": 5.24589087307592e-09,
      "loss": 0.7537061,
      "num_input_tokens_seen": 175524180,
      "step": 8130,
      "time_per_iteration": 3.671020269393921
    },
    {
      "auxiliary_loss_clip": 0.01138303,
      "auxiliary_loss_mlp": 0.01023387,
      "balance_loss_clip": 1.04565454,
      "balance_loss_mlp": 1.01604712,
      "epoch": 0.9776949437864486,
      "flos": 59532314042880.0,
      "grad_norm": 1.6191573484698578,
      "language_loss": 0.65076822,
      "learning_rate": 5.189659418944891e-09,
      "loss": 0.6723851,
      "num_input_tokens_seen": 175554355,
      "step": 8131,
      "time_per_iteration": 3.0367848873138428
    },
    {
      "auxiliary_loss_clip": 0.01154648,
      "auxiliary_loss_mlp": 0.00762476,
      "balance_loss_clip": 1.04982734,
      "balance_loss_mlp": 1.00042927,
      "epoch": 0.9778151866770877,
      "flos": 21178605715200.0,
      "grad_norm": 1.9687647233636476,
      "language_loss": 0.78556746,
      "learning_rate": 5.133730575956674e-09,
      "loss": 0.8047387,
      "num_input_tokens_seen": 175574025,
      "step": 8132,
      "time_per_iteration": 2.637779951095581
    },
    {
      "auxiliary_loss_clip": 0.01134373,
      "auxiliary_loss_mlp": 0.01024428,
      "balance_loss_clip": 1.04454851,
      "balance_loss_mlp": 1.01641083,
      "epoch": 0.9779354295677268,
      "flos": 20886795624960.0,
      "grad_norm": 2.1883436177301876,
      "language_loss": 0.72469652,
      "learning_rate": 5.0781043525953696e-09,
      "loss": 0.74628454,
      "num_input_tokens_seen": 175592090,
      "step": 8133,
      "time_per_iteration": 2.6617393493652344
    },
    {
      "auxiliary_loss_clip": 0.01139039,
      "auxiliary_loss_mlp": 0.01023929,
      "balance_loss_clip": 1.04417634,
      "balance_loss_mlp": 1.01690125,
      "epoch": 0.9780556724583659,
      "flos": 23440618748160.0,
      "grad_norm": 1.6958809253200773,
      "language_loss": 0.73766828,
      "learning_rate": 5.0227807572995605e-09,
      "loss": 0.75929797,
      "num_input_tokens_seen": 175614065,
      "step": 8134,
      "time_per_iteration": 2.6406123638153076
    },
    {
      "auxiliary_loss_clip": 0.01149736,
      "auxiliary_loss_mlp": 0.01024386,
      "balance_loss_clip": 1.04538465,
      "balance_loss_mlp": 1.01672733,
      "epoch": 0.9781759153490049,
      "flos": 20923244951040.0,
      "grad_norm": 2.9883320319738593,
      "language_loss": 0.6731683,
      "learning_rate": 4.967759798461646e-09,
      "loss": 0.69490945,
      "num_input_tokens_seen": 175632410,
      "step": 8135,
      "time_per_iteration": 2.6789355278015137
    },
    {
      "auxiliary_loss_clip": 0.01092109,
      "auxiliary_loss_mlp": 0.01022568,
      "balance_loss_clip": 1.03766418,
      "balance_loss_mlp": 1.01545727,
      "epoch": 0.9782961582396441,
      "flos": 28293191539200.0,
      "grad_norm": 3.611184276766915,
      "language_loss": 0.74905491,
      "learning_rate": 4.913041484428282e-09,
      "loss": 0.77020168,
      "num_input_tokens_seen": 175652885,
      "step": 8136,
      "time_per_iteration": 2.8399739265441895
    },
    {
      "auxiliary_loss_clip": 0.01138389,
      "auxiliary_loss_mlp": 0.01022529,
      "balance_loss_clip": 1.04442954,
      "balance_loss_mlp": 1.01529264,
      "epoch": 0.9784164011302832,
      "flos": 25552955808000.0,
      "grad_norm": 2.767368633716536,
      "language_loss": 0.73905694,
      "learning_rate": 4.858625823500384e-09,
      "loss": 0.76066613,
      "num_input_tokens_seen": 175670585,
      "step": 8137,
      "time_per_iteration": 2.6640772819519043
    },
    {
      "auxiliary_loss_clip": 0.01112632,
      "auxiliary_loss_mlp": 0.01022329,
      "balance_loss_clip": 1.03987598,
      "balance_loss_mlp": 1.01474166,
      "epoch": 0.9785366440209222,
      "flos": 29965945956480.0,
      "grad_norm": 1.913364932166085,
      "language_loss": 0.73296255,
      "learning_rate": 4.80451282393246e-09,
      "loss": 0.75431216,
      "num_input_tokens_seen": 175690570,
      "step": 8138,
      "time_per_iteration": 2.8498215675354004
    },
    {
      "auxiliary_loss_clip": 0.01158799,
      "auxiliary_loss_mlp": 0.01022853,
      "balance_loss_clip": 1.04849052,
      "balance_loss_mlp": 1.01535439,
      "epoch": 0.9786568869115614,
      "flos": 32343591847680.0,
      "grad_norm": 1.7833610235295794,
      "language_loss": 0.67626274,
      "learning_rate": 4.750702493933722e-09,
      "loss": 0.69807923,
      "num_input_tokens_seen": 175710455,
      "step": 8139,
      "time_per_iteration": 2.706942558288574
    },
    {
      "auxiliary_loss_clip": 0.01131884,
      "auxiliary_loss_mlp": 0.0102686,
      "balance_loss_clip": 1.0435102,
      "balance_loss_mlp": 1.01971912,
      "epoch": 0.9787771298022004,
      "flos": 23331414424320.0,
      "grad_norm": 2.7450380627939226,
      "language_loss": 0.85238779,
      "learning_rate": 4.697194841666974e-09,
      "loss": 0.87397516,
      "num_input_tokens_seen": 175729380,
      "step": 8140,
      "time_per_iteration": 2.6947314739227295
    },
    {
      "auxiliary_loss_clip": 0.01119137,
      "auxiliary_loss_mlp": 0.01029451,
      "balance_loss_clip": 1.04260969,
      "balance_loss_mlp": 1.02137399,
      "epoch": 0.9788973726928395,
      "flos": 21468548298240.0,
      "grad_norm": 1.9202137164170228,
      "language_loss": 0.82021272,
      "learning_rate": 4.6439898752492764e-09,
      "loss": 0.84169865,
      "num_input_tokens_seen": 175749520,
      "step": 8141,
      "time_per_iteration": 2.6855409145355225
    },
    {
      "auxiliary_loss_clip": 0.01031034,
      "auxiliary_loss_mlp": 0.01000046,
      "balance_loss_clip": 1.00876832,
      "balance_loss_mlp": 0.99905664,
      "epoch": 0.9790176155834787,
      "flos": 68897459439360.0,
      "grad_norm": 0.7437766402248096,
      "language_loss": 0.63623822,
      "learning_rate": 4.591087602751731e-09,
      "loss": 0.65654904,
      "num_input_tokens_seen": 175811380,
      "step": 8142,
      "time_per_iteration": 3.3300681114196777
    },
    {
      "auxiliary_loss_clip": 0.0113865,
      "auxiliary_loss_mlp": 0.01021145,
      "balance_loss_clip": 1.04279375,
      "balance_loss_mlp": 1.01393008,
      "epoch": 0.9791378584741177,
      "flos": 21430877909760.0,
      "grad_norm": 1.6469344773328471,
      "language_loss": 0.72244656,
      "learning_rate": 4.538488032199916e-09,
      "loss": 0.74404448,
      "num_input_tokens_seen": 175829480,
      "step": 8143,
      "time_per_iteration": 2.6528570652008057
    },
    {
      "auxiliary_loss_clip": 0.01130065,
      "auxiliary_loss_mlp": 0.01024746,
      "balance_loss_clip": 1.04451478,
      "balance_loss_mlp": 1.01693225,
      "epoch": 0.9792581013647568,
      "flos": 20153032594560.0,
      "grad_norm": 2.6525678509866633,
      "language_loss": 0.68768138,
      "learning_rate": 4.486191171572784e-09,
      "loss": 0.70922953,
      "num_input_tokens_seen": 175846750,
      "step": 8144,
      "time_per_iteration": 2.7213168144226074
    },
    {
      "auxiliary_loss_clip": 0.0107992,
      "auxiliary_loss_mlp": 0.01027437,
      "balance_loss_clip": 1.03585136,
      "balance_loss_mlp": 1.01975703,
      "epoch": 0.9793783442553959,
      "flos": 23728191033600.0,
      "grad_norm": 1.7867909138487075,
      "language_loss": 0.77784318,
      "learning_rate": 4.434197028803766e-09,
      "loss": 0.7989167,
      "num_input_tokens_seen": 175865975,
      "step": 8145,
      "time_per_iteration": 2.8977999687194824
    },
    {
      "auxiliary_loss_clip": 0.01141819,
      "auxiliary_loss_mlp": 0.01025892,
      "balance_loss_clip": 1.04306936,
      "balance_loss_mlp": 1.018471,
      "epoch": 0.979498587146035,
      "flos": 23038742407680.0,
      "grad_norm": 2.469162660147936,
      "language_loss": 0.8227765,
      "learning_rate": 4.3825056117805514e-09,
      "loss": 0.84445357,
      "num_input_tokens_seen": 175881860,
      "step": 8146,
      "time_per_iteration": 2.6807265281677246
    },
    {
      "auxiliary_loss_clip": 0.01168548,
      "auxiliary_loss_mlp": 0.01026467,
      "balance_loss_clip": 1.0461489,
      "balance_loss_mlp": 1.01902843,
      "epoch": 0.979618830036674,
      "flos": 14318841951360.0,
      "grad_norm": 2.6645564408293523,
      "language_loss": 0.79584819,
      "learning_rate": 4.331116928344425e-09,
      "loss": 0.81779826,
      "num_input_tokens_seen": 175898175,
      "step": 8147,
      "time_per_iteration": 2.691194534301758
    },
    {
      "auxiliary_loss_clip": 0.01102034,
      "auxiliary_loss_mlp": 0.01027777,
      "balance_loss_clip": 1.04182863,
      "balance_loss_mlp": 1.01999891,
      "epoch": 0.9797390729273132,
      "flos": 16727514215040.0,
      "grad_norm": 2.3323361861202816,
      "language_loss": 0.63071394,
      "learning_rate": 4.28003098629115e-09,
      "loss": 0.65201205,
      "num_input_tokens_seen": 175914310,
      "step": 8148,
      "time_per_iteration": 3.716252088546753
    },
    {
      "auxiliary_loss_clip": 0.01104595,
      "auxiliary_loss_mlp": 0.01023127,
      "balance_loss_clip": 1.03672361,
      "balance_loss_mlp": 1.01550996,
      "epoch": 0.9798593158179523,
      "flos": 24532661986560.0,
      "grad_norm": 2.787426184118382,
      "language_loss": 0.7858789,
      "learning_rate": 4.229247793370305e-09,
      "loss": 0.80715609,
      "num_input_tokens_seen": 175933435,
      "step": 8149,
      "time_per_iteration": 3.702073574066162
    },
    {
      "auxiliary_loss_clip": 0.0111673,
      "auxiliary_loss_mlp": 0.01028698,
      "balance_loss_clip": 1.04310393,
      "balance_loss_mlp": 1.02108693,
      "epoch": 0.9799795587085913,
      "flos": 27308808339840.0,
      "grad_norm": 1.7839306478095553,
      "language_loss": 0.70679682,
      "learning_rate": 4.178767357285951e-09,
      "loss": 0.7282511,
      "num_input_tokens_seen": 175955065,
      "step": 8150,
      "time_per_iteration": 2.8009490966796875
    },
    {
      "auxiliary_loss_clip": 0.01118183,
      "auxiliary_loss_mlp": 0.0102331,
      "balance_loss_clip": 1.04325438,
      "balance_loss_mlp": 1.01549566,
      "epoch": 0.9800998015992305,
      "flos": 26286575184000.0,
      "grad_norm": 2.135192418147983,
      "language_loss": 0.71496236,
      "learning_rate": 4.128589685695516e-09,
      "loss": 0.73637736,
      "num_input_tokens_seen": 175975490,
      "step": 8151,
      "time_per_iteration": 2.85919451713562
    },
    {
      "auxiliary_loss_clip": 0.01139095,
      "auxiliary_loss_mlp": 0.01022449,
      "balance_loss_clip": 1.04806161,
      "balance_loss_mlp": 1.01459324,
      "epoch": 0.9802200444898695,
      "flos": 16723635546240.0,
      "grad_norm": 2.413514634341404,
      "language_loss": 0.84552759,
      "learning_rate": 4.078714786211135e-09,
      "loss": 0.86714303,
      "num_input_tokens_seen": 175991340,
      "step": 8152,
      "time_per_iteration": 2.692779302597046
    },
    {
      "auxiliary_loss_clip": 0.01094133,
      "auxiliary_loss_mlp": 0.0102356,
      "balance_loss_clip": 1.03636408,
      "balance_loss_mlp": 1.01687217,
      "epoch": 0.9803402873805086,
      "flos": 24900459298560.0,
      "grad_norm": 1.7176993036109838,
      "language_loss": 0.76727796,
      "learning_rate": 4.029142666398977e-09,
      "loss": 0.78845489,
      "num_input_tokens_seen": 176011505,
      "step": 8153,
      "time_per_iteration": 2.7912979125976562
    },
    {
      "auxiliary_loss_clip": 0.01069527,
      "auxiliary_loss_mlp": 0.01024402,
      "balance_loss_clip": 1.03465223,
      "balance_loss_mlp": 1.017205,
      "epoch": 0.9804605302711478,
      "flos": 22564937082240.0,
      "grad_norm": 2.0424494913626927,
      "language_loss": 0.80410308,
      "learning_rate": 3.979873333778805e-09,
      "loss": 0.82504237,
      "num_input_tokens_seen": 176029680,
      "step": 8154,
      "time_per_iteration": 2.9750523567199707
    },
    {
      "auxiliary_loss_clip": 0.01076706,
      "auxiliary_loss_mlp": 0.01026653,
      "balance_loss_clip": 1.03659618,
      "balance_loss_mlp": 1.01803994,
      "epoch": 0.9805807731617868,
      "flos": 38905368382080.0,
      "grad_norm": 2.7519877028620883,
      "language_loss": 0.73809206,
      "learning_rate": 3.930906795824862e-09,
      "loss": 0.75912571,
      "num_input_tokens_seen": 176050355,
      "step": 8155,
      "time_per_iteration": 3.201256275177002
    },
    {
      "auxiliary_loss_clip": 0.01117199,
      "auxiliary_loss_mlp": 0.01027277,
      "balance_loss_clip": 1.04093933,
      "balance_loss_mlp": 1.01945066,
      "epoch": 0.9807010160524259,
      "flos": 17821999578240.0,
      "grad_norm": 2.405220774843006,
      "language_loss": 0.76759905,
      "learning_rate": 3.882243059965207e-09,
      "loss": 0.78904384,
      "num_input_tokens_seen": 176068070,
      "step": 8156,
      "time_per_iteration": 2.693127393722534
    },
    {
      "auxiliary_loss_clip": 0.0107081,
      "auxiliary_loss_mlp": 0.01025075,
      "balance_loss_clip": 1.03592253,
      "balance_loss_mlp": 1.01702833,
      "epoch": 0.980821258943065,
      "flos": 13552975140480.0,
      "grad_norm": 4.0027867672108535,
      "language_loss": 0.65701795,
      "learning_rate": 3.833882133582156e-09,
      "loss": 0.67797673,
      "num_input_tokens_seen": 176083730,
      "step": 8157,
      "time_per_iteration": 3.7236924171447754
    },
    {
      "auxiliary_loss_clip": 0.01120445,
      "auxiliary_loss_mlp": 0.0076206,
      "balance_loss_clip": 1.04541314,
      "balance_loss_mlp": 1.00047421,
      "epoch": 0.9809415018337041,
      "flos": 21689794120320.0,
      "grad_norm": 3.2599294486953005,
      "language_loss": 0.78232265,
      "learning_rate": 3.785824024012285e-09,
      "loss": 0.8011477,
      "num_input_tokens_seen": 176102730,
      "step": 8158,
      "time_per_iteration": 2.710052251815796
    },
    {
      "auxiliary_loss_clip": 0.01104154,
      "auxiliary_loss_mlp": 0.010226,
      "balance_loss_clip": 1.04336703,
      "balance_loss_mlp": 1.01562965,
      "epoch": 0.9810617447243432,
      "flos": 23294857357440.0,
      "grad_norm": 2.012470342457511,
      "language_loss": 0.78700709,
      "learning_rate": 3.738068738545541e-09,
      "loss": 0.80827457,
      "num_input_tokens_seen": 176121815,
      "step": 8159,
      "time_per_iteration": 2.7771224975585938
    },
    {
      "auxiliary_loss_clip": 0.01106522,
      "auxiliary_loss_mlp": 0.01030686,
      "balance_loss_clip": 1.0419842,
      "balance_loss_mlp": 1.02255583,
      "epoch": 0.9811819876149822,
      "flos": 18332038748160.0,
      "grad_norm": 2.3722215013635672,
      "language_loss": 0.78800207,
      "learning_rate": 3.6906162844265733e-09,
      "loss": 0.80937409,
      "num_input_tokens_seen": 176138900,
      "step": 8160,
      "time_per_iteration": 2.727283000946045
    },
    {
      "auxiliary_loss_clip": 0.01116237,
      "auxiliary_loss_mlp": 0.01025232,
      "balance_loss_clip": 1.04350209,
      "balance_loss_mlp": 1.01762378,
      "epoch": 0.9813022305056214,
      "flos": 22601961025920.0,
      "grad_norm": 1.9589486121431594,
      "language_loss": 0.70986652,
      "learning_rate": 3.643466668853845e-09,
      "loss": 0.73128116,
      "num_input_tokens_seen": 176156925,
      "step": 8161,
      "time_per_iteration": 2.7551870346069336
    },
    {
      "auxiliary_loss_clip": 0.01106724,
      "auxiliary_loss_mlp": 0.01022907,
      "balance_loss_clip": 1.04063308,
      "balance_loss_mlp": 1.01558125,
      "epoch": 0.9814224733962604,
      "flos": 25413335642880.0,
      "grad_norm": 2.144460985942178,
      "language_loss": 0.75387686,
      "learning_rate": 3.59661989898008e-09,
      "loss": 0.77517319,
      "num_input_tokens_seen": 176177980,
      "step": 8162,
      "time_per_iteration": 2.844827175140381
    },
    {
      "auxiliary_loss_clip": 0.01125234,
      "auxiliary_loss_mlp": 0.0102614,
      "balance_loss_clip": 1.04064834,
      "balance_loss_mlp": 1.0190618,
      "epoch": 0.9815427162868995,
      "flos": 25007185584000.0,
      "grad_norm": 2.691471899313185,
      "language_loss": 0.76969421,
      "learning_rate": 3.5500759819115934e-09,
      "loss": 0.79120791,
      "num_input_tokens_seen": 176198345,
      "step": 8163,
      "time_per_iteration": 2.813246488571167
    },
    {
      "auxiliary_loss_clip": 0.01154646,
      "auxiliary_loss_mlp": 0.01028067,
      "balance_loss_clip": 1.04944682,
      "balance_loss_mlp": 1.01951945,
      "epoch": 0.9816629591775387,
      "flos": 20662604887680.0,
      "grad_norm": 3.8315270636557277,
      "language_loss": 0.81103778,
      "learning_rate": 3.5038349247094034e-09,
      "loss": 0.83286488,
      "num_input_tokens_seen": 176215605,
      "step": 8164,
      "time_per_iteration": 2.723604679107666
    },
    {
      "auxiliary_loss_clip": 0.01112822,
      "auxiliary_loss_mlp": 0.01026719,
      "balance_loss_clip": 1.04088616,
      "balance_loss_mlp": 1.01927423,
      "epoch": 0.9817832020681777,
      "flos": 17712220636800.0,
      "grad_norm": 2.1521635027848967,
      "language_loss": 0.77654529,
      "learning_rate": 3.4578967343878994e-09,
      "loss": 0.79794067,
      "num_input_tokens_seen": 176231810,
      "step": 8165,
      "time_per_iteration": 2.688162326812744
    },
    {
      "auxiliary_loss_clip": 0.01151377,
      "auxiliary_loss_mlp": 0.0102457,
      "balance_loss_clip": 1.04467595,
      "balance_loss_mlp": 1.01731277,
      "epoch": 0.9819034449588168,
      "flos": 22530032040960.0,
      "grad_norm": 1.711465791278678,
      "language_loss": 0.81043452,
      "learning_rate": 3.4122614179161733e-09,
      "loss": 0.83219397,
      "num_input_tokens_seen": 176251770,
      "step": 8166,
      "time_per_iteration": 2.665083646774292
    },
    {
      "auxiliary_loss_clip": 0.01165945,
      "auxiliary_loss_mlp": 0.01019503,
      "balance_loss_clip": 1.04618835,
      "balance_loss_mlp": 1.01276505,
      "epoch": 0.9820236878494559,
      "flos": 20011221699840.0,
      "grad_norm": 1.8765436191848321,
      "language_loss": 0.7830559,
      "learning_rate": 3.36692898221691e-09,
      "loss": 0.80491042,
      "num_input_tokens_seen": 176270135,
      "step": 8167,
      "time_per_iteration": 2.6751549243927
    },
    {
      "auxiliary_loss_clip": 0.01139178,
      "auxiliary_loss_mlp": 0.01026982,
      "balance_loss_clip": 1.04538107,
      "balance_loss_mlp": 1.01926267,
      "epoch": 0.982143930740095,
      "flos": 18807316531200.0,
      "grad_norm": 2.462231035069759,
      "language_loss": 0.73754066,
      "learning_rate": 3.3218994341668305e-09,
      "loss": 0.7592023,
      "num_input_tokens_seen": 176289065,
      "step": 8168,
      "time_per_iteration": 2.7334141731262207
    },
    {
      "auxiliary_loss_clip": 0.01136099,
      "auxiliary_loss_mlp": 0.01018679,
      "balance_loss_clip": 1.04870069,
      "balance_loss_mlp": 1.01175857,
      "epoch": 0.982264173630734,
      "flos": 26578026138240.0,
      "grad_norm": 1.8859768193444797,
      "language_loss": 0.75898135,
      "learning_rate": 3.2771727805971373e-09,
      "loss": 0.78052908,
      "num_input_tokens_seen": 176310450,
      "step": 8169,
      "time_per_iteration": 2.736647129058838
    },
    {
      "auxiliary_loss_clip": 0.01120524,
      "auxiliary_loss_mlp": 0.01023716,
      "balance_loss_clip": 1.04089141,
      "balance_loss_mlp": 1.01566362,
      "epoch": 0.9823844165213732,
      "flos": 22014462176640.0,
      "grad_norm": 2.0329156202607006,
      "language_loss": 0.77264786,
      "learning_rate": 3.232749028292847e-09,
      "loss": 0.79409027,
      "num_input_tokens_seen": 176327415,
      "step": 8170,
      "time_per_iteration": 2.8431806564331055
    },
    {
      "auxiliary_loss_clip": 0.01113778,
      "auxiliary_loss_mlp": 0.01027018,
      "balance_loss_clip": 1.04066145,
      "balance_loss_mlp": 1.01895893,
      "epoch": 0.9825046594120123,
      "flos": 21908166854400.0,
      "grad_norm": 1.7526336361081787,
      "language_loss": 0.88519335,
      "learning_rate": 3.188628183992792e-09,
      "loss": 0.90660131,
      "num_input_tokens_seen": 176347680,
      "step": 8171,
      "time_per_iteration": 2.905452013015747
    },
    {
      "auxiliary_loss_clip": 0.01018526,
      "auxiliary_loss_mlp": 0.01002723,
      "balance_loss_clip": 1.01277685,
      "balance_loss_mlp": 1.0017817,
      "epoch": 0.9826249023026513,
      "flos": 59494610718720.0,
      "grad_norm": 0.7317872879426222,
      "language_loss": 0.62531054,
      "learning_rate": 3.1448102543902844e-09,
      "loss": 0.64552295,
      "num_input_tokens_seen": 176411595,
      "step": 8172,
      "time_per_iteration": 3.2577552795410156
    },
    {
      "auxiliary_loss_clip": 0.01131415,
      "auxiliary_loss_mlp": 0.01022441,
      "balance_loss_clip": 1.04330623,
      "balance_loss_mlp": 1.01463008,
      "epoch": 0.9827451451932905,
      "flos": 16071031296000.0,
      "grad_norm": 2.3984814829575902,
      "language_loss": 0.67414308,
      "learning_rate": 3.1012952461324515e-09,
      "loss": 0.69568169,
      "num_input_tokens_seen": 176430570,
      "step": 8173,
      "time_per_iteration": 2.726412296295166
    },
    {
      "auxiliary_loss_clip": 0.01166839,
      "auxiliary_loss_mlp": 0.0103111,
      "balance_loss_clip": 1.04710126,
      "balance_loss_mlp": 1.02343929,
      "epoch": 0.9828653880839295,
      "flos": 20262775622400.0,
      "grad_norm": 2.8190914370610027,
      "language_loss": 0.73872733,
      "learning_rate": 3.0580831658204575e-09,
      "loss": 0.76070684,
      "num_input_tokens_seen": 176448150,
      "step": 8174,
      "time_per_iteration": 3.930586576461792
    },
    {
      "auxiliary_loss_clip": 0.01053735,
      "auxiliary_loss_mlp": 0.01024919,
      "balance_loss_clip": 1.03687739,
      "balance_loss_mlp": 1.01825845,
      "epoch": 0.9829856309745686,
      "flos": 21616141282560.0,
      "grad_norm": 2.2560643115752455,
      "language_loss": 0.78268933,
      "learning_rate": 3.015174020009281e-09,
      "loss": 0.80347586,
      "num_input_tokens_seen": 176467475,
      "step": 8175,
      "time_per_iteration": 3.7705163955688477
    },
    {
      "auxiliary_loss_clip": 0.01114266,
      "auxiliary_loss_mlp": 0.01021208,
      "balance_loss_clip": 1.04369426,
      "balance_loss_mlp": 1.0144459,
      "epoch": 0.9831058738652078,
      "flos": 23764209396480.0,
      "grad_norm": 1.8318975415388026,
      "language_loss": 0.75066978,
      "learning_rate": 2.9725678152086043e-09,
      "loss": 0.77202451,
      "num_input_tokens_seen": 176486045,
      "step": 8176,
      "time_per_iteration": 2.7722604274749756
    },
    {
      "auxiliary_loss_clip": 0.01092077,
      "auxiliary_loss_mlp": 0.01023536,
      "balance_loss_clip": 1.04044247,
      "balance_loss_mlp": 1.01578188,
      "epoch": 0.9832261167558468,
      "flos": 11320911072000.0,
      "grad_norm": 8.161083726050123,
      "language_loss": 0.82745492,
      "learning_rate": 2.930264557881257e-09,
      "loss": 0.848611,
      "num_input_tokens_seen": 176501230,
      "step": 8177,
      "time_per_iteration": 2.7092416286468506
    },
    {
      "auxiliary_loss_clip": 0.01058019,
      "auxiliary_loss_mlp": 0.01001797,
      "balance_loss_clip": 1.01038671,
      "balance_loss_mlp": 1.00074756,
      "epoch": 0.9833463596464859,
      "flos": 60000304343040.0,
      "grad_norm": 0.8369411439969912,
      "language_loss": 0.58167464,
      "learning_rate": 2.8882642544452163e-09,
      "loss": 0.60227281,
      "num_input_tokens_seen": 176565955,
      "step": 8178,
      "time_per_iteration": 3.239825487136841
    },
    {
      "auxiliary_loss_clip": 0.01122091,
      "auxiliary_loss_mlp": 0.01020024,
      "balance_loss_clip": 1.0405705,
      "balance_loss_mlp": 1.01328897,
      "epoch": 0.983466602537125,
      "flos": 13626699805440.0,
      "grad_norm": 2.3610812636409984,
      "language_loss": 0.74549001,
      "learning_rate": 2.8465669112716083e-09,
      "loss": 0.76691115,
      "num_input_tokens_seen": 176583480,
      "step": 8179,
      "time_per_iteration": 2.7022995948791504
    },
    {
      "auxiliary_loss_clip": 0.01117767,
      "auxiliary_loss_mlp": 0.01022393,
      "balance_loss_clip": 1.04330802,
      "balance_loss_mlp": 1.0141108,
      "epoch": 0.9835868454277641,
      "flos": 22926844563840.0,
      "grad_norm": 1.9280729865785584,
      "language_loss": 0.76477885,
      "learning_rate": 2.8051725346858177e-09,
      "loss": 0.78618044,
      "num_input_tokens_seen": 176603740,
      "step": 8180,
      "time_per_iteration": 2.7818868160247803
    },
    {
      "auxiliary_loss_clip": 0.01154553,
      "auxiliary_loss_mlp": 0.01024599,
      "balance_loss_clip": 1.04463458,
      "balance_loss_mlp": 1.01663005,
      "epoch": 0.9837070883184031,
      "flos": 27673409341440.0,
      "grad_norm": 2.246475523228503,
      "language_loss": 0.70995855,
      "learning_rate": 2.7640811309674883e-09,
      "loss": 0.73175007,
      "num_input_tokens_seen": 176623240,
      "step": 8181,
      "time_per_iteration": 2.6833980083465576
    },
    {
      "auxiliary_loss_clip": 0.01110103,
      "auxiliary_loss_mlp": 0.01017776,
      "balance_loss_clip": 1.04345775,
      "balance_loss_mlp": 1.0106864,
      "epoch": 0.9838273312090423,
      "flos": 29241951425280.0,
      "grad_norm": 1.655269944310944,
      "language_loss": 0.80764365,
      "learning_rate": 2.7232927063498557e-09,
      "loss": 0.82892239,
      "num_input_tokens_seen": 176643615,
      "step": 8182,
      "time_per_iteration": 2.8455567359924316
    },
    {
      "auxiliary_loss_clip": 0.01141008,
      "auxiliary_loss_mlp": 0.01027865,
      "balance_loss_clip": 1.04473364,
      "balance_loss_mlp": 1.02040803,
      "epoch": 0.9839475740996814,
      "flos": 40110207304320.0,
      "grad_norm": 1.9267388188451584,
      "language_loss": 0.69421232,
      "learning_rate": 2.682807267020859e-09,
      "loss": 0.71590102,
      "num_input_tokens_seen": 176666375,
      "step": 8183,
      "time_per_iteration": 3.917966604232788
    },
    {
      "auxiliary_loss_clip": 0.0116733,
      "auxiliary_loss_mlp": 0.010226,
      "balance_loss_clip": 1.04585099,
      "balance_loss_mlp": 1.01436818,
      "epoch": 0.9840678169903204,
      "flos": 24169389788160.0,
      "grad_norm": 1.770840609454645,
      "language_loss": 0.62231982,
      "learning_rate": 2.642624819121808e-09,
      "loss": 0.6442191,
      "num_input_tokens_seen": 176686525,
      "step": 8184,
      "time_per_iteration": 2.7490603923797607
    },
    {
      "auxiliary_loss_clip": 0.01130351,
      "auxiliary_loss_mlp": 0.01023099,
      "balance_loss_clip": 1.04281628,
      "balance_loss_mlp": 1.01575541,
      "epoch": 0.9841880598809596,
      "flos": 14684484447360.0,
      "grad_norm": 2.2224597197666025,
      "language_loss": 0.61910594,
      "learning_rate": 2.6027453687487154e-09,
      "loss": 0.64064038,
      "num_input_tokens_seen": 176703615,
      "step": 8185,
      "time_per_iteration": 2.6568500995635986
    },
    {
      "auxiliary_loss_clip": 0.01108031,
      "auxiliary_loss_mlp": 0.01026844,
      "balance_loss_clip": 1.03948259,
      "balance_loss_mlp": 1.01861227,
      "epoch": 0.9843083027715986,
      "flos": 22344768668160.0,
      "grad_norm": 4.6840194842519205,
      "language_loss": 0.54626185,
      "learning_rate": 2.5631689219509643e-09,
      "loss": 0.56761062,
      "num_input_tokens_seen": 176722295,
      "step": 8186,
      "time_per_iteration": 2.747939348220825
    },
    {
      "auxiliary_loss_clip": 0.0113443,
      "auxiliary_loss_mlp": 0.0102668,
      "balance_loss_clip": 1.04459381,
      "balance_loss_mlp": 1.01977515,
      "epoch": 0.9844285456622377,
      "flos": 21800111765760.0,
      "grad_norm": 2.5984082391445087,
      "language_loss": 0.83617687,
      "learning_rate": 2.523895484732197e-09,
      "loss": 0.85778797,
      "num_input_tokens_seen": 176741750,
      "step": 8187,
      "time_per_iteration": 2.6869614124298096
    },
    {
      "auxiliary_loss_clip": 0.01143888,
      "auxiliary_loss_mlp": 0.00761986,
      "balance_loss_clip": 1.04680753,
      "balance_loss_mlp": 1.00033736,
      "epoch": 0.9845487885528769,
      "flos": 18035380321920.0,
      "grad_norm": 4.918108963490269,
      "language_loss": 0.74918449,
      "learning_rate": 2.4849250630505357e-09,
      "loss": 0.76824319,
      "num_input_tokens_seen": 176759995,
      "step": 8188,
      "time_per_iteration": 2.751999855041504
    },
    {
      "auxiliary_loss_clip": 0.01120508,
      "auxiliary_loss_mlp": 0.0102448,
      "balance_loss_clip": 1.04309952,
      "balance_loss_mlp": 1.01699102,
      "epoch": 0.9846690314435159,
      "flos": 25228610974080.0,
      "grad_norm": 1.9439465075167965,
      "language_loss": 0.73432243,
      "learning_rate": 2.4462576628172528e-09,
      "loss": 0.75577223,
      "num_input_tokens_seen": 176778625,
      "step": 8189,
      "time_per_iteration": 2.7830793857574463
    },
    {
      "auxiliary_loss_clip": 0.01126013,
      "auxiliary_loss_mlp": 0.01024752,
      "balance_loss_clip": 1.04515278,
      "balance_loss_mlp": 1.01759696,
      "epoch": 0.984789274334155,
      "flos": 18552171248640.0,
      "grad_norm": 2.023133752272113,
      "language_loss": 0.74280274,
      "learning_rate": 2.407893289898766e-09,
      "loss": 0.76431036,
      "num_input_tokens_seen": 176797655,
      "step": 8190,
      "time_per_iteration": 2.728640079498291
    },
    {
      "auxiliary_loss_clip": 0.01139474,
      "auxiliary_loss_mlp": 0.01025539,
      "balance_loss_clip": 1.04320621,
      "balance_loss_mlp": 1.01793003,
      "epoch": 0.984909517224794,
      "flos": 27345437233920.0,
      "grad_norm": 2.673623407070034,
      "language_loss": 0.8418138,
      "learning_rate": 2.3698319501144202e-09,
      "loss": 0.863464,
      "num_input_tokens_seen": 176818640,
      "step": 8191,
      "time_per_iteration": 2.781122922897339
    },
    {
      "auxiliary_loss_clip": 0.01117026,
      "auxiliary_loss_mlp": 0.01023551,
      "balance_loss_clip": 1.0416106,
      "balance_loss_mlp": 1.01502168,
      "epoch": 0.9850297601154332,
      "flos": 18734058743040.0,
      "grad_norm": 2.893978661566953,
      "language_loss": 0.73806989,
      "learning_rate": 2.3320736492382644e-09,
      "loss": 0.75947565,
      "num_input_tokens_seen": 176837475,
      "step": 8192,
      "time_per_iteration": 2.694711685180664
    },
    {
      "auxiliary_loss_clip": 0.01126735,
      "auxiliary_loss_mlp": 0.00761464,
      "balance_loss_clip": 1.04144108,
      "balance_loss_mlp": 1.0004003,
      "epoch": 0.9851500030060723,
      "flos": 22308247514880.0,
      "grad_norm": 1.6337450254128425,
      "language_loss": 0.68631089,
      "learning_rate": 2.29461839299816e-09,
      "loss": 0.70519292,
      "num_input_tokens_seen": 176857190,
      "step": 8193,
      "time_per_iteration": 2.6960246562957764
    },
    {
      "auxiliary_loss_clip": 0.01130184,
      "auxiliary_loss_mlp": 0.01027413,
      "balance_loss_clip": 1.04320002,
      "balance_loss_mlp": 1.0201745,
      "epoch": 0.9852702458967113,
      "flos": 26353691746560.0,
      "grad_norm": 2.104740288548236,
      "language_loss": 0.79855919,
      "learning_rate": 2.257466187076229e-09,
      "loss": 0.82013518,
      "num_input_tokens_seen": 176876395,
      "step": 8194,
      "time_per_iteration": 2.6955466270446777
    },
    {
      "auxiliary_loss_clip": 0.01129052,
      "auxiliary_loss_mlp": 0.01022761,
      "balance_loss_clip": 1.04379106,
      "balance_loss_mlp": 1.01475596,
      "epoch": 0.9853904887873505,
      "flos": 20883599314560.0,
      "grad_norm": 1.8611053977701317,
      "language_loss": 0.71730423,
      "learning_rate": 2.2206170371081854e-09,
      "loss": 0.73882234,
      "num_input_tokens_seen": 176894980,
      "step": 8195,
      "time_per_iteration": 2.770660877227783
    },
    {
      "auxiliary_loss_clip": 0.01120789,
      "auxiliary_loss_mlp": 0.01027533,
      "balance_loss_clip": 1.04052794,
      "balance_loss_mlp": 1.01949787,
      "epoch": 0.9855107316779895,
      "flos": 25263444188160.0,
      "grad_norm": 1.8703973315797122,
      "language_loss": 0.84949791,
      "learning_rate": 2.1840709486842247e-09,
      "loss": 0.87098122,
      "num_input_tokens_seen": 176914600,
      "step": 8196,
      "time_per_iteration": 2.759788751602173
    },
    {
      "auxiliary_loss_clip": 0.01145913,
      "auxiliary_loss_mlp": 0.01026829,
      "balance_loss_clip": 1.04251909,
      "balance_loss_mlp": 1.01890194,
      "epoch": 0.9856309745686286,
      "flos": 19062102677760.0,
      "grad_norm": 1.967430795803385,
      "language_loss": 0.79780078,
      "learning_rate": 2.1478279273481335e-09,
      "loss": 0.81952822,
      "num_input_tokens_seen": 176933085,
      "step": 8197,
      "time_per_iteration": 2.6750411987304688
    },
    {
      "auxiliary_loss_clip": 0.01117095,
      "auxiliary_loss_mlp": 0.01026709,
      "balance_loss_clip": 1.0439055,
      "balance_loss_mlp": 1.01886487,
      "epoch": 0.9857512174592677,
      "flos": 34130758060800.0,
      "grad_norm": 2.118346592424291,
      "language_loss": 0.80364573,
      "learning_rate": 2.1118879785981815e-09,
      "loss": 0.82508379,
      "num_input_tokens_seen": 176953225,
      "step": 8198,
      "time_per_iteration": 2.9527101516723633
    },
    {
      "auxiliary_loss_clip": 0.0112486,
      "auxiliary_loss_mlp": 0.01020556,
      "balance_loss_clip": 1.04237056,
      "balance_loss_mlp": 1.0136832,
      "epoch": 0.9858714603499068,
      "flos": 25994693266560.0,
      "grad_norm": 1.8775642586458088,
      "language_loss": 0.79595006,
      "learning_rate": 2.0762511078862288e-09,
      "loss": 0.81740421,
      "num_input_tokens_seen": 176973570,
      "step": 8199,
      "time_per_iteration": 3.7794573307037354
    },
    {
      "auxiliary_loss_clip": 0.01095169,
      "auxiliary_loss_mlp": 0.01028859,
      "balance_loss_clip": 1.04140735,
      "balance_loss_mlp": 1.02078271,
      "epoch": 0.9859917032405459,
      "flos": 23696230907520.0,
      "grad_norm": 2.103357818166951,
      "language_loss": 0.65050375,
      "learning_rate": 2.0409173206186183e-09,
      "loss": 0.67174411,
      "num_input_tokens_seen": 176992810,
      "step": 8200,
      "time_per_iteration": 2.856308937072754
    },
    {
      "auxiliary_loss_clip": 0.0114073,
      "auxiliary_loss_mlp": 0.01024213,
      "balance_loss_clip": 1.04496396,
      "balance_loss_mlp": 1.01701307,
      "epoch": 0.986111946131185,
      "flos": 19938287134080.0,
      "grad_norm": 2.042237242585777,
      "language_loss": 0.87155402,
      "learning_rate": 2.0058866221550617e-09,
      "loss": 0.8932035,
      "num_input_tokens_seen": 177011050,
      "step": 8201,
      "time_per_iteration": 4.611984014511108
    },
    {
      "auxiliary_loss_clip": 0.0114514,
      "auxiliary_loss_mlp": 0.01020917,
      "balance_loss_clip": 1.04175878,
      "balance_loss_mlp": 1.01334167,
      "epoch": 0.9862321890218241,
      "flos": 19828831415040.0,
      "grad_norm": 2.2501037436826916,
      "language_loss": 0.74832946,
      "learning_rate": 1.971159017809976e-09,
      "loss": 0.76998997,
      "num_input_tokens_seen": 177029340,
      "step": 8202,
      "time_per_iteration": 2.7132134437561035
    },
    {
      "auxiliary_loss_clip": 0.01137324,
      "auxiliary_loss_mlp": 0.01028953,
      "balance_loss_clip": 1.04261327,
      "balance_loss_mlp": 1.02122819,
      "epoch": 0.9863524319124631,
      "flos": 21652051904640.0,
      "grad_norm": 2.176809659042925,
      "language_loss": 0.77993524,
      "learning_rate": 1.93673451285159e-09,
      "loss": 0.80159795,
      "num_input_tokens_seen": 177048390,
      "step": 8203,
      "time_per_iteration": 2.681992530822754
    },
    {
      "auxiliary_loss_clip": 0.0102483,
      "auxiliary_loss_mlp": 0.01000473,
      "balance_loss_clip": 1.00926757,
      "balance_loss_mlp": 0.99939388,
      "epoch": 0.9864726748031023,
      "flos": 52769977920000.0,
      "grad_norm": 0.7342844420934571,
      "language_loss": 0.56529665,
      "learning_rate": 1.9026131125019495e-09,
      "loss": 0.58554965,
      "num_input_tokens_seen": 177105760,
      "step": 8204,
      "time_per_iteration": 3.21983003616333
    },
    {
      "auxiliary_loss_clip": 0.01105744,
      "auxiliary_loss_mlp": 0.01024452,
      "balance_loss_clip": 1.03745556,
      "balance_loss_mlp": 1.01725173,
      "epoch": 0.9865929176937414,
      "flos": 23364631526400.0,
      "grad_norm": 2.101791646429402,
      "language_loss": 0.87225449,
      "learning_rate": 1.8687948219371363e-09,
      "loss": 0.89355642,
      "num_input_tokens_seen": 177124985,
      "step": 8205,
      "time_per_iteration": 2.7747581005096436
    },
    {
      "auxiliary_loss_clip": 0.01143702,
      "auxiliary_loss_mlp": 0.01023509,
      "balance_loss_clip": 1.04238081,
      "balance_loss_mlp": 1.01517582,
      "epoch": 0.9867131605843804,
      "flos": 21616679986560.0,
      "grad_norm": 2.210830298376718,
      "language_loss": 0.88769048,
      "learning_rate": 1.835279646287491e-09,
      "loss": 0.90936255,
      "num_input_tokens_seen": 177142995,
      "step": 8206,
      "time_per_iteration": 2.6576104164123535
    },
    {
      "auxiliary_loss_clip": 0.01142055,
      "auxiliary_loss_mlp": 0.01029288,
      "balance_loss_clip": 1.04822838,
      "balance_loss_mlp": 1.0212357,
      "epoch": 0.9868334034750196,
      "flos": 22271403139200.0,
      "grad_norm": 2.026404718836563,
      "language_loss": 0.76844507,
      "learning_rate": 1.8020675906371685e-09,
      "loss": 0.79015851,
      "num_input_tokens_seen": 177162390,
      "step": 8207,
      "time_per_iteration": 2.676474094390869
    },
    {
      "auxiliary_loss_clip": 0.0113225,
      "auxiliary_loss_mlp": 0.01023763,
      "balance_loss_clip": 1.04279661,
      "balance_loss_mlp": 1.01540029,
      "epoch": 0.9869536463656586,
      "flos": 25809573548160.0,
      "grad_norm": 4.037365270077843,
      "language_loss": 0.74991989,
      "learning_rate": 1.7691586600243612e-09,
      "loss": 0.77148008,
      "num_input_tokens_seen": 177181290,
      "step": 8208,
      "time_per_iteration": 3.677286386489868
    },
    {
      "auxiliary_loss_clip": 0.01109415,
      "auxiliary_loss_mlp": 0.01021404,
      "balance_loss_clip": 1.03910089,
      "balance_loss_mlp": 1.01374829,
      "epoch": 0.9870738892562977,
      "flos": 16398500613120.0,
      "grad_norm": 2.61547870875552,
      "language_loss": 0.87553144,
      "learning_rate": 1.7365528594415202e-09,
      "loss": 0.89683962,
      "num_input_tokens_seen": 177195360,
      "step": 8209,
      "time_per_iteration": 2.694946765899658
    },
    {
      "auxiliary_loss_clip": 0.01127697,
      "auxiliary_loss_mlp": 0.01022807,
      "balance_loss_clip": 1.04175055,
      "balance_loss_mlp": 1.01533008,
      "epoch": 0.9871941321469369,
      "flos": 35481358373760.0,
      "grad_norm": 2.419430119929413,
      "language_loss": 0.6775918,
      "learning_rate": 1.7042501938346888e-09,
      "loss": 0.69909686,
      "num_input_tokens_seen": 177218090,
      "step": 8210,
      "time_per_iteration": 2.8232345581054688
    },
    {
      "auxiliary_loss_clip": 0.01112225,
      "auxiliary_loss_mlp": 0.01022634,
      "balance_loss_clip": 1.0398488,
      "balance_loss_mlp": 1.01530302,
      "epoch": 0.9873143750375759,
      "flos": 21434217874560.0,
      "grad_norm": 2.0705425109243834,
      "language_loss": 0.76961392,
      "learning_rate": 1.6722506681043913e-09,
      "loss": 0.79096246,
      "num_input_tokens_seen": 177237050,
      "step": 8211,
      "time_per_iteration": 2.723874807357788
    },
    {
      "auxiliary_loss_clip": 0.01120986,
      "auxiliary_loss_mlp": 0.01023334,
      "balance_loss_clip": 1.04354048,
      "balance_loss_mlp": 1.01580596,
      "epoch": 0.987434617928215,
      "flos": 16326499800960.0,
      "grad_norm": 2.140687955614431,
      "language_loss": 0.69211322,
      "learning_rate": 1.640554287104745e-09,
      "loss": 0.71355647,
      "num_input_tokens_seen": 177255325,
      "step": 8212,
      "time_per_iteration": 2.800013780593872
    },
    {
      "auxiliary_loss_clip": 0.01150358,
      "auxiliary_loss_mlp": 0.01021633,
      "balance_loss_clip": 1.04411006,
      "balance_loss_mlp": 1.01359487,
      "epoch": 0.9875548608188541,
      "flos": 17851984456320.0,
      "grad_norm": 3.0324692893131395,
      "language_loss": 0.8011567,
      "learning_rate": 1.609161055644348e-09,
      "loss": 0.82287663,
      "num_input_tokens_seen": 177271250,
      "step": 8213,
      "time_per_iteration": 2.6137046813964844
    },
    {
      "auxiliary_loss_clip": 0.01134471,
      "auxiliary_loss_mlp": 0.01021949,
      "balance_loss_clip": 1.04303634,
      "balance_loss_mlp": 1.01361012,
      "epoch": 0.9876751037094932,
      "flos": 26132876887680.0,
      "grad_norm": 12.183291178058246,
      "language_loss": 0.68490183,
      "learning_rate": 1.5780709784849467e-09,
      "loss": 0.70646602,
      "num_input_tokens_seen": 177288270,
      "step": 8214,
      "time_per_iteration": 2.7973275184631348
    },
    {
      "auxiliary_loss_clip": 0.01137422,
      "auxiliary_loss_mlp": 0.01032852,
      "balance_loss_clip": 1.04662967,
      "balance_loss_mlp": 1.02459121,
      "epoch": 0.9877953466001322,
      "flos": 15991344973440.0,
      "grad_norm": 2.15330726257056,
      "language_loss": 0.82345611,
      "learning_rate": 1.5472840603436565e-09,
      "loss": 0.84515887,
      "num_input_tokens_seen": 177305500,
      "step": 8215,
      "time_per_iteration": 2.688744068145752
    },
    {
      "auxiliary_loss_clip": 0.01111196,
      "auxiliary_loss_mlp": 0.01025215,
      "balance_loss_clip": 1.03842282,
      "balance_loss_mlp": 1.01778483,
      "epoch": 0.9879155894907714,
      "flos": 18806777827200.0,
      "grad_norm": 2.1560434408915814,
      "language_loss": 0.7797482,
      "learning_rate": 1.5168003058900757e-09,
      "loss": 0.80111235,
      "num_input_tokens_seen": 177323500,
      "step": 8216,
      "time_per_iteration": 2.7086734771728516
    },
    {
      "auxiliary_loss_clip": 0.01080457,
      "auxiliary_loss_mlp": 0.01024998,
      "balance_loss_clip": 1.03877687,
      "balance_loss_mlp": 1.01726723,
      "epoch": 0.9880358323814105,
      "flos": 22382044007040.0,
      "grad_norm": 1.9693956457026858,
      "language_loss": 0.92051315,
      "learning_rate": 1.4866197197491715e-09,
      "loss": 0.94156772,
      "num_input_tokens_seen": 177342860,
      "step": 8217,
      "time_per_iteration": 2.9065728187561035
    },
    {
      "auxiliary_loss_clip": 0.01135739,
      "auxiliary_loss_mlp": 0.01034142,
      "balance_loss_clip": 1.04631114,
      "balance_loss_mlp": 1.02547538,
      "epoch": 0.9881560752720495,
      "flos": 15668831733120.0,
      "grad_norm": 2.9263147514601693,
      "language_loss": 0.78235626,
      "learning_rate": 1.4567423064988371e-09,
      "loss": 0.80405509,
      "num_input_tokens_seen": 177360210,
      "step": 8218,
      "time_per_iteration": 2.6454222202301025
    },
    {
      "auxiliary_loss_clip": 0.01119601,
      "auxiliary_loss_mlp": 0.01032558,
      "balance_loss_clip": 1.04665637,
      "balance_loss_mlp": 1.02422547,
      "epoch": 0.9882763181626887,
      "flos": 21500113374720.0,
      "grad_norm": 2.2518786020254313,
      "language_loss": 0.78434241,
      "learning_rate": 1.4271680706718913e-09,
      "loss": 0.80586404,
      "num_input_tokens_seen": 177377885,
      "step": 8219,
      "time_per_iteration": 2.756058931350708
    },
    {
      "auxiliary_loss_clip": 0.01132133,
      "auxiliary_loss_mlp": 0.01026254,
      "balance_loss_clip": 1.0461905,
      "balance_loss_mlp": 1.01788545,
      "epoch": 0.9883965610533277,
      "flos": 28034598551040.0,
      "grad_norm": 1.8197975463886082,
      "language_loss": 0.82715893,
      "learning_rate": 1.3978970167543013e-09,
      "loss": 0.84874284,
      "num_input_tokens_seen": 177398065,
      "step": 8220,
      "time_per_iteration": 2.7361600399017334
    },
    {
      "auxiliary_loss_clip": 0.01131359,
      "auxiliary_loss_mlp": 0.01026401,
      "balance_loss_clip": 1.04385519,
      "balance_loss_mlp": 1.01825643,
      "epoch": 0.9885168039439668,
      "flos": 14098601710080.0,
      "grad_norm": 28.080031177353835,
      "language_loss": 0.77797639,
      "learning_rate": 1.3689291491867372e-09,
      "loss": 0.79955399,
      "num_input_tokens_seen": 177416380,
      "step": 8221,
      "time_per_iteration": 2.687563180923462
    },
    {
      "auxiliary_loss_clip": 0.01155397,
      "auxiliary_loss_mlp": 0.01027637,
      "balance_loss_clip": 1.04597855,
      "balance_loss_mlp": 1.01981115,
      "epoch": 0.988637046834606,
      "flos": 26432013352320.0,
      "grad_norm": 4.265945155633227,
      "language_loss": 0.73925185,
      "learning_rate": 1.3402644723636836e-09,
      "loss": 0.76108217,
      "num_input_tokens_seen": 177438410,
      "step": 8222,
      "time_per_iteration": 2.648322820663452
    },
    {
      "auxiliary_loss_clip": 0.01088547,
      "auxiliary_loss_mlp": 0.01022833,
      "balance_loss_clip": 1.03941798,
      "balance_loss_mlp": 1.01492345,
      "epoch": 0.988757289725245,
      "flos": 25229113764480.0,
      "grad_norm": 6.801860134690427,
      "language_loss": 0.8394568,
      "learning_rate": 1.311902990633218e-09,
      "loss": 0.86057067,
      "num_input_tokens_seen": 177457375,
      "step": 8223,
      "time_per_iteration": 2.875194787979126
    },
    {
      "auxiliary_loss_clip": 0.01121026,
      "auxiliary_loss_mlp": 0.0102314,
      "balance_loss_clip": 1.0418607,
      "balance_loss_mlp": 1.01602352,
      "epoch": 0.9888775326158841,
      "flos": 26359042872960.0,
      "grad_norm": 1.7137287418195306,
      "language_loss": 0.71370369,
      "learning_rate": 1.2838447082978987e-09,
      "loss": 0.73514539,
      "num_input_tokens_seen": 177478530,
      "step": 8224,
      "time_per_iteration": 2.7321596145629883
    },
    {
      "auxiliary_loss_clip": 0.01111353,
      "auxiliary_loss_mlp": 0.0102394,
      "balance_loss_clip": 1.03956115,
      "balance_loss_mlp": 1.01633477,
      "epoch": 0.9889977755065231,
      "flos": 24316120846080.0,
      "grad_norm": 95.36441919643734,
      "language_loss": 0.83631176,
      "learning_rate": 1.2560896296143208e-09,
      "loss": 0.85766464,
      "num_input_tokens_seen": 177496995,
      "step": 8225,
      "time_per_iteration": 2.7777445316314697
    },
    {
      "auxiliary_loss_clip": 0.01119551,
      "auxiliary_loss_mlp": 0.01024271,
      "balance_loss_clip": 1.04266644,
      "balance_loss_mlp": 1.0167073,
      "epoch": 0.9891180183971623,
      "flos": 18951066760320.0,
      "grad_norm": 2.4992113129611817,
      "language_loss": 0.82486105,
      "learning_rate": 1.2286377587926722e-09,
      "loss": 0.84629929,
      "num_input_tokens_seen": 177513785,
      "step": 8226,
      "time_per_iteration": 3.6259584426879883
    },
    {
      "auxiliary_loss_clip": 0.01168771,
      "auxiliary_loss_mlp": 0.01022293,
      "balance_loss_clip": 1.04648829,
      "balance_loss_mlp": 1.01500893,
      "epoch": 0.9892382612878013,
      "flos": 26176580760960.0,
      "grad_norm": 2.2173356499880956,
      "language_loss": 0.75117272,
      "learning_rate": 1.2014890999973992e-09,
      "loss": 0.77308333,
      "num_input_tokens_seen": 177530705,
      "step": 8227,
      "time_per_iteration": 4.458129644393921
    },
    {
      "auxiliary_loss_clip": 0.01132896,
      "auxiliary_loss_mlp": 0.01023666,
      "balance_loss_clip": 1.04008722,
      "balance_loss_mlp": 1.01586962,
      "epoch": 0.9893585041784404,
      "flos": 25449605400960.0,
      "grad_norm": 1.7735982906165992,
      "language_loss": 0.78531426,
      "learning_rate": 1.1746436573472073e-09,
      "loss": 0.80687988,
      "num_input_tokens_seen": 177552440,
      "step": 8228,
      "time_per_iteration": 2.718994140625
    },
    {
      "auxiliary_loss_clip": 0.01122756,
      "auxiliary_loss_mlp": 0.01027684,
      "balance_loss_clip": 1.04147673,
      "balance_loss_mlp": 1.01970947,
      "epoch": 0.9894787470690796,
      "flos": 20189302352640.0,
      "grad_norm": 2.176846250108278,
      "language_loss": 0.69074321,
      "learning_rate": 1.1481014349141726e-09,
      "loss": 0.71224761,
      "num_input_tokens_seen": 177569660,
      "step": 8229,
      "time_per_iteration": 2.7645773887634277
    },
    {
      "auxiliary_loss_clip": 0.01123714,
      "auxiliary_loss_mlp": 0.01033187,
      "balance_loss_clip": 1.04330063,
      "balance_loss_mlp": 1.02490163,
      "epoch": 0.9895989899597186,
      "flos": 24644308435200.0,
      "grad_norm": 1.9163544665929764,
      "language_loss": 0.84612584,
      "learning_rate": 1.121862436724852e-09,
      "loss": 0.86769485,
      "num_input_tokens_seen": 177588500,
      "step": 8230,
      "time_per_iteration": 2.6972134113311768
    },
    {
      "auxiliary_loss_clip": 0.01114119,
      "auxiliary_loss_mlp": 0.0102491,
      "balance_loss_clip": 1.04432511,
      "balance_loss_mlp": 1.01776898,
      "epoch": 0.9897192328503577,
      "flos": 21799034357760.0,
      "grad_norm": 4.73187890566837,
      "language_loss": 0.70287812,
      "learning_rate": 1.0959266667598388e-09,
      "loss": 0.72426844,
      "num_input_tokens_seen": 177607315,
      "step": 8231,
      "time_per_iteration": 2.742626190185547
    },
    {
      "auxiliary_loss_clip": 0.01140233,
      "auxiliary_loss_mlp": 0.01025612,
      "balance_loss_clip": 1.04483378,
      "balance_loss_mlp": 1.01706421,
      "epoch": 0.9898394757409968,
      "flos": 21325229032320.0,
      "grad_norm": 2.031904804064391,
      "language_loss": 0.75111932,
      "learning_rate": 1.0702941289533196e-09,
      "loss": 0.7727778,
      "num_input_tokens_seen": 177625990,
      "step": 8232,
      "time_per_iteration": 2.6417534351348877
    },
    {
      "auxiliary_loss_clip": 0.01152003,
      "auxiliary_loss_mlp": 0.01024392,
      "balance_loss_clip": 1.044209,
      "balance_loss_mlp": 1.01732314,
      "epoch": 0.9899597186316359,
      "flos": 18545024442240.0,
      "grad_norm": 2.2937067179852657,
      "language_loss": 0.88825822,
      "learning_rate": 1.0449648271939615e-09,
      "loss": 0.91002214,
      "num_input_tokens_seen": 177642335,
      "step": 8233,
      "time_per_iteration": 2.6182680130004883
    },
    {
      "auxiliary_loss_clip": 0.01154419,
      "auxiliary_loss_mlp": 0.00762128,
      "balance_loss_clip": 1.04533315,
      "balance_loss_mlp": 1.00043344,
      "epoch": 0.990079961522275,
      "flos": 23766723348480.0,
      "grad_norm": 2.2397784413516835,
      "language_loss": 0.73004448,
      "learning_rate": 1.0199387653240243e-09,
      "loss": 0.74920994,
      "num_input_tokens_seen": 177662025,
      "step": 8234,
      "time_per_iteration": 3.8363518714904785
    },
    {
      "auxiliary_loss_clip": 0.01130035,
      "auxiliary_loss_mlp": 0.01024063,
      "balance_loss_clip": 1.04385304,
      "balance_loss_mlp": 1.01624858,
      "epoch": 0.9902002044129141,
      "flos": 16399182971520.0,
      "grad_norm": 3.645521802805884,
      "language_loss": 0.71176934,
      "learning_rate": 9.952159471400267e-10,
      "loss": 0.73331034,
      "num_input_tokens_seen": 177679065,
      "step": 8235,
      "time_per_iteration": 2.671959161758423
    },
    {
      "auxiliary_loss_clip": 0.01134897,
      "auxiliary_loss_mlp": 0.01024356,
      "balance_loss_clip": 1.04369664,
      "balance_loss_mlp": 1.0167892,
      "epoch": 0.9903204473035532,
      "flos": 22559657783040.0,
      "grad_norm": 2.528899769327356,
      "language_loss": 0.84318453,
      "learning_rate": 9.707963763923022e-10,
      "loss": 0.86477709,
      "num_input_tokens_seen": 177698115,
      "step": 8236,
      "time_per_iteration": 2.7922558784484863
    },
    {
      "auxiliary_loss_clip": 0.01114027,
      "auxiliary_loss_mlp": 0.01030711,
      "balance_loss_clip": 1.04262447,
      "balance_loss_mlp": 1.02235985,
      "epoch": 0.9904406901941922,
      "flos": 16144001775360.0,
      "grad_norm": 1.7853718469970097,
      "language_loss": 0.79123664,
      "learning_rate": 9.466800567854427e-10,
      "loss": 0.81268406,
      "num_input_tokens_seen": 177716715,
      "step": 8237,
      "time_per_iteration": 2.7068543434143066
    },
    {
      "auxiliary_loss_clip": 0.01133615,
      "auxiliary_loss_mlp": 0.01030791,
      "balance_loss_clip": 1.04415238,
      "balance_loss_mlp": 1.02225602,
      "epoch": 0.9905609330848314,
      "flos": 26651499408000.0,
      "grad_norm": 1.7799803205040001,
      "language_loss": 0.68588054,
      "learning_rate": 9.228669919778553e-10,
      "loss": 0.7075246,
      "num_input_tokens_seen": 177735640,
      "step": 8238,
      "time_per_iteration": 2.6888959407806396
    },
    {
      "auxiliary_loss_clip": 0.01144736,
      "auxiliary_loss_mlp": 0.0102403,
      "balance_loss_clip": 1.04353523,
      "balance_loss_mlp": 1.01637709,
      "epoch": 0.9906811759754705,
      "flos": 23111820627840.0,
      "grad_norm": 2.1714737379927467,
      "language_loss": 0.79641354,
      "learning_rate": 8.993571855817617e-10,
      "loss": 0.81810129,
      "num_input_tokens_seen": 177754470,
      "step": 8239,
      "time_per_iteration": 2.6617624759674072
    },
    {
      "auxiliary_loss_clip": 0.01139816,
      "auxiliary_loss_mlp": 0.01027104,
      "balance_loss_clip": 1.04536462,
      "balance_loss_mlp": 1.02005911,
      "epoch": 0.9908014188661095,
      "flos": 22090593052800.0,
      "grad_norm": 1.860914263261462,
      "language_loss": 0.75121474,
      "learning_rate": 8.761506411638642e-10,
      "loss": 0.77288395,
      "num_input_tokens_seen": 177773935,
      "step": 8240,
      "time_per_iteration": 2.6574296951293945
    },
    {
      "auxiliary_loss_clip": 0.01156526,
      "auxiliary_loss_mlp": 0.01023557,
      "balance_loss_clip": 1.04481745,
      "balance_loss_mlp": 1.0163058,
      "epoch": 0.9909216617567487,
      "flos": 19242948677760.0,
      "grad_norm": 2.3418749272904416,
      "language_loss": 0.73906279,
      "learning_rate": 8.53247362244236e-10,
      "loss": 0.7608636,
      "num_input_tokens_seen": 177792745,
      "step": 8241,
      "time_per_iteration": 2.685185670852661
    },
    {
      "auxiliary_loss_clip": 0.01098309,
      "auxiliary_loss_mlp": 0.01026248,
      "balance_loss_clip": 1.03926754,
      "balance_loss_mlp": 1.01808858,
      "epoch": 0.9910419046473877,
      "flos": 23621213352960.0,
      "grad_norm": 1.7156701721197254,
      "language_loss": 0.68462741,
      "learning_rate": 8.306473522976532e-10,
      "loss": 0.70587295,
      "num_input_tokens_seen": 177812150,
      "step": 8242,
      "time_per_iteration": 2.8360631465911865
    },
    {
      "auxiliary_loss_clip": 0.01136267,
      "auxiliary_loss_mlp": 0.01027287,
      "balance_loss_clip": 1.04130793,
      "balance_loss_mlp": 1.01933599,
      "epoch": 0.9911621475380268,
      "flos": 22711380831360.0,
      "grad_norm": 2.834007817969741,
      "language_loss": 0.7198295,
      "learning_rate": 8.083506147522623e-10,
      "loss": 0.74146497,
      "num_input_tokens_seen": 177831545,
      "step": 8243,
      "time_per_iteration": 2.7742769718170166
    },
    {
      "auxiliary_loss_clip": 0.01082094,
      "auxiliary_loss_mlp": 0.00762391,
      "balance_loss_clip": 1.03541124,
      "balance_loss_mlp": 1.00044763,
      "epoch": 0.991282390428666,
      "flos": 13516956777600.0,
      "grad_norm": 2.0447462385187873,
      "language_loss": 0.8522296,
      "learning_rate": 7.863571529906909e-10,
      "loss": 0.87067437,
      "num_input_tokens_seen": 177847130,
      "step": 8244,
      "time_per_iteration": 2.7867212295532227
    },
    {
      "auxiliary_loss_clip": 0.01037097,
      "auxiliary_loss_mlp": 0.01001103,
      "balance_loss_clip": 1.00928617,
      "balance_loss_mlp": 1.00014293,
      "epoch": 0.991402633319305,
      "flos": 61830492071040.0,
      "grad_norm": 0.7719139336340772,
      "language_loss": 0.59628445,
      "learning_rate": 7.646669703489372e-10,
      "loss": 0.61666644,
      "num_input_tokens_seen": 177911440,
      "step": 8245,
      "time_per_iteration": 3.4157962799072266
    },
    {
      "auxiliary_loss_clip": 0.0116783,
      "auxiliary_loss_mlp": 0.01019843,
      "balance_loss_clip": 1.04796779,
      "balance_loss_mlp": 1.01243424,
      "epoch": 0.9915228762099441,
      "flos": 18770148933120.0,
      "grad_norm": 2.1139620803160972,
      "language_loss": 0.57450277,
      "learning_rate": 7.432800701177023e-10,
      "loss": 0.59637952,
      "num_input_tokens_seen": 177929440,
      "step": 8246,
      "time_per_iteration": 2.5463216304779053
    },
    {
      "auxiliary_loss_clip": 0.01045408,
      "auxiliary_loss_mlp": 0.01001396,
      "balance_loss_clip": 1.01010633,
      "balance_loss_mlp": 1.00042427,
      "epoch": 0.9916431191005832,
      "flos": 65936660244480.0,
      "grad_norm": 0.7958542670080103,
      "language_loss": 0.57793403,
      "learning_rate": 7.221964555415017e-10,
      "loss": 0.59840202,
      "num_input_tokens_seen": 177989100,
      "step": 8247,
      "time_per_iteration": 3.1683924198150635
    },
    {
      "auxiliary_loss_clip": 0.0111667,
      "auxiliary_loss_mlp": 0.0102022,
      "balance_loss_clip": 1.04015827,
      "balance_loss_mlp": 1.01311159,
      "epoch": 0.9917633619912223,
      "flos": 16581573256320.0,
      "grad_norm": 1.9184659872875236,
      "language_loss": 0.75068533,
      "learning_rate": 7.01416129818222e-10,
      "loss": 0.77205431,
      "num_input_tokens_seen": 178006720,
      "step": 8248,
      "time_per_iteration": 2.703279733657837
    },
    {
      "auxiliary_loss_clip": 0.01125513,
      "auxiliary_loss_mlp": 0.01022853,
      "balance_loss_clip": 1.04451609,
      "balance_loss_mlp": 1.01527691,
      "epoch": 0.9918836048818613,
      "flos": 25411108999680.0,
      "grad_norm": 1.9854555672030338,
      "language_loss": 0.58976912,
      "learning_rate": 6.809390961006745e-10,
      "loss": 0.61125278,
      "num_input_tokens_seen": 178026850,
      "step": 8249,
      "time_per_iteration": 2.7551655769348145
    },
    {
      "auxiliary_loss_clip": 0.01096726,
      "auxiliary_loss_mlp": 0.01024187,
      "balance_loss_clip": 1.04041219,
      "balance_loss_mlp": 1.01702237,
      "epoch": 0.9920038477725005,
      "flos": 25046867134080.0,
      "grad_norm": 1.8199420939611128,
      "language_loss": 0.68557441,
      "learning_rate": 6.607653574948191e-10,
      "loss": 0.70678353,
      "num_input_tokens_seen": 178047630,
      "step": 8250,
      "time_per_iteration": 2.8633742332458496
    },
    {
      "auxiliary_loss_clip": 0.01138488,
      "auxiliary_loss_mlp": 0.01030558,
      "balance_loss_clip": 1.04120111,
      "balance_loss_mlp": 1.02339971,
      "epoch": 0.9921240906631396,
      "flos": 21829773421440.0,
      "grad_norm": 3.3902839667406504,
      "language_loss": 0.81945014,
      "learning_rate": 6.408949170613187e-10,
      "loss": 0.84114051,
      "num_input_tokens_seen": 178066895,
      "step": 8251,
      "time_per_iteration": 3.5733065605163574
    },
    {
      "auxiliary_loss_clip": 0.01121704,
      "auxiliary_loss_mlp": 0.01025582,
      "balance_loss_clip": 1.04401445,
      "balance_loss_mlp": 1.01775575,
      "epoch": 0.9922443335537786,
      "flos": 24864225454080.0,
      "grad_norm": 1.988676651544001,
      "language_loss": 0.81870854,
      "learning_rate": 6.213277778144288e-10,
      "loss": 0.84018135,
      "num_input_tokens_seen": 178088540,
      "step": 8252,
      "time_per_iteration": 4.609842538833618
    },
    {
      "auxiliary_loss_clip": 0.01074479,
      "auxiliary_loss_mlp": 0.01021441,
      "balance_loss_clip": 1.03422296,
      "balance_loss_mlp": 1.01394224,
      "epoch": 0.9923645764444178,
      "flos": 21613088626560.0,
      "grad_norm": 3.0052032442588046,
      "language_loss": 0.67057413,
      "learning_rate": 6.020639427224416e-10,
      "loss": 0.69153333,
      "num_input_tokens_seen": 178106185,
      "step": 8253,
      "time_per_iteration": 2.7910356521606445
    },
    {
      "auxiliary_loss_clip": 0.01134663,
      "auxiliary_loss_mlp": 0.01023491,
      "balance_loss_clip": 1.04499125,
      "balance_loss_mlp": 1.01605844,
      "epoch": 0.9924848193350568,
      "flos": 25001798544000.0,
      "grad_norm": 2.5611865200489747,
      "language_loss": 0.7285111,
      "learning_rate": 5.831034147076864e-10,
      "loss": 0.75009257,
      "num_input_tokens_seen": 178123435,
      "step": 8254,
      "time_per_iteration": 2.709933042526245
    },
    {
      "auxiliary_loss_clip": 0.01065605,
      "auxiliary_loss_mlp": 0.01001703,
      "balance_loss_clip": 1.00953317,
      "balance_loss_mlp": 1.00070119,
      "epoch": 0.9926050622256959,
      "flos": 68912543151360.0,
      "grad_norm": 0.6855382613980922,
      "language_loss": 0.5571574,
      "learning_rate": 5.644461966463065e-10,
      "loss": 0.57783043,
      "num_input_tokens_seen": 178191045,
      "step": 8255,
      "time_per_iteration": 3.304326295852661
    },
    {
      "auxiliary_loss_clip": 0.0110499,
      "auxiliary_loss_mlp": 0.01029588,
      "balance_loss_clip": 1.04467463,
      "balance_loss_mlp": 1.02209568,
      "epoch": 0.9927253051163349,
      "flos": 20923675914240.0,
      "grad_norm": 1.8152611814504396,
      "language_loss": 0.75976288,
      "learning_rate": 5.460922913687049e-10,
      "loss": 0.78110868,
      "num_input_tokens_seen": 178210135,
      "step": 8256,
      "time_per_iteration": 2.820828437805176
    },
    {
      "auxiliary_loss_clip": 0.01127139,
      "auxiliary_loss_mlp": 0.01023766,
      "balance_loss_clip": 1.04379106,
      "balance_loss_mlp": 1.01584435,
      "epoch": 0.9928455480069741,
      "flos": 22308211601280.0,
      "grad_norm": 2.0912321973981967,
      "language_loss": 0.75197369,
      "learning_rate": 5.280417016593208e-10,
      "loss": 0.7734828,
      "num_input_tokens_seen": 178229925,
      "step": 8257,
      "time_per_iteration": 2.7280185222625732
    },
    {
      "auxiliary_loss_clip": 0.01156218,
      "auxiliary_loss_mlp": 0.01025226,
      "balance_loss_clip": 1.04667568,
      "balance_loss_mlp": 1.01703644,
      "epoch": 0.9929657908976132,
      "flos": 17383889393280.0,
      "grad_norm": 2.1470301560206435,
      "language_loss": 0.75040334,
      "learning_rate": 5.102944302559642e-10,
      "loss": 0.77221775,
      "num_input_tokens_seen": 178247420,
      "step": 8258,
      "time_per_iteration": 2.666407585144043
    },
    {
      "auxiliary_loss_clip": 0.01159135,
      "auxiliary_loss_mlp": 0.01027042,
      "balance_loss_clip": 1.04970145,
      "balance_loss_mlp": 1.01881647,
      "epoch": 0.9930860337882522,
      "flos": 22674680110080.0,
      "grad_norm": 2.4621143559115874,
      "language_loss": 0.80053067,
      "learning_rate": 4.9285047985137e-10,
      "loss": 0.82239246,
      "num_input_tokens_seen": 178266840,
      "step": 8259,
      "time_per_iteration": 2.672119140625
    },
    {
      "auxiliary_loss_clip": 0.01069241,
      "auxiliary_loss_mlp": 0.0102622,
      "balance_loss_clip": 1.0415318,
      "balance_loss_mlp": 1.01810801,
      "epoch": 0.9932062766788914,
      "flos": 28147789284480.0,
      "grad_norm": 1.6667550100368935,
      "language_loss": 0.74375921,
      "learning_rate": 4.757098530916436e-10,
      "loss": 0.76471388,
      "num_input_tokens_seen": 178287285,
      "step": 8260,
      "time_per_iteration": 4.337630987167358
    },
    {
      "auxiliary_loss_clip": 0.01172883,
      "auxiliary_loss_mlp": 0.0076223,
      "balance_loss_clip": 1.04928398,
      "balance_loss_mlp": 1.00045013,
      "epoch": 0.9933265195695304,
      "flos": 20156659868160.0,
      "grad_norm": 3.038619982463179,
      "language_loss": 0.77328372,
      "learning_rate": 4.5887255257670563e-10,
      "loss": 0.79263484,
      "num_input_tokens_seen": 178304325,
      "step": 8261,
      "time_per_iteration": 2.9803857803344727
    },
    {
      "auxiliary_loss_clip": 0.0112348,
      "auxiliary_loss_mlp": 0.01024295,
      "balance_loss_clip": 1.04046071,
      "balance_loss_mlp": 1.01681507,
      "epoch": 0.9934467624601695,
      "flos": 21362037494400.0,
      "grad_norm": 2.1960703101750245,
      "language_loss": 0.76632386,
      "learning_rate": 4.4233858086117906e-10,
      "loss": 0.78780162,
      "num_input_tokens_seen": 178322850,
      "step": 8262,
      "time_per_iteration": 2.694382429122925
    },
    {
      "auxiliary_loss_clip": 0.01153683,
      "auxiliary_loss_mlp": 0.01024252,
      "balance_loss_clip": 1.04433703,
      "balance_loss_mlp": 1.01667905,
      "epoch": 0.9935670053508087,
      "flos": 19756040503680.0,
      "grad_norm": 3.616107495241777,
      "language_loss": 0.6805619,
      "learning_rate": 4.261079404528356e-10,
      "loss": 0.70234126,
      "num_input_tokens_seen": 178342330,
      "step": 8263,
      "time_per_iteration": 2.66475510597229
    },
    {
      "auxiliary_loss_clip": 0.0114,
      "auxiliary_loss_mlp": 0.0102918,
      "balance_loss_clip": 1.04478717,
      "balance_loss_mlp": 1.02136016,
      "epoch": 0.9936872482414477,
      "flos": 21978838863360.0,
      "grad_norm": 3.776441218862851,
      "language_loss": 0.69540846,
      "learning_rate": 4.1018063381437205e-10,
      "loss": 0.71710026,
      "num_input_tokens_seen": 178362715,
      "step": 8264,
      "time_per_iteration": 2.6945483684539795
    },
    {
      "auxiliary_loss_clip": 0.01057024,
      "auxiliary_loss_mlp": 0.01000797,
      "balance_loss_clip": 1.00978565,
      "balance_loss_mlp": 0.99980736,
      "epoch": 0.9938074911320868,
      "flos": 69810667839360.0,
      "grad_norm": 0.862223822467689,
      "language_loss": 0.61114293,
      "learning_rate": 3.9455666336141167e-10,
      "loss": 0.63172114,
      "num_input_tokens_seen": 178426495,
      "step": 8265,
      "time_per_iteration": 3.275158643722534
    },
    {
      "auxiliary_loss_clip": 0.01082646,
      "auxiliary_loss_mlp": 0.00761995,
      "balance_loss_clip": 1.03533471,
      "balance_loss_mlp": 1.00041294,
      "epoch": 0.9939277340227259,
      "flos": 15084170058240.0,
      "grad_norm": 3.6556956360323096,
      "language_loss": 0.83917677,
      "learning_rate": 3.7923603146450267e-10,
      "loss": 0.85762322,
      "num_input_tokens_seen": 178442555,
      "step": 8266,
      "time_per_iteration": 2.7646989822387695
    },
    {
      "auxiliary_loss_clip": 0.01119544,
      "auxiliary_loss_mlp": 0.01024519,
      "balance_loss_clip": 1.03877151,
      "balance_loss_mlp": 1.01687145,
      "epoch": 0.994047976913365,
      "flos": 17712364291200.0,
      "grad_norm": 14.48943991459738,
      "language_loss": 0.80599546,
      "learning_rate": 3.642187404473418e-10,
      "loss": 0.82743609,
      "num_input_tokens_seen": 178460715,
      "step": 8267,
      "time_per_iteration": 2.6884045600891113
    },
    {
      "auxiliary_loss_clip": 0.01137864,
      "auxiliary_loss_mlp": 0.01024339,
      "balance_loss_clip": 1.04732633,
      "balance_loss_mlp": 1.01683486,
      "epoch": 0.994168219804004,
      "flos": 19171558396800.0,
      "grad_norm": 2.4584365713376632,
      "language_loss": 0.86163092,
      "learning_rate": 3.495047925885508e-10,
      "loss": 0.88325298,
      "num_input_tokens_seen": 178479050,
      "step": 8268,
      "time_per_iteration": 2.618936538696289
    },
    {
      "auxiliary_loss_clip": 0.01118407,
      "auxiliary_loss_mlp": 0.01023138,
      "balance_loss_clip": 1.04316545,
      "balance_loss_mlp": 1.01572609,
      "epoch": 0.9942884626946432,
      "flos": 17851589406720.0,
      "grad_norm": 3.24002436925833,
      "language_loss": 0.83235586,
      "learning_rate": 3.350941901199e-10,
      "loss": 0.85377133,
      "num_input_tokens_seen": 178495970,
      "step": 8269,
      "time_per_iteration": 2.7510976791381836
    },
    {
      "auxiliary_loss_clip": 0.01087546,
      "auxiliary_loss_mlp": 0.01023455,
      "balance_loss_clip": 1.0389452,
      "balance_loss_mlp": 1.01574802,
      "epoch": 0.9944087055852823,
      "flos": 18796578364800.0,
      "grad_norm": 3.0038903401642463,
      "language_loss": 0.83790946,
      "learning_rate": 3.2098693522764066e-10,
      "loss": 0.85901946,
      "num_input_tokens_seen": 178509170,
      "step": 8270,
      "time_per_iteration": 2.765131950378418
    },
    {
      "auxiliary_loss_clip": 0.01119574,
      "auxiliary_loss_mlp": 0.01025599,
      "balance_loss_clip": 1.0419302,
      "balance_loss_mlp": 1.01746893,
      "epoch": 0.9945289484759213,
      "flos": 20996969616000.0,
      "grad_norm": 2.171332357404602,
      "language_loss": 0.80968177,
      "learning_rate": 3.071830300516165e-10,
      "loss": 0.83113348,
      "num_input_tokens_seen": 178527000,
      "step": 8271,
      "time_per_iteration": 2.6918208599090576
    },
    {
      "auxiliary_loss_clip": 0.01096132,
      "auxiliary_loss_mlp": 0.01024553,
      "balance_loss_clip": 1.03989613,
      "balance_loss_mlp": 1.01644683,
      "epoch": 0.9946491913665605,
      "flos": 14756952136320.0,
      "grad_norm": 1.9583224359394047,
      "language_loss": 0.71169758,
      "learning_rate": 2.9368247668615234e-10,
      "loss": 0.73290449,
      "num_input_tokens_seen": 178545590,
      "step": 8272,
      "time_per_iteration": 2.7911014556884766
    },
    {
      "auxiliary_loss_clip": 0.01095413,
      "auxiliary_loss_mlp": 0.01024982,
      "balance_loss_clip": 1.04204249,
      "balance_loss_mlp": 1.01696563,
      "epoch": 0.9947694342571995,
      "flos": 12669931186560.0,
      "grad_norm": 4.824513737147081,
      "language_loss": 0.61342865,
      "learning_rate": 2.804852771789434e-10,
      "loss": 0.63463259,
      "num_input_tokens_seen": 178558890,
      "step": 8273,
      "time_per_iteration": 2.7969143390655518
    },
    {
      "auxiliary_loss_clip": 0.0109697,
      "auxiliary_loss_mlp": 0.01026846,
      "balance_loss_clip": 1.03833437,
      "balance_loss_mlp": 1.01933539,
      "epoch": 0.9948896771478386,
      "flos": 18843442634880.0,
      "grad_norm": 1.7437595770695626,
      "language_loss": 0.56156081,
      "learning_rate": 2.675914335321661e-10,
      "loss": 0.58279902,
      "num_input_tokens_seen": 178577645,
      "step": 8274,
      "time_per_iteration": 2.735664129257202
    },
    {
      "auxiliary_loss_clip": 0.01114895,
      "auxiliary_loss_mlp": 0.01031569,
      "balance_loss_clip": 1.0419836,
      "balance_loss_mlp": 1.02245593,
      "epoch": 0.9950099200384778,
      "flos": 24900207903360.0,
      "grad_norm": 2.7424598613011666,
      "language_loss": 0.79962957,
      "learning_rate": 2.550009477018111e-10,
      "loss": 0.82109416,
      "num_input_tokens_seen": 178596415,
      "step": 8275,
      "time_per_iteration": 2.7693042755126953
    },
    {
      "auxiliary_loss_clip": 0.01144131,
      "auxiliary_loss_mlp": 0.00762368,
      "balance_loss_clip": 1.04304194,
      "balance_loss_mlp": 1.0004636,
      "epoch": 0.9951301629291168,
      "flos": 23733613987200.0,
      "grad_norm": 2.51386918181101,
      "language_loss": 0.64018041,
      "learning_rate": 2.4271382159790634e-10,
      "loss": 0.65924543,
      "num_input_tokens_seen": 178613845,
      "step": 8276,
      "time_per_iteration": 2.714388608932495
    },
    {
      "auxiliary_loss_clip": 0.01137133,
      "auxiliary_loss_mlp": 0.0102739,
      "balance_loss_clip": 1.04479134,
      "balance_loss_mlp": 1.01963508,
      "epoch": 0.9952504058197559,
      "flos": 22236893147520.0,
      "grad_norm": 2.436857013828781,
      "language_loss": 0.86024272,
      "learning_rate": 2.3073005708429406e-10,
      "loss": 0.88188791,
      "num_input_tokens_seen": 178633490,
      "step": 8277,
      "time_per_iteration": 4.162052154541016
    },
    {
      "auxiliary_loss_clip": 0.01111227,
      "auxiliary_loss_mlp": 0.01023432,
      "balance_loss_clip": 1.03896821,
      "balance_loss_mlp": 1.01639271,
      "epoch": 0.995370648710395,
      "flos": 21211032718080.0,
      "grad_norm": 1.9055983673431087,
      "language_loss": 0.72141993,
      "learning_rate": 2.190496559788535e-10,
      "loss": 0.74276656,
      "num_input_tokens_seen": 178651775,
      "step": 8278,
      "time_per_iteration": 3.6308350563049316
    },
    {
      "auxiliary_loss_clip": 0.01138878,
      "auxiliary_loss_mlp": 0.01026435,
      "balance_loss_clip": 1.04287803,
      "balance_loss_mlp": 1.01824498,
      "epoch": 0.9954908916010341,
      "flos": 14866731077760.0,
      "grad_norm": 2.719105986476604,
      "language_loss": 0.77124232,
      "learning_rate": 2.0767262005372265e-10,
      "loss": 0.79289544,
      "num_input_tokens_seen": 178669290,
      "step": 8279,
      "time_per_iteration": 3.5487570762634277
    },
    {
      "auxiliary_loss_clip": 0.0111016,
      "auxiliary_loss_mlp": 0.01027421,
      "balance_loss_clip": 1.03930235,
      "balance_loss_mlp": 1.01995826,
      "epoch": 0.9956111344916732,
      "flos": 19208259118080.0,
      "grad_norm": 2.5858717784310485,
      "language_loss": 0.75521421,
      "learning_rate": 1.965989510346322e-10,
      "loss": 0.77659005,
      "num_input_tokens_seen": 178688410,
      "step": 8280,
      "time_per_iteration": 2.7091596126556396
    },
    {
      "auxiliary_loss_clip": 0.01119977,
      "auxiliary_loss_mlp": 0.010246,
      "balance_loss_clip": 1.03973317,
      "balance_loss_mlp": 1.01691115,
      "epoch": 0.9957313773823123,
      "flos": 20047060494720.0,
      "grad_norm": 2.549549815953917,
      "language_loss": 0.70999599,
      "learning_rate": 1.8582865060134955e-10,
      "loss": 0.73144174,
      "num_input_tokens_seen": 178706600,
      "step": 8281,
      "time_per_iteration": 2.7905726432800293
    },
    {
      "auxiliary_loss_clip": 0.01038698,
      "auxiliary_loss_mlp": 0.01000405,
      "balance_loss_clip": 1.01033044,
      "balance_loss_mlp": 0.99938601,
      "epoch": 0.9958516202729514,
      "flos": 57483253768320.0,
      "grad_norm": 0.7690282112073168,
      "language_loss": 0.55723619,
      "learning_rate": 1.7536172038790098e-10,
      "loss": 0.57762724,
      "num_input_tokens_seen": 178766910,
      "step": 8282,
      "time_per_iteration": 3.341707468032837
    },
    {
      "auxiliary_loss_clip": 0.01086252,
      "auxiliary_loss_mlp": 0.01021416,
      "balance_loss_clip": 1.0371635,
      "balance_loss_mlp": 1.01384068,
      "epoch": 0.9959718631635904,
      "flos": 27782900974080.0,
      "grad_norm": 4.1199434596563185,
      "language_loss": 0.69305158,
      "learning_rate": 1.651981619819054e-10,
      "loss": 0.71412826,
      "num_input_tokens_seen": 178784060,
      "step": 8283,
      "time_per_iteration": 2.896991014480591
    },
    {
      "auxiliary_loss_clip": 0.01136469,
      "auxiliary_loss_mlp": 0.01026277,
      "balance_loss_clip": 1.04273796,
      "balance_loss_mlp": 1.01878738,
      "epoch": 0.9960921060542296,
      "flos": 24024095274240.0,
      "grad_norm": 3.595495064604763,
      "language_loss": 0.7103976,
      "learning_rate": 1.5533797692546257e-10,
      "loss": 0.73202509,
      "num_input_tokens_seen": 178802795,
      "step": 8284,
      "time_per_iteration": 2.995058536529541
    },
    {
      "auxiliary_loss_clip": 0.01172087,
      "auxiliary_loss_mlp": 0.01025188,
      "balance_loss_clip": 1.04840958,
      "balance_loss_mlp": 1.01746643,
      "epoch": 0.9962123489448687,
      "flos": 18697393935360.0,
      "grad_norm": 2.3902264292631843,
      "language_loss": 0.84340465,
      "learning_rate": 1.4578116671404296e-10,
      "loss": 0.86537743,
      "num_input_tokens_seen": 178821075,
      "step": 8285,
      "time_per_iteration": 2.710378885269165
    },
    {
      "auxiliary_loss_clip": 0.0113454,
      "auxiliary_loss_mlp": 0.01023069,
      "balance_loss_clip": 1.0453794,
      "balance_loss_mlp": 1.01533842,
      "epoch": 0.9963325918355077,
      "flos": 20010754823040.0,
      "grad_norm": 3.0919443228302117,
      "language_loss": 0.71758687,
      "learning_rate": 1.3652773279759777e-10,
      "loss": 0.73916292,
      "num_input_tokens_seen": 178837725,
      "step": 8286,
      "time_per_iteration": 3.6147706508636475
    },
    {
      "auxiliary_loss_clip": 0.01105908,
      "auxiliary_loss_mlp": 0.00762369,
      "balance_loss_clip": 1.04053235,
      "balance_loss_mlp": 1.00050664,
      "epoch": 0.9964528347261468,
      "flos": 33108488991360.0,
      "grad_norm": 1.9903878058553177,
      "language_loss": 0.63074899,
      "learning_rate": 1.2757767657989305e-10,
      "loss": 0.64943177,
      "num_input_tokens_seen": 178861515,
      "step": 8287,
      "time_per_iteration": 2.8528597354888916
    },
    {
      "auxiliary_loss_clip": 0.01078508,
      "auxiliary_loss_mlp": 0.01028672,
      "balance_loss_clip": 1.0360496,
      "balance_loss_mlp": 1.0210247,
      "epoch": 0.9965730776167859,
      "flos": 23109342589440.0,
      "grad_norm": 1.9688047608002965,
      "language_loss": 0.87315726,
      "learning_rate": 1.1893099941850948e-10,
      "loss": 0.89422899,
      "num_input_tokens_seen": 178880410,
      "step": 8288,
      "time_per_iteration": 2.8060920238494873
    },
    {
      "auxiliary_loss_clip": 0.01122639,
      "auxiliary_loss_mlp": 0.01024427,
      "balance_loss_clip": 1.04000318,
      "balance_loss_mlp": 1.01716971,
      "epoch": 0.996693320507425,
      "flos": 22965843755520.0,
      "grad_norm": 7.268886074375672,
      "language_loss": 0.78056628,
      "learning_rate": 1.105877026252866e-10,
      "loss": 0.80203694,
      "num_input_tokens_seen": 178898740,
      "step": 8289,
      "time_per_iteration": 2.703273296356201
    },
    {
      "auxiliary_loss_clip": 0.01097094,
      "auxiliary_loss_mlp": 0.01021346,
      "balance_loss_clip": 1.04087484,
      "balance_loss_mlp": 1.01360309,
      "epoch": 0.996813563398064,
      "flos": 13222740476160.0,
      "grad_norm": 2.2646888097694524,
      "language_loss": 0.72347832,
      "learning_rate": 1.0254778746565663e-10,
      "loss": 0.74466276,
      "num_input_tokens_seen": 178914015,
      "step": 8290,
      "time_per_iteration": 2.7414000034332275
    },
    {
      "auxiliary_loss_clip": 0.01112385,
      "auxiliary_loss_mlp": 0.01027918,
      "balance_loss_clip": 1.04505479,
      "balance_loss_mlp": 1.01977038,
      "epoch": 0.9969338062887032,
      "flos": 14647855553280.0,
      "grad_norm": 2.3441890914613173,
      "language_loss": 0.73347348,
      "learning_rate": 9.481125515953259e-11,
      "loss": 0.75487649,
      "num_input_tokens_seen": 178932075,
      "step": 8291,
      "time_per_iteration": 2.687997579574585
    },
    {
      "auxiliary_loss_clip": 0.01136716,
      "auxiliary_loss_mlp": 0.01023228,
      "balance_loss_clip": 1.04453719,
      "balance_loss_mlp": 1.01505065,
      "epoch": 0.9970540491793423,
      "flos": 25735741142400.0,
      "grad_norm": 1.7533350001653951,
      "language_loss": 0.7998507,
      "learning_rate": 8.737810688064228e-11,
      "loss": 0.82145011,
      "num_input_tokens_seen": 178951910,
      "step": 8292,
      "time_per_iteration": 2.719245433807373
    },
    {
      "auxiliary_loss_clip": 0.01144267,
      "auxiliary_loss_mlp": 0.01021929,
      "balance_loss_clip": 1.04249787,
      "balance_loss_mlp": 1.01378143,
      "epoch": 0.9971742920699813,
      "flos": 21470236237440.0,
      "grad_norm": 2.0787527633074463,
      "language_loss": 0.78728598,
      "learning_rate": 8.024834375608414e-11,
      "loss": 0.80894804,
      "num_input_tokens_seen": 178970500,
      "step": 8293,
      "time_per_iteration": 2.673407554626465
    },
    {
      "auxiliary_loss_clip": 0.01036262,
      "auxiliary_loss_mlp": 0.01001049,
      "balance_loss_clip": 1.00954771,
      "balance_loss_mlp": 1.0,
      "epoch": 0.9972945349606205,
      "flos": 72211223629440.0,
      "grad_norm": 0.8201875017458184,
      "language_loss": 0.62785017,
      "learning_rate": 7.342196686788149e-11,
      "loss": 0.64822328,
      "num_input_tokens_seen": 179023665,
      "step": 8294,
      "time_per_iteration": 3.169585704803467
    },
    {
      "auxiliary_loss_clip": 0.01147494,
      "auxiliary_loss_mlp": 0.00762526,
      "balance_loss_clip": 1.04608881,
      "balance_loss_mlp": 1.00046372,
      "epoch": 0.9974147778512595,
      "flos": 19678293515520.0,
      "grad_norm": 2.129779923443368,
      "language_loss": 0.68726599,
      "learning_rate": 6.689897725142834e-11,
      "loss": 0.70636624,
      "num_input_tokens_seen": 179043140,
      "step": 8295,
      "time_per_iteration": 2.684890031814575
    },
    {
      "auxiliary_loss_clip": 0.01151609,
      "auxiliary_loss_mlp": 0.01025299,
      "balance_loss_clip": 1.04414368,
      "balance_loss_mlp": 1.01730061,
      "epoch": 0.9975350207418986,
      "flos": 15960821391360.0,
      "grad_norm": 2.0943934088351925,
      "language_loss": 0.88205588,
      "learning_rate": 6.067937589615545e-11,
      "loss": 0.90382493,
      "num_input_tokens_seen": 179061215,
      "step": 8296,
      "time_per_iteration": 2.6486868858337402
    },
    {
      "auxiliary_loss_clip": 0.01043536,
      "auxiliary_loss_mlp": 0.00999591,
      "balance_loss_clip": 1.01108432,
      "balance_loss_mlp": 0.99868554,
      "epoch": 0.9976552636325378,
      "flos": 59961879768960.0,
      "grad_norm": 0.7595178331297082,
      "language_loss": 0.57665938,
      "learning_rate": 5.476316374575241e-11,
      "loss": 0.59709066,
      "num_input_tokens_seen": 179124700,
      "step": 8297,
      "time_per_iteration": 3.2446346282958984
    },
    {
      "auxiliary_loss_clip": 0.01145733,
      "auxiliary_loss_mlp": 0.01026638,
      "balance_loss_clip": 1.04441333,
      "balance_loss_mlp": 1.01836157,
      "epoch": 0.9977755065231768,
      "flos": 22487872452480.0,
      "grad_norm": 1.9222428833530105,
      "language_loss": 0.72385782,
      "learning_rate": 4.9150341697723476e-11,
      "loss": 0.74558151,
      "num_input_tokens_seen": 179144590,
      "step": 8298,
      "time_per_iteration": 2.756594181060791
    },
    {
      "auxiliary_loss_clip": 0.01122442,
      "auxiliary_loss_mlp": 0.01025068,
      "balance_loss_clip": 1.04501104,
      "balance_loss_mlp": 1.01722133,
      "epoch": 0.9978957494138159,
      "flos": 26030280666240.0,
      "grad_norm": 2.5874982223470946,
      "language_loss": 0.66996002,
      "learning_rate": 4.384091060338768e-11,
      "loss": 0.69143522,
      "num_input_tokens_seen": 179165060,
      "step": 8299,
      "time_per_iteration": 2.786309242248535
    },
    {
      "auxiliary_loss_clip": 0.0115265,
      "auxiliary_loss_mlp": 0.00761656,
      "balance_loss_clip": 1.0469476,
      "balance_loss_mlp": 1.00042439,
      "epoch": 0.998015992304455,
      "flos": 22637835734400.0,
      "grad_norm": 2.0113361041387234,
      "language_loss": 0.73959243,
      "learning_rate": 3.883487126810081e-11,
      "loss": 0.75873554,
      "num_input_tokens_seen": 179184320,
      "step": 8300,
      "time_per_iteration": 2.682182550430298
    },
    {
      "auxiliary_loss_clip": 0.01110664,
      "auxiliary_loss_mlp": 0.01022738,
      "balance_loss_clip": 1.03935432,
      "balance_loss_mlp": 1.0148468,
      "epoch": 0.9981362351950941,
      "flos": 18223444955520.0,
      "grad_norm": 1.7816289386529385,
      "language_loss": 0.79274756,
      "learning_rate": 3.41322244516995e-11,
      "loss": 0.81408155,
      "num_input_tokens_seen": 179202265,
      "step": 8301,
      "time_per_iteration": 2.758431911468506
    },
    {
      "auxiliary_loss_clip": 0.01093492,
      "auxiliary_loss_mlp": 0.01024451,
      "balance_loss_clip": 1.03670478,
      "balance_loss_mlp": 1.01665771,
      "epoch": 0.9982564780857331,
      "flos": 33474095573760.0,
      "grad_norm": 2.9666231509430125,
      "language_loss": 0.6316061,
      "learning_rate": 2.9732970866946925e-11,
      "loss": 0.65278548,
      "num_input_tokens_seen": 179222145,
      "step": 8302,
      "time_per_iteration": 2.8489110469818115
    },
    {
      "auxiliary_loss_clip": 0.01112699,
      "auxiliary_loss_mlp": 0.01029704,
      "balance_loss_clip": 1.03905296,
      "balance_loss_mlp": 1.02100158,
      "epoch": 0.9983767209763723,
      "flos": 15523465392000.0,
      "grad_norm": 2.136253548889351,
      "language_loss": 0.7817499,
      "learning_rate": 2.563711118175327e-11,
      "loss": 0.80317396,
      "num_input_tokens_seen": 179239030,
      "step": 8303,
      "time_per_iteration": 3.76017165184021
    },
    {
      "auxiliary_loss_clip": 0.01117358,
      "auxiliary_loss_mlp": 0.01023776,
      "balance_loss_clip": 1.03945136,
      "balance_loss_mlp": 1.01682901,
      "epoch": 0.9984969638670114,
      "flos": 19974377324160.0,
      "grad_norm": 1.7262337920765514,
      "language_loss": 0.83546406,
      "learning_rate": 2.184464601717728e-11,
      "loss": 0.85687542,
      "num_input_tokens_seen": 179257345,
      "step": 8304,
      "time_per_iteration": 4.524463415145874
    },
    {
      "auxiliary_loss_clip": 0.01160901,
      "auxiliary_loss_mlp": 0.01031248,
      "balance_loss_clip": 1.0515089,
      "balance_loss_mlp": 1.02262282,
      "epoch": 0.9986172067576504,
      "flos": 20375750874240.0,
      "grad_norm": 2.3919048101626283,
      "language_loss": 0.78016293,
      "learning_rate": 1.8355575948758585e-11,
      "loss": 0.80208439,
      "num_input_tokens_seen": 179275330,
      "step": 8305,
      "time_per_iteration": 2.592745780944824
    },
    {
      "auxiliary_loss_clip": 0.01078071,
      "auxiliary_loss_mlp": 0.01025774,
      "balance_loss_clip": 1.03679788,
      "balance_loss_mlp": 1.01836491,
      "epoch": 0.9987374496482896,
      "flos": 23727903724800.0,
      "grad_norm": 2.79105274199412,
      "language_loss": 0.73836589,
      "learning_rate": 1.5169901505407424e-11,
      "loss": 0.75940442,
      "num_input_tokens_seen": 179292395,
      "step": 8306,
      "time_per_iteration": 2.696301221847534
    },
    {
      "auxiliary_loss_clip": 0.01098861,
      "auxiliary_loss_mlp": 0.01025005,
      "balance_loss_clip": 1.04296935,
      "balance_loss_mlp": 1.01785851,
      "epoch": 0.9988576925389286,
      "flos": 25044029959680.0,
      "grad_norm": 2.0114872784037705,
      "language_loss": 0.74188107,
      "learning_rate": 1.228762317073695e-11,
      "loss": 0.7631197,
      "num_input_tokens_seen": 179311225,
      "step": 8307,
      "time_per_iteration": 2.667921781539917
    },
    {
      "auxiliary_loss_clip": 0.01131804,
      "auxiliary_loss_mlp": 0.00761691,
      "balance_loss_clip": 1.04192996,
      "balance_loss_mlp": 1.00040853,
      "epoch": 0.9989779354295677,
      "flos": 31285627637760.0,
      "grad_norm": 1.9030944040651265,
      "language_loss": 0.79072094,
      "learning_rate": 9.70874138195299e-12,
      "loss": 0.8096559,
      "num_input_tokens_seen": 179333135,
      "step": 8308,
      "time_per_iteration": 2.626349687576294
    },
    {
      "auxiliary_loss_clip": 0.01088944,
      "auxiliary_loss_mlp": 0.01021752,
      "balance_loss_clip": 1.0375396,
      "balance_loss_mlp": 1.01371098,
      "epoch": 0.9990981783202069,
      "flos": 19573398823680.0,
      "grad_norm": 2.3394974206883523,
      "language_loss": 0.74308908,
      "learning_rate": 7.433256530076093e-12,
      "loss": 0.76419604,
      "num_input_tokens_seen": 179353090,
      "step": 8309,
      "time_per_iteration": 2.832685947418213
    },
    {
      "auxiliary_loss_clip": 0.0111455,
      "auxiliary_loss_mlp": 0.01024358,
      "balance_loss_clip": 1.03934813,
      "balance_loss_mlp": 1.01699078,
      "epoch": 0.9992184212108459,
      "flos": 17199667514880.0,
      "grad_norm": 2.5471959836347535,
      "language_loss": 0.75813538,
      "learning_rate": 5.46116896038562e-12,
      "loss": 0.77952445,
      "num_input_tokens_seen": 179367500,
      "step": 8310,
      "time_per_iteration": 2.592132806777954
    },
    {
      "auxiliary_loss_clip": 0.0115449,
      "auxiliary_loss_mlp": 0.00762372,
      "balance_loss_clip": 1.0437603,
      "balance_loss_mlp": 1.0004189,
      "epoch": 0.999338664101485,
      "flos": 46497853681920.0,
      "grad_norm": 4.379190995150306,
      "language_loss": 0.62004292,
      "learning_rate": 3.792478972197699e-12,
      "loss": 0.63921154,
      "num_input_tokens_seen": 179388085,
      "step": 8311,
      "time_per_iteration": 2.861922025680542
    },
    {
      "auxiliary_loss_clip": 0.01147503,
      "auxiliary_loss_mlp": 0.01020332,
      "balance_loss_clip": 1.04362011,
      "balance_loss_mlp": 1.01282763,
      "epoch": 0.9994589069921241,
      "flos": 15158253859200.0,
      "grad_norm": 73.47778718962074,
      "language_loss": 0.70428944,
      "learning_rate": 2.4271868181990895e-12,
      "loss": 0.72596776,
      "num_input_tokens_seen": 179405250,
      "step": 8312,
      "time_per_iteration": 3.6590211391448975
    },
    {
      "auxiliary_loss_clip": 0.01089192,
      "auxiliary_loss_mlp": 0.01022892,
      "balance_loss_clip": 1.03696775,
      "balance_loss_mlp": 1.0149467,
      "epoch": 0.9995791498827632,
      "flos": 12531460256640.0,
      "grad_norm": 2.4715848493604744,
      "language_loss": 0.81190753,
      "learning_rate": 1.3652927060014973e-12,
      "loss": 0.83302838,
      "num_input_tokens_seen": 179420845,
      "step": 8313,
      "time_per_iteration": 2.699702024459839
    },
    {
      "auxiliary_loss_clip": 0.01124618,
      "auxiliary_loss_mlp": 0.01025587,
      "balance_loss_clip": 1.0460068,
      "balance_loss_mlp": 1.01711142,
      "epoch": 0.9996993927734023,
      "flos": 19245175320960.0,
      "grad_norm": 2.787200370915733,
      "language_loss": 0.64020818,
      "learning_rate": 6.067967965872612e-13,
      "loss": 0.66171026,
      "num_input_tokens_seen": 179440455,
      "step": 8314,
      "time_per_iteration": 2.670785665512085
    },
    {
      "auxiliary_loss_clip": 0.01157361,
      "auxiliary_loss_mlp": 0.01027418,
      "balance_loss_clip": 1.04686129,
      "balance_loss_mlp": 1.02018774,
      "epoch": 0.9998196356640414,
      "flos": 62952804518400.0,
      "grad_norm": 1.5908295124522422,
      "language_loss": 0.77292866,
      "learning_rate": 1.5169920497548615e-13,
      "loss": 0.79477644,
      "num_input_tokens_seen": 179465075,
      "step": 8315,
      "time_per_iteration": 2.9979214668273926
    },
    {
      "auxiliary_loss_clip": 0.01069573,
      "auxiliary_loss_mlp": 0.01014865,
      "balance_loss_clip": 1.02453637,
      "balance_loss_mlp": 1.01029348,
      "epoch": 0.9999398785546805,
      "flos": 50922375073920.0,
      "grad_norm": 1.387161080083939,
      "language_loss": 0.54892969,
      "learning_rate": 0.0,
      "loss": 0.56977409,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "time_per_iteration": 3.286201238632202
    },
    {
      "epoch": 0.9999398785546805,
      "num_input_tokens_seen": 179513955,
      "step": 8316,
      "total_flos": 6.996749092776837e+17,
      "train_loss": 0.7897510985418716,
      "train_runtime": 25154.9547,
      "train_samples_per_second": 13.224,
      "train_steps_per_second": 0.331
    }
  ],
  "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
}